PROPERTY^ THE UNITED^TATES UNITED STATES REPORTS VOLUME 482 CASES ADJUDGED IN THE SUPREME COURT AT OCTOBER TERM, 1986 June 1 Through June 19, 1987 FRANK D. WAGNER REPORTER OF DECISIONS UNITED STATES GOVERNMENT PRINTING OFFICE WASHINGTON : 1990 Errata 460 U. S. 546, n. 11, line 3: “differrent” should be “different”. 481 U. S. 672, n. 3, line 2: “521” should be “510”. il JUSTICES OF THE SUPREME COURT DURING THE TIME OF THESE REPORTS WILLIAM H. REHNQUIST, Chief Justice. WILLIAM J. BRENNAN, Jr., Associate Justice. BYRON R. WHITE, Associate Justice. THURGOOD MARSHALL, Associate Justice. HARRY A. BLACKMUN, Associate Justice. LEWIS F. POWELL, Jr., Associate Justice. JOHN PAUL STEVENS, Associate Justice. SANDRA DAY O’CONNOR, Associate Justice. ANTONIN SCALIA, Associate Justice. retired WARREN E. BURGER, Chief Justice. OFFICERS OF THE COURT EDWIN MEESE III, Attorney General. CHARLES FRIED, Solicitor General. JOSEPH F. SPANIOL, Jr., Clerk. FRANK D. WAGNER, Reporter of Decisions. ALFRED WONG, Marshal. STEPHEN G. MARGETON, Librarian. hi SUPREME COURT OF THE UNITED STATES Allotment of Justices It is ordered that the following allotment be made of the Chief Justice and Associate Justices of this Court among the circuits, pursuant to Title 28, United States Code, Section 42, and that such allotment be entered of record, viz.: For the District of Columbia Circuit, William H. Rehnquist, Chief Justice. For the First Circuit, William J. Brennan, Jr., Associate Justice. For the Second Circuit, Thurgood Marshall, Associate Justice. For the Third Circuit, William J. Brennan, Jr., Associate Justice. For the Fourth Circuit, William H. Rehnquist, Chief Justice. For the Fifth Circuit, Byron R. White, Associate Justice. For the Sixth Circuit, Antonin Scalia, Associate Justice. For the Seventh Circuit, John Paul Stevens, Associate Justice. For the Eighth Circuit, Harry A. Blackmun, Associate Justice. For the Ninth Circuit, Sandra Day O’Connor, Associate Justice. For the Tenth Circuit, Byron R. White, Associate Justice. For the Eleventh Circuit, Lewis F. Powell, Jr., Associate Justice. For the Federal Circuit, William H. Rehnquist, Chief Justice. October 6, 1986. (For next previous allotment, and modifications, see 453 U. S., p. vi, 459 U. S., p. iv, and 478 U. S., p. v.) IV TABLE OF CASES REPORTED Note: All undesignated references herein to the United States Code are to the 1982 edition. Cases reported before page 901 are those decided with opinions of the Court or decisions per curiam. Cases reported on page 901 et seq. are those in which orders were entered. Page Abbott v. Florida.............................................. 924 Adams; Owens v................................................. 907 Adkins v. Tennessee........................................... 909 Advanced Lightweight Concrete; Laborers Health & Welf. Tr. v. . 904 Agan v. Florida................................................ 932 A. G. Becker Inc. v. First Biscayne Corp....................... 924 Aguillard; Edwards v........................................... 578 Ahlberg v. Department of Health and Human Services............. 913 Aiken; Lucas v............................................... 931 Akzo N. V. v. United States International Trade Comm’n....... 909 Alabama; Hammond v............................................. 917 Alabama; Tarver v.............................................. 920 Allain; Burrell v. ............................................ 910 Allen; Board of Pardons v...................................... 369 Allendale Mut. Ins. Co.; AZL Engineering, Inc. v............... 914 Allendale Mut. Ins. Co.; Kaiser Engineers v.................... 914 Alton Packaging Corp.; Cahoon v................................ 922 American Savings & Loan Assn. v. Morgan Guaranty Trust Co.... 929 Annonson v. Grover............................................. 907 Appleby v. Maryland............................................ 930 Aragon v. California........................................... 913 Arizona; Tison v............................................... 921 Arizona State Real Estate Dept.; Wetzel v. ..................... 914 Arkansas v. Foster............................................. 929 Arlington v. Byrd.............................................. 929 Armstrong v. Kansas ........................................... 929 Asphalt Products Co. v. Commissioner........................... 117 Asphalt Products Co.; Commissioner v........................... 117 Astrike; Perry v............................................... 921 Attorney General of Mich.; 511 Detroit Street, Inc. v.......... 928 VI TABLE OF CASES REPORTED Page Attwell v. Metropolitan Atlanta Rapid Transit Authority...... 908 Averbach v. Rival Mfg. Co..................................... 915 Ayodeji v. United States...................................... 916 AZL Engineering, Inc. v. Allendale Mut. Ins. Co............... 914 Badart; Merrill Lynch, Pierce, Fenner & Smith, Inc. v. ....... 923 Bailey v. United States....................................... 908 Baker v. United States........................................ 921 Bankers Life & Casualty Co. v. Crenshaw....................... 925 Bascos v. United States....................................... 908 Basic Inc. v. Levinson........................................ 904 Batayola-v. Metropolitan Seattle............................. 914 Bates v. Blackbum ............................................ 916 Battles v. Oklahoma .......................................... 917 Bauer; Marshall v............................................. 910 Beasley v. United States...................................... 907 Beck; Communications Workers v................................ 904 Becker Inc. v. First Biscayne Corp............................ 924 Belk v. Chrans................................................ 931 Bell; Burlington Northern R. Co. v. .......................... 919 Berry v. Phelps .............................................. 910 Billings; Skurdal v. ......................................... 910 Bio-Rad Laboratories, Inc.; Nicolet Instrument Corp. v. ...... 915 Blackbum; Bates v............................................. 916 Blackburn; Williams v......................................... 917 Blinn v. Massachusetts........................................ 921 Blumenthal; Dean Witter Reynolds Inc. v. ..................... 923 Boag v. Bramlett.............................................. 930 Board of Airport Comm’rs of Los Angeles v. Jews for Jesus, Inc. . 569 Board of Pardons v. Allen..................................... 369 Bollinger; Commissioner v. ................................... 913 Booth v. Maryland............................................. 496 Bourke v. Shuman.............................................. 906 Bowen v. Dixon................................................ 922 Bowen v. Johnson ............................................. 922 Bowen; Monsour Medical Center v............................... 905 Bowen v. Polaski.............................................. 927 Bowen; Taylor v............................................... 913 Bowen v. Wilson............................................... 923 Bowen v. Yuckert.............................................. 137 Boyle v. United Technologies Corp..................... 904,911,925 Brady; Parks v. .............................................. 930 Bramlett; Boag v.............................................. 930 Brewer; Hampton v............................................. 915 Brickie, In re................................................ 902 TABLE OF CASES REPORTED VII Page Brotherhood. For labor union, see name of trade. Brown, In re..................................................... 903 Brown; Coleman v. .............................................. 909 Brown v. Dugger.................................................. 917 Brown v. Schweitzer.............................................. 926 Brown-Bey v. U. S. Marshal....................................... 908 Bruno v. Western Pacific R. Co. ................................. 927 Bryant v. Kansas................................................. 917 Buerkle Buick Honda Co.; Lancaster v............................. 928 Burger; New York v. ............................................. 691 Burke v. General Motors Corp..................................... 927 Burlington Northern R. Co. v. Bell............................... 919 Burrell v. Allain................................................ 910 Burris; Paine, Webber, Jackson & Curtis, Inc. v.................. 923 Business Electronics Corp. v. Sharp Electronics Corp............. 912 Butcher v. Detroit............................................... 905 Butler; Glass v.................................................. 921 Butler; Moore v.................................................. 920 Butler v. Welschmeyer............................................ 918 Butler; Wingo v.................................s........ J.. 924,925 Byrd; Arlington v................................................ 929 Cadwalader, Wickersham & Taft v. Gottlieb........................ 931 Cadwalader, Wickersham & Taft v. U. S. District Court.......... 931 Cahoon v. Alton Packaging Corp................................... 922 Caifano v. United States ........................................ 905 California; Aragon v............................................. 913 California; Charrette v.......................................... 917 California v. Smolin............................................. 400 California v. Superior Court of Cal., San Bernardino County.... 400 Camer v. Seattle Post-Intelligencer.............................. 916 Cardelle v. Delta Air Lines, Inc................................. 921 Carwile v. Kentucky.............................................. 917 Cason v. Cook.................................................... 930 Cason; Cook v. .................................................. 930 Caterpillar, Inc. v. Williams.........-.......................... 386 Champion International Corp. v. Woodworkers...................... 437 Charrette v. California.......................................... 917 Chief Judge, U. S. District Court; Frazier v..................... 641 Chirino v. Jordan Marsh Co....................................... 901 Chrans; Belk v................................................... 931 City. See name of city. Cloutier v. United States........................................ 915 Cofield v. Merit Systems Protection Bd........................... 916 Coker v. Gielow.................................................. 906 VIII TABLE OF CASES REPORTED Page Coleman v. Brown............................................. 909 Colonial Beach; Heislup v.................................... 909 Commercial Office Products Co.; EEOC v. ..................... 926 Commissioner v. Asphalt Products Co.......................... 117 Commissioner; Asphalt Products Co. v......................... 117 Commissioner v. Bollinger.................................... 913 Commissioner; Fleming v...................................... 929 Commissioner of Internal Revenue. See Commissioner. Commonwealth. See name of Commonwealth. Communications Workers v. Beck............................... 904 Connell, In re............................................... 903 Conover; Dean Witter Reynolds Inc. v......................... 923 Consolidated Rail Corp.; Monnin v..................... ,j. .. 907 Cooper, In re................................................ 926 Cooper v. Simon.............................................. 918 Corbett v. Tisch............................................. 905 Correll v. Virginia.......................................... 931 Costello v. New York City Police Dept...................... 912 Coulter v. Tennessee......................................... 914 County. See name of county. Cox; Vermont v............................................... 904 Coxon; Lyne v............................................... 905 Coyne; Fort Halifax Packing Co. v.............................. 1 Crawford Fitting Co. v. J. T. Gibbons, Inc................... 437 Crenshaw; Bankers Life & Casualty Co. v...................... 925 Cross v. Ohio................................................ 901 Cunard Line Ltd.; Ferrino v.............................. 915 Darnell v. Jasper............................................ 915 Davis v. Oklahoma............................................ 916 Davis v. United States....................................... 916 Deakins v. Monaghan...................................... 904,912 Dean Witter Reynolds Inc. v. Blumenthal...................... 923 Dean Witter Reynolds Inc. v. Conover......................... 923 DeBartolo Corp. v. Fla. Gulf Coast Bldg. & Constr. Trades Coun.. 913 Delman; Merrill Lynch, Pierce, Fenner & Smith, Inc. v........ 923 Delta Air Lines, Inc.; Cardelle v............................ 921 Demouchette v. Texas......................................... 920 Dempsey v. Wisconsin ........................................ 930 Denton; Wilson v............................................. 931 Department of Health and Human Services; Ahlberg v......... 913 Department of Health and Human Services; Sumida v.......... 913 Department of Justice v. Julian.............................. 926 Department of Revenue of Fla.; Shell Oil Co. v. ............. 904 Detroit; Butcher v. ......................................... 905 TABLE OF CASES REPORTED IX Page Di Bella v. United States....................................... 914 Director, OWCP; Mullins Coal Co. v.............................. 911 Director of penal or correctional institution. See name or title of director. District Court. See U. S. District Court. Dixon; Bowen v.................................................. 922 Dixon v. Pennsylvania........................................... 929 Doe; Honig v................................................ 912,926 Doe; Webster v.................................................. 913 Donnelly; Shedrick v. ............................... i ; ..t 906 Dorrill v. United States........................................ 917 Drane v. Mississippi ........................................... 916 Drexel Burnham Lambert, Inc. v. King............................ 922 Dugger; Brown v. ............................................... 917 Dugger v. Porter................................................ 919 Dugger; Porter v................................................ 918 Easley v. Petsock............................................... 918 Ed Miniat, Inc.; Globe Life Ins. Co. v.......................... 915 Edward J. DeBartolo Corp. v. Fla. Gulf Bldg. & Constr. Trades Coun. 913 Edwards v. Aguillard............................................ 578 Edwards; Thomas v............................................... 908 Edwards v. Turner............................................... 922 Effinger v. Greene.............................................. 911 E. F. Hutton & Co. v. Wolfe ................................... 923 Envirotech Corp.; Twin Falls v.................................. 914 Envirotech Systems, Inc.; Twin Falls v. ........................ 914 EEOC v. Commercial Office Products Co........................... 926 Ernst, In re.......................;............................ 903 Erwin; Westfall v............................................... 925 Estate. See name of party. Evans v. Oberle-Jordre Co....................................... 917 Ewers v. Garland................................................ 930 Fall River Dyeing & Finishing Corp. v. National Labor Relations Bd. 27 Fausto; United States v. ....................................... 925 FCC; Telecommunications Research & Action Center v.............. 919 Federal National Mortgage Assn.; Millsap v. .................... 902 Ferrino v. Cunard Line Ltd...................................... 915 First Biscayne Corp.; A. G. Becker Inc. v....................... 924 First English Evangelical Lutheran Church v. Los Angeles County 304 511 Detroit Street, Inc. v. Kelley.............................. 928 Fixel, In re . 910 Fixel v. Slansky................................................ 917 Flaugh v. Michigan.............................................. 906 Fleischer, In re ............................................... 903 X TABLE OF CASES REPORTED Page Fleming v. Commissioner.......................................... 929 Flight Attendants; Trans World Airlines, Inc. v.................. 913 Florida; Abbott v................................................ 924 Florida; Agan v.................................................. 932 Florida; Jackson v............................................... 920 Florida; Lusk v.................................................. 921 Florida; Mars v.................................................. 929 Florida; Miller v. .............................................. 423 Florida; Patterson v............................................. 924 Florida; Rhoden v................................................ 907 Florida; Sueiro v................................................ 922 Florida; Wilkerson v. ........................................... 924 Florida Bar; Mirabole v.......................................... 905 Florida Gulf Coast Bldg. & Constr. Trades Coun. ; DeBartolo Corp. v. 913 Foltz v. Thigpen................................................. 918 Ford Motor Co.; Vinson v......................................... 906 Ford Motor Co.; Walsh v. ........................................ 915 Fort Halifax Packing Co. v. Coyne.................................. 1 Fortin v. Michigan............................................... 928 Foster; Arkansas v............................................... 929 Frazier v. Heebe................................................. 641 Fuccillo; United States v........................................ 905 Fuerst; Turner v................................................. 917 Gabel v. Lynaugh................................................. 929 Gaggi v. United States........................................... 929 Gardebring v. Jenkins............................................ 926 Garland; Ewers v. ............................................... 930 Garris v. Lyles.................................................. 930 Gaskins v. South Carolina........................................ 909 Geinglass; Goulding v............................................ 929 General Motors Corp.; Burke v.................................... 927 Georgia; Hicks v................................................. 931 Gerns, In re..................................................... 903 Gibbons, Inc.; Crawford Fitting Co. v............................ 437 Gielow; Coker v.................................................. 906 Glantz v. United States ......................................... 929 Glass v. Butler.................................................. 921 Glenn v. Ohio ................................................... 931 Glenn v. Sowders................................................. 916 Globe Life Ins. Co. v. Ed Miniat, Inc............................ 915 Goldberg, In re ................................................. 903 Gomez v. United States........................................... 908 Goodman v. Lukens Steel Co....................................... 656 Goodman; Steelworkers v.......................................... 656 TABLE OF CASES REPORTED XI Page Gopman v. United States......................................... 917 Gordon v. Lynaugh............................................... 907 Gottlieb; Cadwalader, Wickersham & Taft v....................... 931 Goulding v. Feinglass........................................... 929 Governor of La. v. Aguillard.................................... 578 Governor of Miss.; Burrell v.................................... 910 Granviel v. Texas............................................... 902 Greene; Effinger v. ............................................ 911 Greenhaw Cotton Co. v. Merrill Lynch, Pierce, Fenner & Smith .. 928 Gresk; Perry v.................................................. 921 Griffin v. Hilke................................................ 914 Griffin; Hilke v. .............................................. 914 Griffith v. Illinois............................................ 910 Grover; Annonson v.............................................. 907 Grubbs v. Missouri ............................................. 931 Guillory v. St. Landry Parish Police Jury....................... 916 Haeberle, In re ................................................ 903 Hamilton v. Reid................................................ 930 Hammond v. Alabama.............................................. 917 Hampton v. Brewer............................................... 915 Haney v. Louisiana Training Institute........................... 906 Harris v. Indiana............................................... 909 Harrison School Dist. No. 1 v. Lewis............................ 905 Hatton v. Minnesota............................................. 911 Haynie v. Ross Gear Division of TRW, Inc........................ 901 Hazelwood School Dist. v. Kuhlmeier......................... 912,926 Heebe; Frazier v................................................ 641 Heislup v. Colonial Beach....................................... 909 Helms; Hewitt v................................................. 755 Henrich v. Texas................................................ 913 Hensiek; Watson v............................................... 906 Her Majesty the Queen in Right of Ontario v. Thomas ............ 919 Hernandez-Cano v. United States................................. 918 Hewitt v. Helms................................................. 755 Hicks v. Georgia................................................ 931 Hilke v. Griffin................................................ 914 Hilke; Griffin v. .............................................. 914 Hill; Houston v. ............................................... 451 Hoagland, In re................................................. 903 Hohri; United States v. ......................................... 64 Holy Name Hospital; LaVergne v.................................. 931 Honig v. Doe................................................ 912,926 Horan; Lay v.................................................... 910 Houston v. Hill................................................. 451 Houston; Savely v............................................... 928 XII TABLE OF CASES REPORTED Page Huiskamp v. New York Connecting R. Co............................... 915 Hutton & Co. v. Wolfe............................................... 923 Hymen v. Merit Systems Protection Bd................................ 909 lannelli v. New York................................................ 914 Illinois; Griffith v. .............................................. 910 Illinois; Lewis v................................................... 907 Illinois; Morrison v................................................ 915 Illinois; Phillips v. .............................................. 907 Illinois; Sawyer v. ................................................ 930 Illinois; Tripp v................................................... 931 Illinois Dept, of Revenue; Rockford Life Ins. Co. v................. 182 Immigration and Naturalization Service; Schellong v................. 921 Indiana; Harris v. ................................................. 909 In re. See name of party. Institute for Behavioral Res.; Int’l Primate Protection League v. .. 909 International. For labor union, see name of trade. International Distribution Centers, Inc. v. Walsh Trucking Co..... 915 International Primate Protection League v. Inst, for Behav. Res... 909 Interstate Commerce Comm’n v. Locomotive Engineers.................. 270 Interstate Commerce Comm’n; Locomotive Engineers v.................. 927 Interstate Commerce Comm’n; Transportation Union v.................. 927 Intre Sport, Ltd.; Kidder, Peabody & Co. v. ........................ 922 Iowa; Koenig v. .................................................... 917 Irving v. Mississippi............................................... 921 Jackson v. Florida ................................................. 920 Jacobson; Merrill Lynch, Pierce, Fenner & Smith, Inc. v........... 923 Jasper; Darnell v................................................... 915 Jenkins; Gardebring v............................................... 926 Jews for Jesus, Inc.; Board of Airport Comm’rs of Los Angeles v. . 569 Johns; Margoles v................................................... 905 Johnson; Bowen v.................................................... 922 Johnstone; Kelly v. ................................................ 928 Jones v. United States.............................................. 917 Jordan Marsh Co.; Chirino v. ....................................... 901 J. T. Gibbons, Inc.; Crawford Fitting Co. v......................... 437 Julian; Department of Justice v.................................... 926 Junction City; Riley v. ........................................ 911,917 Kaiser Engineers v. Allendale Mut. Ins. Co.......................... 914 Kansas; Armstrong v................................................. 929 Kansas; Bryant v. .................................................. 917 Kelley; 511 Detroit Street, Inc. v.................................. 928 Kelly v. Johnstone.................................................. 928 Kemp; McCleskey v................................................... 920 Kemp; Morris v..................................................... 907 TABLE OF CASES REPORTED xni Page Kennedy, In re................................................. 912 Kentucky; Carwile v............................................ 917 Kentucky v. Stincer............................................ 730 Kentucky; Tyler v.............................................. 929 Kidder, Peabody & Co. v. Intre Sport, Ltd...................... 922 Kile v. Pacific Telephone & Telegraph ......................... 914 Kimble v. United States........................................ 930 King; Drexel Burnham Lambert, Inc. v. ......................... 922 Klein v. United States ........................................ 909 Koenig v. Iowa................................................. 917 Kornowski, In re............................................... 911 Kowalik, In re................................................. 921 Kuhlmann; Meadows v............................................ 915 Kuhlmeier; Hazelwood School Dist. v. .................... 912,926 Kuriansky; Y v................................................. 928 Laborers Health & Welf. Tr. v. Advanced Lightweight Concrete 904 Labor Union. See name of trade. LaMothe v. Michigan............................................ 930 Lancaster v. Buerkle Buick Honda Co............................ 928 Lane; Max v..:................................................. 918 LaSalvia; United Dairymen of Ariz. v........................... 928 LaVergne v. Holy Name Hospital................................. 931 Lay v. Horan................................................... 910 LeFevre; Soto v. .............................................. 907 Lensink; Minotti v. ........................................... 906 Levinson; Basic Inc. v......................................... 904 Levy v. United States.......................................... 902 Lewis; Harrison School Dist. No. 1 v. ......................... 905 Lewis v. Illinois.............................................. 907 Lewis v. Lynaugh............................................... 908 Liquid Transporters, Inc. v. Revenue Cabinet of Ky............. 901 Little, Brown & Co.; Martin v.................................. 930 Local. For labor union, see name of trade. Locomotive Engineers v. Interstate Commerce Comm’n............. 927 Locomotive Engineers; Interstate Commerce Comm’n v............. 270 Locomotive Engineers; Missouri-Kansas-Texas R. Co. v. ......... 270 Lohmar; Ruether v.............................................. 930 Loney v. United States......................................... 908 Los Angeles County; First English Evangelical Lutheran Church v. 304 Louisiana Training Institute; Haney v.......................... 906 Lowenfield v. Phelps........................................... 932 Lucas v. Aiken................................................. 931 Lukens Steel Co.; Goodman v. .................................. 656 Lusk v. Florida................................................ 921 XIV TABLE OF CASES REPORTED Page Lyles; Garris v. ................................................ 930 Lynaugh; Gabel v................................................. 929 Lynaugh; Gordon v................................................ 907 Lynaugh; Lewis v................................................. 908 Lyne v. Coxon ................................................... 905 Malia; RCA Corp. v............................................... 927 Margoles v. Johns ............................................... 905 Marino v. Ortiz.................................................. 912 Marmott v. Maryland Lumber Co.................................... 929 Marquez v. United States ........................................ 908 Mars v. Florida.................................................. 929 Marshall v. Bauer................................................ 910 Marsh Co.; Chirino v............................................. 901 Martin v. Little, Brown & Co..................................... 930 Martir v. United States.......................................... 907 Maryland; Appleby v.............................................. 930 Maryland; Booth v. .............................................. 496 Maryland; Sears v................................................ 918 Maryland Lumber Co.; Marmott v.................................. 929 Massachusetts; Blinn v. ........................................ 921 Max v. Lane...................................................... 918 May v. Pro-Guard, Inc............................................ 932 Mayaja, Inc. v. Shearson Lehman Brothers, Inc.................... 927 Mayaja, Inc.; Shearson Lehman Brothers, Inc. v................... 923 Mays v. United States ........................................... 913 May wood; Vaden v................................................ 908 McCleskey v. Kemp................................................ 920 McDermott v. Tate................................................ 928 McHarris v. Spears............................................... 932 McKeever; Ratcliff v. ........................................... 921 McKenna; Wilson v................................................ 918 McMahon; Shearson/American Express Inc. v.................... 220 Meadows v. Kuhlmann.............................................. 915 Melton v. United States.......................................... 916 Merit Systems Protection Bd.; Cofield v....................... 916 Merit Systems Protection Bd.; Hymen v. ...................... 909 Merrill Lynch, Pierce, Fenner & Smith v. Badart.................. 923 Merrill Lynch, Pierce, Fenner & Smith v. Delman.................. 923 Merrill Lynch, Pierce, Fenner & Smith v. Jacobson................ 923 Merrill Lynch, Pierce, Fenner & Smith; Phillips v................ 931 Merrill Lynch, Pierce, Fenner & Smith; Smoky Greenhaw Cotton v. 928 Metropolitan Atlanta Rapid Transit Authority; Attwell v.... 908 Metropolitan Seattle; Batayola v................................. 914 Michigan; Flaugh v.t............................... 906 TABLE OF CASES REPORTED xv Page Michigan; Fortin v. ............................................ 928 Michigan; LaMothe v............................................. 930 Miller v. Florida............................................... 423 Millsap v. Federal National Mortgage Assn....................... 902 Miniat, Inc.; Globe Life Ins. Co. v............................. 915 Minnesota; Hatton v............................................. 911 Minotti v. Lensink.............................................. 906 Mirabole v. Florida Bar......................................... 905 Misenko v. Wisconsin............................................ 916 Mississippi; Drane v............................................ 916 Mississippi; Irving v........................................... 921 Mississippi; Mississippi Power & Light Co. v..................... 902 Mississippi Power & Light Co. v. Mississippi............... 902 Missouri; Grubbs v.............................................. 931 Missouri; Schlup v. ............................................ 920 Missouri-Kansas-Texas R. Co. v. Locomotive Engineers........... 270 Missouri Pacific R. Co.; Transportation Union v................. 927 Monaghan; Deakins v......................................... 904,912 Monnin v. Consolidated Rail Corp................................ 907 Monsour Medical Center v. Bowen ................................ 905 Montgomery v. United States..................................... 918 Moore v. Butler................................................. 920 Moore v. Ohio................................................... 918 Moore v. State Bank of Burden................................... 906 Moore v. United States......................................... 918 Morgan Guaranty Trust Co. ; American Savings & Loan Assn. v. .. 929 Morris v. Kemp.................................................. 907 Morrison v. Illinois ........................................... 915 Mosley; New York v. ............................................ 914 Mullins Coal Co. v. Director, OWCP.............................. 911 Multibanco Comermex, S. A.; West v.............................. 906 NLRB; Fall River Dyeing & Finishing Corp. v. .................... 27 NLRB; Quality Aluminum Products, Inc. v......................... 924 Neuroth v. United States........................................ 916 Nevada; Woolery v............................................... 914 New Jersey; Reiboldt v.......................................... 928 New Mexico; Texas v............................................. 124 New York v. Burger.............................................. 691 New York; lannelli v............................................ 914 New York v. Mosley.............................................. 914 New York; Polur v............................................... 921 New York; Standifer v. ......................... s..r.......i.. 907 New York v. Thomas ............................................. 919 New York City Loft Bd.; Spring Realty Co. v.................... 911 XVI TABLE OF CASES REPORTED Page New York City Police Dept.; Costello v........................... 912 New York Connecting R. Co.; Huiskamp v........................... 915 Nicolet Instrument Corp. v. Bio-Rad Laboratories, Inc............ 915 Oberle-Jordre Co.; Evans v....................................... 917 Oglala Sioux Tribe v. United States.............................. 913 Ohio; Cross v.................................................... 901 Ohio; Glenn v.................................................... 931 Ohio; Moore v. .................................................. 918 Oklahoma; Battles v.............................................. 917 Oklahoma; Davis v. .............................................. 916 Oklahoma; Ross v................................................. 926 Oklahoma; Webb v................................................. 930 O’Lone v. Shabazz’ Estate........................................ 342 Orlowski v. United States........................................ 927 Ortiz; Marino v.................................................. 912 Owens v. Adams................................................... 907 Oxfeld, In re.................................................... 903 Pace Resources, Inc. v. Shrewsbury............................... 906 Pacific First Federal Savings Bank v. Rembold.................... 905 Pacific Telephone & Telegraph; Kile v............................ 914 Page v. United States............................................ 918 Paine, Webber, Jackson & Curtis, Inc. v. Burris.................. 923 Palomo, In re.................... .............................. 911 Parks v. Brady................................................... 930 Patterson v. Florida............................................. 924 Patterson v. South Carolina...................................... 902 Pennsauken; Sweetman v........................................... 907 Pennsylvania; Dixon v............................................ 929 Pennsylvania; Strelkoff v. ...................................... 929 Pennsylvania; Terry v............................................ 920 Perry v. Astrike............................................... 921 Perry v. Gresk.................................................. 921 Perry v. Thomas................................................. 483 Petsock; Easley v............................................... 918 Phelps; Berry v.................................................. 910 Phelps; Lowenfield v. ........................................... 932 Phillips v. Illinois............................................. 907 Phillips v. Merrill Lynch, Pierce, Fenner & Smith, Inc........... 931 Plumbers & Steamfitters v. Wash. Public Power Supply System 905 Polaski; Bowen v................................................. 927 Polo Fashions, Inc. v. Stock Buyers International, Inc........... 905 Polur v. New York................................................ 921 Portal v. United States.......................................... 908 Porter v. Dugger................................................. 918 Porter; Dugger v................................................. 919 TABLE OF CASES REPORTED XVII Page Postmaster General; Corbett v. ................................. 905 Pro-Guard, Inc.; May v.......................................... 932 Pruett v. United States......................................... 931 Pruett v. Virginia.............................................. 931 Pulido-Zorrilla v. United States................................ 913 Quality Aluminum Products, Inc. v. National Labor Relations Bd. . 924 Raffoul v. United States........................................ 908 Ramirez v. United States........................................ 916 Ratcliff v. McKeever............................................ 921 Rawson; Steelworkers v.......................................... 901 Rawwad v. United States ........................................ 909 Rayburn v. United States........................................ 906 RCA Corp. v. Malia ............................................. 927 Reiboldt v. New Jersey.......................................... 928 Reid; Hamilton v................................................ 930 Rembold; Pacific First Federal Savings Bank v................... 905 Revenue Cabinet of Ky.; Liquid Transporters, Inc. v............. 901 Rhoden v. Florida............................................... 907 Riley v. Junction City...................................... 911,917 Rival Mfg. Co.; Averbach v...................................... 915 Robinson v. United States....................................... 921 Rockford Life Ins. Co. v. Illinois Dept, of Revenue ............ 182 Rodrigues v. United States........................................ 908 Rogers; Vester v................................................ 916 Rohrbach v. United States......................................... 909 Roque-Pino v. United States....................................... 908 Ross v. Oklahoma................................................ 926 Ross Gear Division of TRW, Inc.; Haynie v....................... 901 Rothballer; Wanless v........................................... 929 Roy, In re...................................................... 912 Ruether v. Lohmar................................................. 930 Russell v. Shearson Lehman Brothers, Inc.......................... 927 Safley; Turner v. ............................................... 78 St. Landry Parish Police Jury; Guillory v......................... 916 Satterwhite v. Texas.............................................. 905 Savely v. Houston............................................... 928 Sawyer v. Illinois................................................ 930 Schellong v. Immigration and Naturalization Service............... 921 Schlup v. Missouri................................................ 920 Schweitzer; Brown v. ............................................. 926 Scully; Washington v............................................ 907 Sears v. Maryland............................................... 918 Seattle Post-Intelligencer; Camer v............................... 916 Secretary of Health and Human Services v. Dixon................... 922 XVIII TABLE OF CASES REPORTED Page Secretary of Health and Human Services v. Johnson................ 922 Secretary of Health and Human Services; Monsour Medical Center v. 905 Secretary of Health and Human Services v. Polaski.............. 927 Secretary of Health and Human Services; Taylor v............... 913 Secretary of Health and Human Services v. Wilson .............. 923 Secretary of Health and Human Services v. Yuckert.............. 137 Secretary of Navy; Torres v...................................... 927 Setera v. Texas A & M Univ....................................... 928 Shabazz’ Estate; O’Lone v. ...................................... 342 Share, In re .................................................... 902 Sharp Electronics Corp.; Business Electronics Corp. v............ 912 Shearson/American Express Inc. v. McMahon........................ 220 Shearson Lehman Brothers, Inc. v. Mayaja, Inc.................... 923 Shearson Lehman Brothers, Inc.; Mayaja, Inc. v................... 927 Shearson Lehman Brothers, Inc.; Russell v........................ 927 Shedrick v. Donnelly............................................. 906 Shell Oil Co. v. Department of Revenue of Fla.................... 904 Shrewsbury; Pace Resources, Inc. v. ............................. 906 Shuman; Bourke v. ............................................... 906 Simon; Cooper v.................................................. 918 Skurdal v. Billings.............................................. 910 Slansky; Fixel v................................................. 917 Smoky Greenhaw Cotton v. Merrill Lynch, Pierce, Fenner & Smith 928 Smolin; California v. ........................................... 400 Société Nationale Industrielle Aérospatiale v. U. S. District Court 522 Soto v. LeFevre.................................................. 907 South Carolina; Gaskins v........................................ 909 South Carolina; Patterson v...................................... 902 Sowders; Glenn v. ............................................... 916 Spears; McHarris v. ............................................. 932 Spring Realty Co. v. New York City Loft Bd....................... 911 Standifer v. New York............................................ 907 State. See name of State. State Bank of Burden; Moore v.................................... 906 Steelworkers v. Goodman.......................................... 656 Steelworkers v. Rawson........................................... 901 Sterner v. Wayne County Dept, of Health.......................... 921 Stincer; Kentucky v.............................................. 730 Stock Buyers International, Inc.; Polo Fashions, Inc. v. ........ 905 Strelkoff v. Pennsylvania........................................ 929 Sueiro v. Florida................................................ 922 Sumida v. Department of Health and Human Services................ 913 Superintendent of penal or correctional institution. See name or title of superintendent. TABLE OF CASES REPORTED xix Page Superior Court of Cal., San Bernardino County; California v.. 400 Sweetman v. Pennsauken....................................... 907 Tarver v. Alabama............................................ 920 Tate; McDermott v............................................ 928 Taylor v. Bowen.............................................. 913 Telecommunications Research & Action Center v. FCC........... 919 Tennessee; Adkins v.......................................... 909 Tennessee; Coutler v......................................... 914 Terry v. Pennsylvania........................................ 920 Texas; Demouchette v......................................... 920 Texas; Granviel v............................................ 902 Texas; Henrich v............................................. 913 Texas v. New Mexico.......................................... 124 Texas; Satterwhite v......................................... 905 Texas v. Williams............................................ 902 Texas A & M Univ.; Setera v.................................. 928 Thigpen; Foltz v............................................. 918 Thomas v. Edwards............................................ 908 Thomas; Her Majesty the Queen in Right of Ontario v.......... 919 Thomas; New York v........................................... 919 Thomas; Perry v.............................................. 483 Thomas v. United States...................................... 930 Thompson v. United States.................................... 918 Tisch; Corbett v. ........................................... 905 Tison v. Arizona............................................. 921 Torres v. Webb............................................... 927 Town. See name of town. Transportation Union v. Interstate Commerce Comm’n........... 927 Transportation Union v. Missouri Pacific R. Co............... 927 Trans World Airlines, Inc. v. Flight Attendants.............. 913 Treasury Employees v. Von Raab........................... 902,912 Tripp v. Illinois............................................ 931 Trullo v. United States...................................... 916 Turner; Edwards v............................................ 922 Turner v. Fuerst............................................. 917 Turner v. Safley.............................................. 78 Twin Falls v. Envirotech Corp................................ 914 Twin Falls v. Envirotech Systems, Inc........................ 914 Tyler, In re................................................. 912 Tyler v. Kentucky............................................ 929 Union. For labor union, see name of trade. Union Pacific R. Co.; Walsh v................................ 928 United. For labor union, see name of trade. United Dairymen of Ariz. v. LaSalvia ........................ 928 XX TABLE OF CASES REPORTED Page United States. See name of other party. U. S. District Court; Cadwalader, Wickersham & Taft v........ 931 U. S. District Court; Société Nationale Industrielle Aérospatiale v. 522 U. S. District Court; Yamhill Yamaha, Inc. v................... 906 U. S. International Trade Comm’n; Akzo N. V. v. ............... 909 U. S. Marshal; Brown-Bey v..................................... 908 United Technologies Corp.; Boyle v...................... 904,911,925 United Technologies Corp.; Velilla v............................ 921 Utah Division of State Lands v. United States................... 193 Vaden v. Maywood ............................................... 908 Velilla v. United Technologies Corp............................. 921 Vermont v. Cox.................................................. 904 Vester v. Rogers................................................ 916 Via v. Williams................................................. 914 Village. See name of village. Vinson v. Ford Motor Co......................................... 906 Virginia; Correll v............................................. 931 Virginia; Pruett v.............................................. 931 Von Raab; Treasury Employees v. ............................ 902,912 Walsh v. Ford Motor Co.......................................... 915 Walsh v. Union Pacific R. Co.................................... 928 Walsh Trucking Co.; International Distribution Centers, Inc. v. ... 915 Wanless v. Rothballer........................................... 929 Warden. See name of warden. Washington v. Scully............................................ 907 Washington Public Power Supply Sys.; Plumbers & Steamfitters v. 905 Watkins v. United States........................................ 909 Watson v. Hensiek............................................... 906 Wayne County Dept, of Health; Sterner v. ....................... 921 Webb v. Oklahoma................................................ 930 Webb; Torres v.................................................. 927 Webster v. Doe.................................................. 913 Wechsler, In re................................................. 911 Wells Fargo Bank; United States v............................... 925 Welschmeyer; Butler v. ......................................... 918 West v. Multibanco Comermex, S. A............................... 906 Western Pacific R. Co.; Bruno v................................. 927 Westfall v. Erwin............................................... 925 Wetzel v. Arizona State Real Estate Dept........................ 914 Wilkerson v. Florida............................................ 924 Williams v. Blackburn........................................... 917 Williams; Caterpillar, Inc. v................................... 386 Williams; Texas v............................................... 902 Williams v. United States ...................................... 906 TABLE OF CASES REPORTED XXI Page Williams; Via v. ............................................... 914 Wilson; Bowen v................................................. 923 Wilson v. Denton................................................ 931 Wilson v. McKenna............................................... 918 Wingo v. Butler............................................. 924,925 Wisconsin; Dempsey v............................................ 930 Wisconsin; Misenko v............................................ 916 Wolfe; E. F. Hutton & Co. -y.................................... 923 Wolfe v. United States.......................................... 927 Woodcock v. United States....................................... 908 Woodworkers; Champion International Corp. v. ................... 437 Woolery v. Nevada............................................... 914 Y v. Kuriansky.................................................. 928 Yamaha Motor Corp.; Yamhill Yamaha, Inc. v................... 906 Yamhill Yamaha, Inc. v. U. S. District Court.................... 906 Yamhill Yamaha, Inc. v. Yamaha Motor Corp.................... 906 Young v. United States.......................................... 913 Yuckert; Bowen v................................................ 137 TABLE OF CASES CITED Page Abdul Wali v. Coughlin, 754 F. 2d 1015 101, 349, 354, 358, 359 Abel v. United States, 362 U.S. 217 724 Abington School Dist. v. Schempp, 374 U.S. 203 583- 585, 587, 589, 590, 606, 608, 615 Abrams v. United States, 449 F. 2d 662 120, 121 Acker v. United States, 298 U.S. 426 292 Agins v. Tiburon, 447 U.S. 255 311, 314, 319, 320, 329, 330, 332, 333 Agins v. Tiburon, 24 Cal. 3d 266 308- 314, 317, 324, 337, 338 Aguilar v. Felton, 473 U.S. 402 640 Air France v. Saks, 470 U.S. 392 534 Alabama v. King & Boozer, 314 U.S. 1 191 Alabama Elec. Cooperative, Inc. v. First National Bank, 684 F. 2d 789 74 Alberto-Culver Co. v. Scherk, 484 F. 2d 611 268 Alessi v. Raybestos-Manhat-tan, Inc., 451 U.S. 504 10, 19 Alexander v. Gardner-Denver Co., 415 U.S. 36 689 Al-Khazraji v. St. Francis College, 784 F. 2d 505 663 Allegaert v. Perot, 548 F. 2d 432 225, 248 Allen v. City and County of Honolulu, 58 Haw. 432 341 Allis-Chalmers Corp. v. Lueck, 471 U.S. 202 394-396 Almeida-Sanchez v. United States, 413 U.S. 266 724, 727 Alyeska Pipeline Service Co. v. Wilderness Society, 421 U.S. 240 439, 440, 444, 449, 450 Page American Farm Lines v. Black Ball Freight Service, 397 U.S. 532 285 Amoco Production Co. v. Gambell, 480 U.S. 531 212 Anderson v. Bessemer City, 470 U.S. 564 665 Anderson v. Winsett, 449 U.S. 1093 378 Andrus v. Allard, 444 U.S. 51 330, 340 Anschuetz & Co., GmbH, In re, 754 F. 2d 602 536, 539, 541, 553 Appleyard v. Massachusetts, 203 U.S. 222 419 Application of Varona, 38 Wash. 2d 833 410 Aptos Seascape Corp. v. County of Santa Cruz, 138 Cal. App. 3d 484 312 Arizona v. California, 283 U.S. 423 208, 218 Arizona v. California, 373 U.S. 546 202, 218 Arlington Heights v. Metropolitan Housing Dev. Corp., 429 U.S. 252 595, 685 Armstrong v. United States, 364 U.S. 40 315, 319 Atari, Inc. v. JS & A Group, Inc., 747 F. 2d 1422 69 Atchison, T. & S. F. R. Co. v. United States, 284 U.S. 248 278, 292 Avco Corp. v. Machinists, 390 U.S. 557 391, 393, 394 Axelrod & Co. v. Kordich, Victor & Neufeld, 451 F. 2d 838 235, 237 Axelrod & Co. v. Kordich, Victor & Neufeld, 320 F. Supp. 193 232 Ayres v. Merrill Lynch, Pierce, Fenner & Smith, Inc., 538 F. 2d 532 248, 264 XXIII XXIV TABLE OF CASES CITED Page Babbitt v. Farm Workers, 442 U.S. 289 467, 468, 575 Baeder v. Heckler, 768 F. 2d 547 156, 157, 166 Baggett v. Bullitt, 377 U.S. 360 467, 468, 471, 575 Barber v. Page, 390 U.S. 719 738 Barnett v. Rodgers, 133 U.S. App. D. C. 296 368 Baumann v. Arizona Dept, of Corrections, 754 F. 2d 841 378 Beazell v. Ohio, 269 U.S. 167 429, 430 Belgenland, The, 114 U.S. 355 554 Bell v. Wolfish, 441 U.S. 520 87- 90, 95, 348-350, 355 Bellotti v. Baird, 428 U.S. 132 470, 475 Berard v. State of Vt. Parole Bd., 730 F. 2d 71 379 Berizzi Bros. Co. v. The Pesaro, 271 U.S. 562 554 Berman v. Parker, 348 U.S. 26 321 Bethel School Dist. No. 403 v. Fraser, 478 U.S. 675 583, 584 Beth Israel Hospital v. NLRB, 437 U.S. 483 42 B. F. Goodrich Co. v. Northwest Industries, Inc., 424 F. 2d 1349 293 Biddinger v. Commissioner of Police, 245 U.S. 128 407, 408, 411, 416, 417 Bionic Auto Parts & Sales, Inc. v. Fahner, 721 F. 2d 1072 699 Block v. Rutherford, 468 U.S. 576 87, 89, 90, 353 Blue Chip Stamps v. Manor Drug Stores, 421 U.S. 723 69, 597 Bluvband v. Heckler, 730 F. 2d 886 147 Board of Comm’rs of Montgomery County v. Elston, 32 Ind. 27 190 Board of Ed. v. Allen, 392 U.S. 236 615, 616, 627 Board of Ed. v. Pico, 457 U.S. 853 583, 605 Page Board of Regents v. Roth, 408 U.S. 564 381, 382 Boothe v. Hammock, 605 F. 2d 661 379 Boreri v. Fiat S. P. A., 763 F. 2d 17 554 Bowen v. City of New York, 476 U.S. 467 142, 150 Bowman Transportation, Inc. v. Arkansas-Best Freight System, Inc., 419 U.S. 281 278 Boys Market, Inc. v. Retail Clerks, 398 U.S. 235 268 Brady v. Heckler, 724 F. 2d 914 156, 158 Bremen, The v. Zapata Off-Shore Co., 407 U.S. 1 554 Brener v. Becker Paribas Inc., 628 F. Supp. 442 260 Broadrick v. Oklahoma, 413 U.S. 601 456, 458, 574 Brockett v. Spokane Arcades, Inc., 472 U.S. 491 469, 475, 574, 609 Brooks v. NLRB, 348 U.S. 96 37, 38, 41, 50 Brotherhood of Locomotive Engineers v. Chicago & N. W. R. Co., 314 F. 2d 424 300 Brown v. Duchesne, 19 How. 183 554 Brown v. Gillian, Will & Co., 287 F. Supp. 766 233 Brown v. Heckler, 786 F. 2d 870 145, 156, 165 Brown v. Herald Co., 464 U.S. 928 926 Brown Shoe Co. v. United States, 370 U.S. 294 475, 481 Bryant v. Wilkins, 258 N. Y. S. 2d 455 363 Burlington Northern Inc. v. United States, 459 U.S. 131 275 Burlington Northern R. Co. v. Woods, 480 U.S. 1 653 Burlington Truck Lines, Inc. v. United States, 371 U.S. 156 290 Burr, Ex parte, 9 Wheat. 529 646, 655 Butler v. Wilson, 415 U.S. 953 96 Calder v. Bull, 3 Dall. 386 429,430 TABLE OF CASES CITED xxv Page Califano v. Sanders, 430 U .S. 99 292,293 California v. Green, 399 U.S. 149 736-738, 740, 748, 750 California v. Ramos, 463 U.S. 992 502 California v. United States, 438 U.S. 645 198, 199, 208 Camara v. Municipal Court, 387 U.S. 523 722, 724, 725 Candelaria v. Griffin, 641 F. 2d 868 379 Cantwell v. Connecticut, 310 U.S. 296 583 Cappadora v. Celebrezze, 356 F. 2d 1 293 Capps v. Commonwealth, 560 S. W. 2d 559 741, 742 Carter/Mondale Presidential Committee, Inc. v. FEC, 249 U.S. App. D. C. 349 292 Case Co. v. NLRB, 321 U.S. 332 395, 396 Cavanaugh v. Crist, 189 Mont. 274 377 Celotex Corp. v. Catrett, 477 U.S. 317 595 Chaplinsky v. New Hampshire, 315 U.S. 568 462, 464, 472, 479 Chardon v. Fumero Soto, 462 U.S. 650 609 Charles v. Diamond, 41 N. Y. 2d 318 341 Chevron Oil Co. v. Huson, 404 U.S. 97 662, 663 Chicago, B. & Q. R. Co. v. Chicago, 166 U.S. 226 310 Choctaw Nation v. Oklahoma, 397 U.S. 620 198, 212 Choctaw Nation v. United States, 318 U.S. 423 534 Christiansburg Garment Co. v. EEOC, 434 U.S. 412 446 Citizens to Preserve Overton Park v. Volpe, 401 U.S. 402 382 City. See name of city. City Council v. Taxpayers for Vincent, 466 U.S. 789 574 CAB v. Delta Air Lines, Inc., 367 U.S. 316 285 Page Claire v. Cuyler, 634 F. 2d 109 367 Clearly v. American Airlines, Inc., Ill Cal. App. 3d 443 389 Cleavinger v. Saxner, 474 U.S. 193 371 Clements v. Fashing, 457 U.S. 957 618 Coates v. Cincinnati, 402 U.S. 611 465 Coenen v. R. W. Pressprich & Co., 453 F. 2d 1209 253 Cohen v. California, 403 U.S. 15 576 Colgrove v. Battin, 413 U.S. 149 654 Colonnade Catering Corp. v. United States, 397 U.S. 72 700-702, 717, 719 Colorado River Water Conservation Dist. v. United States, 424 U.S. 800 467 Colten v. Kentucky, 407 U.S. 104 463^65 Committee for Public Ed. & Religious Liberty v. Nyquist, 413 U.S. 756 597, 606, 613, 614, 616, 617, 627 Committee for Public Ed. & Religious Liberty v. Regan, 444 U.S. 646 613, 627, 640 Commonwealth. See also name of Commonwealth. Commonwealth v. Lipomi, 385 Mass. 370 727 Cone Memorial Hospital v. Mercury Construction Corp., 460 U.S. 1 226, 232, 489, 492 Connecticut Bd. of Pardons v. Dumschat, 452 U.S. 458 373, 387, 383 Conover v. Dean Witter Reyn- olds, Inc., 794 F. 2d 520 225, 245, 249 Container Corp. v. Franchise Tax Bd., 463 U.S. 159 553 Cook v. Georgetown Steel Corp., 770 F. 2d 1272 399 County. See name of county. Cox Broadcasting Corp. v. Cohn, 420 U.S. 469 310 XXVI TABLE OF CASES CITED Page Cruz v. Beto, 405 U.S. 319 348, 365 Cupp v. Naughten, 414 U.S. 141 651 Curtis v. Loether, 415 U.S. 189 677 Curtiss-Wright Corp., Wright Aeronautical Div. v. NLRB, 347 F. 2d 61 54 Dace v. Mickelson, 797 F. 2d 574 379, 380 Daimler-Benz Aktiengesellschaft v. United States District Court, 805 F. 2d 340 541 Dandridge v. Williams, 397 U.S. 471 324, 747 Danforth v. United States, 308 U.S. 271 319, 320, 333 Davis v. Alaska, 415 U.S. 308 736, 738, 749 Davis v. Scherer, 468 U.S. 183 341 Davis v. United States Steel Supply, Div. of United States Steel Corp., 581 F. 2d 335 662 Davis v. Weber, 93 Ariz. 312 743 Dean Witter Reynolds Inc. v. Byrd, 470 U.S. 213 224- 227, 232, 248, 249, 256, 268, 488, 490, 492 De Lancie v. Birr, Wilson & Co., 648 F. 2d 1255 248 Delaware v. Fensterer, 474 U.S. 15 737, 739 Delaware v. Van Arsdall, 475 U.S. 673 738 District of Columbia v. Carter, 409 U.S. 418 676 Dixon v. Heckler, 785 F. 2d 1102 156 Dobbert v. Florida, 432 U.S. 282 428-133, 435 Dock v. Latimer, 729 F. 2d 1287 379 Doe v. Bolton, 410 U.S. 179 459 Dombrowski v. Pfister, 380 U.S. 479 467, 476 Donovan v. Dewey, 452 U.S. 594 700-703, 705, 710, 711, 719, 721-724 Donovan v. Dillingham, 688 F. 2d 1367 12 Page Douglas v. Alabama, 380 U.S. 415 736, 748 Douglas v. United States, 206 Ct. Cl. 96 525 Dow Pharmaceuticals, Inc. v. Thompson, 478 U.S. 804 392 Drayer v. Krasner, 572 F. 2d 348 487 Drew v. Thaw, 235 U.S. 432 408, 411-415, 417 Dudley v. Textron, Inc., 386 F. Supp. 602 679 Dufresne v. Baer, 744 F. 2d 1543 434, 435 Dupree v. Hertz Corp., 419 F. Supp. 764 679 Dyer v. Sims, 341 U.S. 22 128 Eastport S.S. Corp. v. United States, 178 Ct. Cl. 599 72 Eccles v. Peoples Bank of Lake- wood, 333 U.S. 426 762 Eddings v. Oklahoma, 455 U.S. 104 502, 517 Electrical Workers v. Foust, 442 U.S. 42 689 Electrical Workers v. Hechler, 481 U.S. 851 394 Electrical Workers v. NLRB, 341 U.S. 694 688 Ellis v. Dyson, 421 U.S. 426 459 Emory v. Grenough, 3 Dall. 369 543 Engle v. Vitale, 370 U.S. 421 585, 596, 605 Enmund v. Florida, 458 U.S. 782 502, 519 Epperson v. Arkansas, 393 U.S. 97 582, 585, 590, 593-595, 599, 603, 604, 608, 613, 614, 630, 638 Erie R. Co. v. Tompkins, 304 U.S. 64 491 Ernst & Ernst v. Hochfelder, 425 U.S. 185 256 Erznoznik v. City of Jacksonville, 422 U.S. 205 314 Espinoza v. Farah Mfg. Co., 414 U.S. 86 178 Estran v. Heckler, 745 F. 2d 340 156, 158 TABLE OF CASES CITED XXVII Page Evans v. Heckler, 734 F. 2d 1012 156, 158 Everson v. Board of Ed., 330 U.S. 1 605, 606 Ex parte. See name of party. Fair, The v. Kohler Die & Specialty Co., 228 U.S. 22 392 Faretta v. California, 422 U.S. 806 753, 754 Farmer v. Arabian American Oil Co., 379 U.S. 227 442, 447-449 Farris v. Secretary of Health and Human Services, 773 F. 2d 85 145, 156 FPC v. Texaco, Inc., 417 U.S. 380 290 Federated Department Stores, Inc. v. Moitie, 452 U.S. 394 397 First National Bank of Atlanta v. Bartow County Tax Assessors, 470 U.S. 583 183 First National Bank of Guthrie Center v. Anderson, 269 U.S. 341 325 First National City Bank v. Banco Para el Comercio Exterior de Cuba, 462 U.S. 611 555 Fletcher v. Peck, 6 Cranch 87 638 Flynn v. Heckler, 768 F. 2d 1273 145 Ford Motor Co. v. EEOC, 458 U.S. 219 687 Ford Motor Co. v. Huffman, 345 U.S. 330 689 Franchise Tax Bd. of Cal. v. Construction Laborers Vacation Trust for Southern Cal., 463 U.S. 1 391-394, 396 Francis v. Heckler, 749 F. 2d 1562 144 Frank v. Maryland, 359 U.S. 360 725 Franks Bros. Co. v. NLRB, 321 U.S. 702 50 French v. Merrill Lynch, Pierce, Fenner & Smith, Inc., 784 F. 2d 902 259 Furman v. Georgia, 408 U.S. 238 506, 515 Furnco Construction Corp. v. Waters, 438 U.S. 567 687 Page Gale v. Moore, 763 F. 2d 341 379 Gannett Co. v. DePasquale, 443 U.S. 368 739 Gardner v. Florida, 430 U.S. 349 507, 508 Garland, Ex parte, 4 Wall. 333 655 Garment Workers v. NLRB, 366 U.S. 731 * 51 Garner v. Teamsters, 346 U.S. 485 20 Gebr. Eickhoff Maschinenfabrik und Eisengieberei mbH v. Starcher, 328 S. E. 2d 492 536, 538, 549 General Box Co. v. United States, 351 U.S. 159 609 General Building Contractors Assn., Inc. v. Pennsylvania, 458 U.S. 375 667, 674, 684 Geneseo, Inc. v. T. Kakiuchi & Co., 815 F. 2d 840 238 Gilbert v. Burlington Industries, Inc., 765 F. 2d 320 7, 17, 19, 24, 25 Gillespie v. United States Steel Corp., 379 U.S. 148 475 Gillette v. United States, 401 U.S. 437 618 Giova v. Rosenberg, 379 U.S. 18 291 Godfrey v. Georgia, 446 U.S. 420 506 Goff v. State, 139 Mont. 641 377 Goldblatt v. Hempstead, 369 U.S. 590 313, 325 Golden v. Zwickler, 394 U.S. 103 459 Golden State Bottling Co. v. NLRB, 414 U.S. 168 40, 42-44, 46, 56, 58 Golden State Transit Corp. v. City of Los Angeles, 475 U. S. 608 19, 20 Goldschmidt A. G. v. Smith, 676 S. W. 2d 443 549 Gooding v. Wilson, 405 U.S. 518 462, 464, 472, 474 Goodrich Co. v. Northwest Industries, Inc., 424 F. 2d 1349 293 XXVIII TABLE OF CASES CITED Page Gore v. United States, 357 U.S. 386 515 Graco, Inc. v. Kremlin, Inc. 101 F. R. D. 503 567 Grand Rapids School Dist. v. Ball, 473 U.S. 373 584, 597, 613, 616, 628, 640 Graver Mfg. Co. v. Linde Co., 336 U.S. 271 665 Graves v. New York ex rel. O’Keefe, 306 U.S. 466 191 Greater Continental Corp. v. Schechter, 422 F. 2d 1100 225, 237, 243 Great North R. Co. v. Alexan- der, 246 U.S. 276 392 Green v. Black, 755 F. 2d 687 380 Greenholtz v. Nebraska Penal Inmates, 442 U.S. 1 371- 379, 381, 385 Gregg v. Georgia, 428 U.S. 153 502, 515, 909, 910, 920, 921, 924, 925, 931, 932 Gregory v. Chicago, 394 U.S. Ill 465 Greyhound Corp. v. ICC, 215 U.S. App. D. C. 322 289 Grifaldo v. State, 182 Mont. 287 377 Griggs v. Duke Power Co., 401 U.S. 424 686 Gulf Offshore Co. v. Mobil Oil Corp., 453 U.S. 473 662 Gully v. First National Bank, 299 U.S. 109 392 Hadacheck v. Sebastian, 239 U.S. 394 313, 325 Haffner v. Dobrinski, 215 U.S. 446 131 Haines v. Kerner, 404 U.S. 519 84 Hampton v. Bowen, 785 F. 2d 1308 145 Hanna v. Plumer, 380 U.S. 460 653 Hanrahan v. Hampton, 446 U.S. 754 760 Hansen v. Heckler, 783 F. 2d 170 145, 156, 164, 165 Haring v. Prosise, 462 U.S. 306 609 Page Harlow v. Fitzgerald, 457 U.S. 800 341, 758 Harman v. Forssenius, 380 U.S. 528 468, 476, 468, 469, 575 Harris v. McRae, 448 U.S. 297 605, 615 Hart, In re, 145 Mont. 203 377 Hawaii Housing Authority v. Midkiff, 467 U.S. 229 321, 468, 469 Hawk, Ex parte, 321 U.S. 114 339 Heckler v. Campbell, 461 U.S. 458 145, 146, 150, 153, 166, 174, 747 Heckler v. Chaney, 470 U.S. 821 282, 289, 294, 382 Helms v. Hewitt, 655 F. 2d 487 758, 760-764 Hendricks v. Commonwealth, 550 S. W. 2d 551 741, 743 Henkel v. Chicago, S. P., M. & O. R. Co., 284 U.S. 444 443, 444, 448 Herman v. Powell, 139 Mont. 583 377 Herman & MacLean v. Huddleston, 459 U.S. 375 246- 248, 256 Hewitt v. Helms, 459 U.S. 460 373, 374, 378, 379, 383 Hibernia Savings Society v. San Francisco, 200 U.S. 310 191 Hickman v. Taylor, 329 U.S. 495 540 Hilton v. Guyot, 159 U.S. 113 546 Hinckley v. Peay, 22 Utah 21 205 Hobbie v. Unemployment Appeals Comm’n of Fla., 480 U.S. 136 617, 618 Hodel v. Irving, 481 U.S. 704 340 Hodel v. Virginia Surface Mining & Reclamation Assn., Inc., 452 U.S. 264 314, 329 Hoffman Estates v. The Flipside, Hoffman Estates, Inc., 455 U.S. 489 458 Hogan v. O’Neill, 255 U.S. 52 421 Holland v. Burlington Industries, Inc., 772 F. 2d 1140 7, 17, 19, 26 Hopt v. Utah, 110 U.S. 574 433 TABLE OF CASES CITED XXIX Page Houston v. Hill, 482 U.S. 451 574, 576 Howard Johnson Co. v. Hotel Employees, 417 U.S. 249 40, 44, 46, 59 Huggins v. Isenbarger, 798 F. 2d 203 379 Hughes v. Washington, 389 U.S. 290 334 Hurley v. Kincaid, 285 U.S. 95 314, 336 Hyatt v. People ex rel. Corkran, 188 U.S. 691 406, 413 Icicle Seafoods, Inc. v. Worthington, 475 U.S. 709 101 Illinois v. ICC, 713 F. 2d 305 289 Illinois v. Krull, 480 U.S. 340 618, 717 Illinois Central R. Co. v. Illinois, 146 U.S. 387 203 Illinois ex rel. McCollum v. Board of Ed., 333 U.S. 203 584 INS v. Bagamasbad, 429 U.S. 24 289 INS v. Jong Ha Wang, 450 U.S. 139 292 INS v. Rios-Pineda, 471 U.S. 444 292 Ingbar v. Drexel Burnham Lambert Inc., 683 F. 2d 603 248 In re. See name of party or proceeding. International Society for Krishna Consciousness, Inc. v. Lee, 105 F. R. D. 435 567 International Union, United Auto, Aerospace, & Ag. Implement Workers v. Brock, 251 U.S. App. D. C. 239 289 ICC v. Jersey City, 322 U.S. 503 278,292 Irving v. Thigpen, 732 F. 2d 1215 379 Jackson v. Reese, 608 F. 2d 159 379 Jacobs v. United States, 290 U.S. 13 315, 316 Jacobson v. Merrill Lynch, Pierce, Fenner & Smith, Inc., 797 F. 2d 1197 225, 249 Page James v. Dravo Contracting Co., 302 U.S. 134 191 James v. United States, 366 U.S. 213 429 Jay v. Boyd, 351 U.S. 345 595 J. I. Case Co. v. NLRB, 321 U.S. 332 395, 396 Johnson v. Avery, 393 U.S. 483 84 Johnson v. Heckler, 769 F. 2d 1202 145, 156, 157, 165 Johnson v. Railway Express Agency, Inc., 421 U.S. 454 660, 674, 680 Johnson v. Rockefeller, 365 F. Supp. 377 96 John Wiley & Sons, Inc. v. Livingston, 376 U.S. 543 40 Joint Anti-Fascist Refugee Committee v. McGrath, 341 U.S. 123 619 Jones v. Alfred H. Mayer Co., 392 U.S. 409 674 Jones v. Helms, 452 U.S. 412 418 Jones v. North Carolina Prisoners’ Union, 433 U.S. 119 86-90, 92, 348, 349, 360, 367 Kaiser Aetna v. United States, 444 U.S. 164 317 Katz v. United States, 389 U.S. 347 699, 700 Keith v. Louisiana Dept, of Ed., 553 F. Supp. 295 581 Kelly v. Robinson, 479 U.S. 36 69, 687 Kentucky v. Dennison, 24 How. 66 406, 407, 418 Kerr v. United States District Court, 426 U.S. 394 527, 554 Keystone Bituminous Coal Assn. v. DeBenedictis, 480 U.S. 470 325, 326, 330, 331 Kimball Laundry Co. v. United States, 338 U.S. 1 318, 332 King v. Drexel Burnham Lambert, Inc., 796 F. 2d 59 225,249 King, The v. Cook, 11 Can. Crim. Cas. Ann. 32 463 Kirby v. United States, 174 U.S. 47 737, 751 Kirby Forest Industries, Inc. v. United States, 467 U.S. 1 316, 317, 320 XXX TABLE OF CASES CITED Page Knott v. State, 648 S. W. 2d 20 461 Knudsen v. Omanson, 10 Utah 124 205 Kolender v. Lawson, 461 U.S. 352 459, 465, 576 Korematsu v. United States, 323 U.S. 214 357 Kuszynski v. Oakland, 479 F. 2d 1130 572 Laker Airways, Ltd. v. Pan American World Airways, 103 F. R. D. 42 542 Laker Airways, Ltd. v. Sabena, Belgian World Airlines, 235 U.S. App. D. C. 207 552, 553, 567 Landreth Timber Co. v. Landreth, 471 U.S. 681 268 Larkin v. Grendel’s Den, Inc., 459 U.S. 116 596, 613 Lauritzen v. Larsen, 345 U.S. 571 554 Lee v. Illinois, 476 U.S. 530 739, 751 Lee v. Washington, 390 U.S. 333 84 Lehman Brothers v. Schein, 416 U.S. 386 475 Leis v. Flynt, 439 U.S. 438 651 Lemon v. Kurtzman, 403 U.S. 602 583, 585, 596, 597, 612-614, 616-618, 626, 627, 635, 636, 639, 640 Levitt v. Committee for Public Ed. & Religious Liberty, 413 U.S. 472 613, 614, 627 Levy v. Edwards, 1 Car. & P. 40 463 Lewis v. City of New Orleans, 415 U.S. 130 454, 461, 462, 466, 472, 478, 480, 482, 576 Lewis v. Commissioner of Cor- rection of City of New York, 100 Mise. 2d 48 410 Lindsey v. Washington, 301 U.S. 397 432, 433 Little, In re, 404 U.S. 553 753 Litzkuhn v. Clark, 85 Ariz. 355 743 Page Lodowski v. State, 302 Md. 691 498, 501, 504, 506, 507 Longshoremen v. Davis, 476 U.S. 380 475 Lopez v. Crist, 176 Mont. 352 377 Loretto v. Teleprompter Manhattan CATV Corp., 458 U. S. 419 314, 329 Loving v. Virginia, 388 U.S. 1 95 Lukhard v. Reed, 481 U.S. 368 178 Lynch v. Donnelly, 465 U.S. 668 585, 599, 606, 614, 616, 618 MacDonald, Sommer & Frates v. Yolo County, 477 U.S. 340 .310, 311, 314, 333, 335 Machinists v. Wisconsin Employment Relations Comm’n, 427 U.S. 132 20, 397 Maher v. Gagne, 448 U.S. 122 760, 761, 767 Malloy v. South Carolina, 237 U.S. 180 429 Malnak v. Yogi, 440 F. Supp. 1284 599 Malone v. White Motor Corp., 435 U.S. 497 8, 22 Mandel v. Bradley, 432 U.S. 173 96 Manhattan Life Ins. Co. v. Cohen, 234 U.S. 123 314 Mann v. Tacoma Land Co., 153 U.S. 273 203 Mansbach v. Prescott, Ball & Turben, 598 F. 2d 1017 248 Marin County v. United States, 356 U.S. 412 300 Marsh v. Chambers, 463 U.S. . 783 583, 597 Marshall v. Barlow’s, Inc., 436 U.S. 307 700, 701, 703, 711, 719, 722, 723 Marshall v. Dewey, 493 F. Supp. 963 719 Martino v. Santa Clara Valley Water Dist., 703 F. 2d 1141 312 Martori Bros. Distributors v. James-Massengale, 781 F. 2d 1349 12 Maryland v. Louisiana, 451 U.S. 725 130, 132 TABLE OF CASES CITED XXXI Page Mathews v. Eldridge, 424 U.S. 319 146 Mattox v. United States, 156 U.S. 237 737, 750 Mayaja, Inc. v. Bodkin, 803 F. 2d 157 225, 239 Mayes v. Trammell, 751 F. 2d 175 379 Mayor of Philadelphia v. Education Equality League, 415 U.S. 605 478 McCleskey v. Kemp, 481 U.S. 279 517 McCollum v. Board of Ed., 333 U.S. 203 584 McCruter v. Bowen, 791 F. 2d 1544 156 McCulloch v. Maryland, 4 Wheat. 316 190 McDonald v. Board of Election Comm’rs of Chicago, 394 U.S. 802 618 McDonald v. Santa Fe Trail Transportation Co., 427 U.S. 273 674, 678 McDonald v. Secretary of Health and Human Services, 795 F. 2d 1118 145, 156 McGoldrick v. Compagnie Generale Transatlantique, 309 U.S. 430 324, 747 McGowan v. Maryland, 366 U.S. 420 605, 615 McLean v. Arkansas Bd. of Ed., 529 F. Supp. 1255 590, 600-603 McLean Trucking Co. v. United States, 321 U.S. 67 300 McLinn, In re, 739 F. 2d 1395 74 McNichols v. Pease, 207 U.S. 100 417, 419 Meachum v. Fano, 427 U.S. 215 383, 384 Medlin v. Boeing Vertol Co. 620 F. 2d 957 399 Medo Photo Supply Corp, v NLRB, 321 U.S. 678 398 Meek v. Pittenger, 421 U.S. 349 584, 613, 616, 627 Memphis Bank & Trust v. Garner, 459 U.S. 392 187, 188 Page Memphis Light, Gas & Water Division v. Craft, 436 U.S. 1 371 Merrell Dow Pharmaceuticals, Inc. v. Thompson, 478 U.S. 804 392 Merrill Lynch, Pierce, Fenner & Smith, Inc. v. Moore, 590 F. 2d 823 248 Merrill Lynch, Pierce, Fenner & Smith, Inc. v. Ware, 414 U.S. 117 468- 488, 490, 491, 493, 495 Messerschmitt Bolkow Blohm GmbH, In re, 757 F. 2d 729 536, 541 Metropolitan Life Ins. Co. v. Massachusetts, 471 U.S. 724 8, 20, 22, 23 Metropolitan Life Ins. Co. v. Taylor, 481 U.S. 58 392, 393 Meyers v. Pennypack Woods Home Ownership Assn., 559 F. 2d 894 662, 679 Michigan v. Clifford, 464 U.S. 287 724, 726 Michigan v. Doran, 439 U.S. 282 408, 416-418, 421 Michigan v. Long, 463 U.S. 1032 735 Michigan v. Tyler, 436 U.S. 499 724 Microwave Communications, Inc. v. FCC, 169 U.S. App. D. C. 154 280 Miner v. Atlass, 363 U.S. 641 654 Mississippi Power Co. v. Luter, 336 So. 2d 753 920 Missouri Pacific R. Co. v. United Transportation Union, 782 F. 2d 107 300, 301 Mitsubishi Motors Corp. v. Soler Chrysler-Plymouth, Inc., 473 U.S. 614 224, 226, 230, 232, 233, 239-241, 250, 251, 255, 260, 267, 490, 554 Mohasco Corp. v. Silver, 447 U.S. 807 178 Moll v. State, 351 N. W. 2d 639 746 XXXII TABLE OF CASES CITED Page Monell v. New York City Dept, of Social Services, 436 U.S. 658 341 Monongahela Navigation Co. v. United States, 148 U.S. 313 315, 316, 319 Monroe v. Pape, 365 U.S. 167 676 Montana v. Hall, 481 U.S. 400 121 Montana v. United States, 450 U.S. 544 198, 202, 207, 209, 219 Montanye v. Haymes, 427 U.S. 236 383, 384 Moore v. Commonwealth, 384 S. W. 2d 498 741 Moore v. State, 442 So. 2d 215 699 Morales v. Schmidt, 340 F. Supp. 544 355 Moran v. Paine, Webber, Jack-son & Curtis, 389 F. 2d 242 233, 237, 244, 253 Morton v. Mancari, 417 U.S. 535 445 Moses H. Cone Memorial Hospital v. Mercury Construction Corp., 460 U.S. 1 226, 232, 489, 492 Mueller v. Allen, 463 U.S. 388 613, 616, 618 Mugler v. Kansas, 123 U.S. 623 313, 325, 326 Munsey v. Clough, 196 U.S. 364 413, 419 NAACP v. Button, 371 U.S. 415 467 NAACP v. Claiborne Hardware Co., 458 U.S. 886 458 NAACP v. FPC, 425 U.S. 662 289 National Bank of Davis v. Comptroller of Currency, 233 U.S. App. D. C. 284 280 NLRB v. Auto Ventshade, Inc., 276 F. 2d 303 37 NLRB v. Band-Age, Inc., 534 F. 2d 1 45, 46, 57 NLRB v. Burns International Security Services, Inc., 406 U.S. 272 29, 30, 35-41, 43, 46, 47, 54, 56, 60 NLRB v. Daneker Clock Co., 516 F. 2d 315 45 Page NLRB v. Financial Institution Employees, 475 U.S. 192 41, 42, 50 NLRB v. Jeffries Lithograph Co., 752 F. 2d 459 43, 47 NLRB v. Jones & Laughlin Steel Corp., 301 U.S. 1 626 NLRB v. Middleboro Fire Apparatus, Inc., 590 F. 2d 4 46 NLRB v. Nash-Finch Co., 404 U.S. 138 20 NLRB v. Pre-Engineered Building Products, Inc., 603 F. 2d 134 48 New Jersey v. New York, 283 U.S. 805 134 New Jersey v. T. L. O., 469 U.S. 325 702 Newton v. Consolidated Gas Co., 265 U.S. 78 447 New York v. Ferber, 458 U.S. 747 458, 574 New York City v. Heckler, 755 F. 2d 31 150 New York Guardian Mortgage Corp. v. Cleland, 473 F. Supp. 409 189 Northeast Broadcasting, Inc. v. FCC, 130 U.S. App. D. C. 278 293 Nyflot v. Minnesota Comm’r of Public Safety, 474 U.S. 1027 910 O’Connor v. Ortega, 480 U.S. 709 702 O’Connor v. United States, 479 U.S. 27 536 Ohio v. Roberts, 448 U.S. 56 737, 749, 750 Olim v. Wakinekona, 461 U.S. 238 383 Oneida Indian Nation v. County of Oneida, 414 U.S. 661 393 Outland v. CAB, 109 U.S. App. D. C. 90 285 Pace Resources, Inc. v. Shrewsbury Township, 808 F. 2d 1023 325 Pacileo v. Walker, 449 U.S. 86 417 TABLE OF CASES CITED XXXIII Page Page v. Moseley, Hallgarten, Estabrook & Weeden, Inc., 806 F. 2d 291 225, 239, 249 Palestine v. United States, 559 F. 2d 408 298 Palmer v. Thompson, 403 U.S. 217 638 Papachristou v. City of Jacksonville, 405 U.S. 156 465 Parker v. Corrothers, 750 F. 2d 653 379 Patten v. North Dakota Parole Bd., 783 F. 2d 140 379 Payne v. Commonwealth, 623 S. W. 2d 867 742, 743 Pell v. Procunier, 417 U.S. 817 86-90, 92, 95, 348 Penn Central Transportation Co. v. New York City, 438 U.S. 104 314, 319, 325, 330, 331, 340 Pennsylvania v. Ritchie, 480 U.S. 39 738, 739 Pennsylvania Coal Co. v. Mahon, 260 U.S. 393 316, 322, 328-330, 339 Pennzoil Co. v. Texaco, Inc., 481 U.S. 1 477 People v. Barnes, 146 Mich. App. 37 699 People v. Class, 67 N. Y. 2d 431 728 People v. Cusumano, 108 App. Div. 2d 752 707 People v. De Bour, 40 N. Y. 2d 210 717 People v. Easley, 90 Cal. App. 3d 440 699 People v. Krull, 107 Ill. 2d 107 699 People v. Levitt, 156 Cal. App. 3d 500 505 People v. Pace, 65 N. Y. 2d 684 717 People v. Pace, 101 App. Div. 2d 336 696, 697 People v. P. J. Video, Inc., 68 N. Y. 2d 296 728 People v. Tinneny, 99 Mise. 2d 962 707 Page People ex rel. Lewis v. Commissioner of Correction of City of New York, 100 Mise. 2d 48 410 Perrin v. United States, 444 U.S. 37 598 Perry Ed. Assn. v. Perry Local Educators’ Assn., 460 U.S. 37 572-574 Personnel Administrator of Mass. v. Feeney, 442 U.S. 256 684, 685 Peterson, Ex parte, 253 U.S. 300 447, 448 Petty v. Tennessee-Missouri Bridge Comm’n, 359 U.S. 275 128 Philadelphia Gear Corp. v. American Pfauter Corp., 100 F. R. D. 58 542, 549 Phillips v. Merrill Lynch, Pierce, Fenner & Smith, Inc., 795 F. 2d 1393 225, 249 Pierburg GmbH & Co. KG v. Superior Court, 137 Cal. App. 3d 238 549 Pierce v. Creecy, 210 U.S. 387 408, 411, 413, 414, 417 Pierce v. LaVallee, 293 F. 2d 233 363 Pierce v. United States, 255 U.S. 398 133 Pierson v. Ray, 386 U.S. 547 609 Plummer v. Coler, 178 U.S. 115 190, 191 Pointer v. Texas, 380 U.S. 400 736, 749 Pollard v. State, 687 S. W. 2d 373 474 Pollard’s Lessee v. Hagan, 3 How. 212 196 Portley v. Grossman, 444 U.S. 1311 434 Poynter v. Chipman, 8 Utah 442 205 Premium Foods, Inc. v. NLRB, 709 F. 2d 623 49, 61 Price v. Johnston, 334 U.S. 266 348 Prigg v. Pennsylvania, 16 Pet. 539 407 XXXIV TABLE OF CASES CITED Page Prima Paint Corp. v. Flood & Conklin Mfg. Co., 388 U.S. 395 491 Procunier v. Martinez, 416 U.S. 396 83-87, 89, 94, 97, 348, 350, 361, 467 Provisioners Frozen Express, Inc. v. ICC, 536 F. 2d 1303 293 Public Affairs Associates, Inc. v. Rickover, 369 U.S. Ill 762 Pugh v. See’s Candies, Inc., 116 Cal. App. 3d 311 389 Pulliam v. Allen, 466 U.S. 522 766 Pullman-Standard v. Swint, 456 U.S. 273 665 Pumpelly v. Green Bay Co., 13 Wall. 166 316 Radio Corp, of America v. United States, 341 U.S. 412 292 Radzanower v. Touche Ross & Co., 426 U.S. 148 445 Raiford v. Buslease Inc., 745 F. 2d 1419 248 Railroad Comm’n v. Pullman Co., 312 U.S. 496 74, 467, 473, 475-478 Red Lion Broadcasting Co. v. FCC, 395 U.S. 367 178 Reggel, Ex parte, 114 U.S. 642 417-419 Reid v. State, 305 Md. 9 498, 506 Rhode Island v. Massachusetts, 12 Pet. 657 128 Richards v. United States, 369 U.S. 1 594 Roberts v. Reilly, 116 U.S. 80 410, 411, 413, 419 Robinson v. California, 370U.S. 660 501 Roemer v. Maryland Public Works Bd., 426 U.S. 736 616 Romero v. International Terminal Operating Co., 358 U.S. 354 554 Rose v. Lundy, 455 U.S. 509 338 Rosen v. Port of Portland, 641 F. 2d 1243 572 Rotsker v. Goldberg, 453 U.S. 57 618, 619 R. R., In re, 79 N. J. 97 743 Page Ruffalo, In re, 390 U.S. 544 645, 654 Rules of Criminal Procedure (Sentencing Guidelines), In re, 439 So. 2d 848 425 Runyon v. McCrary, 427 U.S. 160 660, 671, 673, 674, 676 Ruthenbeck v. First Criminal Judicial Court of Bergen Cty., 7 N. J. Mise. 969 464 St. Joseph Stock Yards Co. v. United States, 298 U.S. 38 292 Saks & Co. v. NLRB, 634 F. 2d 681 47, 60 Salmi v. Secretary of Health and Human Services, 774 F. 2d 685 156 San Diego Building Trades Council v. Garmon, 359 U.S. 236 22, 398, 904 San Diego Gas & Electric Co. v. San Diego, 450 U.S. 621 309-311, 315, 316, 318, 335, 340 Scherk v. Alberto-Culver Co., 417 U.S. 506 225, 229, 232, 237, 238, 244, 247, 248, 254-256, 268, 554 Schooner Exchange, The v. McFaddon, 7 Cranch 116 554, 557 Schuemann v. Colorado State Bd. of Adult Parole, 624 F. 2d 172 379 Schwabacher v. United States, 334 U.S. 182 299 Schweiker v. Gray Panthers, 453 U.S. 34 145, 147, 156 Scopes v. State, 154 Tenn. 105 590, 634 Scotia, The, 14 Wall. 170 554 Scott v. Gulf Oil Corp., 754 F. 2d 1499 7 Scott v. Illinois Parole and Pardon Bd., 669 F. 2d 1185 379 Seaboard Air Line R. Co. v. Daniel, 333 U.S. 118 299 Seaboard Air Line R. Co. v. United States, 261 U.S. 299 316 Secombe, Ex parte, 19 How. 9 655 TABLE OF CASES CITED xxxv Page SEC v. Chenery Corp., 332 U.S. 194 282, 283, 290, 291 SEC v. Louisiana Public Service Comm’n, 353 U.S. 368 280, 292 Sedima, S. P. R. L. v. Imrex Co., 473 U.S. 479 238, 239, 241 See v. City of Seattle, 387 U.S. 541 699, 718, 724 Selling v. Radford, 243 U.S. 46 648, 654 Shaw v. Delta Airlines, Inc., 463 U.S. 85 8, 10, 24 Shearson/American Express Inc. v. McMahon, 482 U.S. 220 490, 493-495 Sherbert v. Verner, 374 U.S. 398 617 Shirley v. Chestnut, 603 F. 2d 805 379 Shirley v. Commonwealth, 218 Va. 49 699 Shively v. Bowlby, 152 U.S. 1 196, 197, 200, 201, 203, 204, 209 Shuttlesworth v. Birmingham, 394 U.S. 147 469 Sibley v. Tandy Corp., 543 F. 2d 540 248 Simmons v. United States, 390 U.S. 377 753 Skipper v. South Carolina, 476 U.S. 1 506 Smith v. Davis, 323 U.S. UI 188-191 Smith v. Goguen, 415 U.S. 566 464, 465, 468, 481 Smith v. Pittsburgh, 764 F. 2d 188 663 Snyder, In re, 472 U.S. 634 646, 655 Snyder v. Massachusetts, 291 U.S. 97 745, 754 Société Int’le Pour Participations Industrielles et Commerciales, S. A. v. Rogers, 357 U.S. 197 544, 553 Société Nationale Industrielle Aérospatiale v. United States District Court, 788 F. 2d 1408 536 Page Sostre v. McGinnis, 334 F. 2d 906 363 South Carolina v. Bailey, 289 U.S. 412 419 South Carolina v. Moore, 447 F. 2d 1067 920 South Dakota v. North Carolina, 192 U.S. 286 130 Southland Corp. v. Keating, 465 U.S. 1 232, 489, 490, 493-495 Spence v. Washington, 418 U.S. 405 463 Sperry v. Florida ex rel. Fla. Bar, 373 U.S. 379 648 Sprague v. Ticonic National Bank, 307 U.S. 161 447 S. S. Silberblatt, Inc. v. Tax Comm’n of New York, 5 N. Y. 2d 635 190 Standard Oil Co. of Cal. v. Agsalud, 633 F. 2d 760 12, 13, 17, 26 State. See also name of State. State v. Elliott, 171 La. 306 405 State v. Galio, 92 N. M. 266 699 State v. Howard, 57 Ohio App. 2d 1 746 State v. Jackson, 478 So. 2d 1054 428 State v. Mischler, 488 So. 2d 523 426, 435 State v. Tindell, 272 Ind. 479 699 State v. Zinmeister, 27 Ohio App. 3d 313 699 State ex rel. Herman v. Powell, 139 Mont. 583 377 Staton v. Wainwright, 665 F. 2d 686 379 Staub v. City of Baxley, 355 U.S. 313 324 Steffel v. Thompson, 415 U.S. 452 459 Sterne v. Dean Witter Reynolds, Inc., 808 F. 2d 480 225, 249 Stone v. Graham, 449 U.S. 39 584, 585, 587, 589, 590, 593, 594, 604, 608, 613, 614 Stone v. Heckler, 752 F. 2d 1099 156, 179 XXXVI TABLE OF CASES CITED Page Story v. Livingston, 13 Pet. 359 655 Stovall v. Denno, 388 U.S. 293 690 Strassheim v. Daily, 221 U.S. 280 419 Strauss, In re, 197 U.S. 324 408, 414 Sumitomo Shoji America, Inc. v. Avagliano, 457 U.S. 176 536 Supreme Court of N. H. v. Piper, 470 U.S. 274 647, 649-651 Surman v. Merrill Lynch, Pierce, Fenner & Smith, 733 F. 2d 59 248 Swift Industries, Inc. v. Botany Industries, Inc., 466 F. 2d 1125 259 Taggart v. Weinacker’s, Inc., 397 U.S. 223 21 Tashea v. Bache, Halsey, Stuart, Shields, Inc., 802 F. 2d 1337 225 Teamsters v. Morton, 377 U.S. 252 397 Teamsters v. United States, 431 U.S. 324 665, 684 Tennessee v. Street, 471 U.S. 409 737 Terminiello v. Chicago, 337 U.S. 1 461, 470, 472 Terraciano v. Montanye, 493 F. 2d 682 727 Texas Liquor Control Bd. v. Canyon Creek Land Corp., 456 S. W. 2d 891 477 Textile Workers v. Darlington Mfg. Co., 380 U.S. 263 59 Theard v. United States, 354 U.S. 278 646, 648, 654 Th. Goldschmidt A. G. v. Smith, 676 S. W. 2d 443 549 Thomas v. Commonwealth, 300 Ky. 480 741 Thomas v. Review Bd., Ind. Employment Security Div., 450 U.S. 707 617 Thomas v. Sellers, 691 F. 2d 487 379 Page Thornhill v. Alabama, 310 U.S. 88 465 Thorpe v. Housing Authority of Durham, 393 U.S. 268 662 Tigner v. Gardner, 356 F. 2d 647 171 Tillman v. Wheaton-Haven Recreation Assn., Inc., 410 U.S. 431 676 Tilton v. Richardson, 403 U.S. 672 613, 627 Tinker v. Des Moines Independent Community School Dist., 393 U.S. 503 573, 583 Tison v. Arizona, 481 U.S. 137 505, 519 Tom-A-Hawk Transit, Inc. v. NLRB, 419 F. 2d 1025 37 Torasco v. Watkins, 367 U.S. 488 624, 635 Train v. City of New York, 420 U.S. 35 237 Train v. Colorado Public Interest Research Group, 426 U.S. 1 445 Trans World Airlines, Inc. v. Franklin Mint Corp., 466 U.S. 243 533, 537 Trans World Airlines, Inc. v. Hardison, 432 U.S. 63 688 Tullis v. Kohlmeyer & Co., 551 F. 2d 632 235 Turner v. Safley, 482 U.S. 78 348-354, 359-361, 363, 367, 368 Udall v. Tailman, 380 U.S. 1 212 United Housing Foundation, Inc. v. Forman, 421 U.S. 837 179 United Services Life Ins. Co. v. Delaney, 396 S. W. 2d 855 477 United States v. American Railway Express Co., 265 U.S. 425 324 United States v. Biswell, 406 U.S. 311 700-703, 710- 713, 716, 717, 719, 721 United States v. Causby, 328 U.S. 256 316, 317, 319, 329 United States v. Ceccolini, 435 U.S. 268 665 TABLE OF CASES CITED XXXVII Page United States v. Cherokee Nation of Okla., 480 U.S. 700 325 United States v. Clarke, 445 U.S. 253 315 United States v. Dickinson, 331 U.S. 745 316, 317 United States v. Dow, 357 U.S. 17 317, 318, 331 United States v. Feola, 420 U.S. 671 516 United States v. Gagnon, 470 U.S. 522 747, 754 United States v. General Motors Corp., 323 U.S. 373 318, 319, 332 United States v. Hasting, 461 U.S. 499 651 United States v. Holt State Bank, 270 U.S. 49 197, 201, 202, 207, 208 United States v. Hvass, 355 U.S. 570 652 United States v. Jones, 109 U.S. 513 315 United States v. King, 119 F. Supp. 398 67 United States v. Michigan, 190 U.S. 379 130 United States v. Mississippi, 380 U.S. 128 130 United States v. Mottaz, 476 U.S. 834 70 United States v. Munsingwear, Inc., 340 U.S. 36 901 United States v. New Mexico, 455 U.S. 720 191 United States v. Northern Pacific R. Co., 288 U.S. 490 292 United States v. O’Brien, 391 U.S. 367 637, 638 United States v. Paradise, 480 U.S. 149 115 United States v. Petty Motor Co., 327 U.S. 372 318,332 United States v. Pierce Auto Freight Lines, Inc., 327 U.S. 515 278, 288, 289 United States v. Reese, 92 U.S. 214 466 United States v. Schooner Peggy, 1 Cranch 103 662 Page United States v. Seatrain Lines, Inc., 329 U.S. 424 278 United States v. Thirty-seven Photographs, 402 U.S. 363 474 United States v. United Continental Tuna Corp., 425 U.S. 164 445 United States v. Villamonte- Marquez, 462 U.S. 579 716 United States ex rel. Scott v. Illinois Parole and Pardon Bd., 669 F. 2d 1185 379 United States ex rel. Terraciano v. Montanye, 493 F. 2d 682 727 Universal Camera Corp. v. NLRB, 340 U.S. 474 42 Uranium Antitrust Litigation, In re, 480 F. Supp. 1138 553 Vaca v. Sipes, 386 U.S. 171 689 Valencia v. Heckler, 751 F. 2d 1082 144 Vermont v. New York, 417 U.S. 270 134 Vermont Yankee Nuclear Power Corp. v. Natural Resources Defense Council, Inc., 435 U.S. 519 291 Vincent v. Ateliers de la Motobecane, S. A., 193 N. J. Super. 716 549 Virginia v. West Virginia, 246 U.S. 565 130, 131, 133, 134 Virginia Appalachian Lumber Corp. v. ICC, 197 U.S. App. D. C. 13 293 Volkswagenwerk Aktiengesellschaft v. Superior Court, 123 Cal. Rptr. 876 549 Wagner v. Gilligan, 609 F. 2d 866 379 Wallace v. Christensen, 802 F. 2d 1539 434 Wallace v. Jaffree, 472 U.S. 38 583-585, 587, 588, 594, 596, 599, 603, 613, 614, 619, 636 Walz v. Tax Comm’n of New York City, 397 U.S. 664 615- 617 Ware v. Hylton, 3 Dall. 199 533 XXXVIII TABLE OF CASES CITED Page Warren v. United States Parole Comm’n, 212U.S. App. D. C. 137 434 Wausau v. United States, 703 F. 2d 1042 293 Weaver v. Graham, 450 U.S. 24 428, 430, 433, 435 Weissbuch v. Merrill Lynch, Pierce, Fenner & Smith, Inc., 588 F. 2d 831 248, 256 Wesley v. Eells, 177 U.S. 370 131 Weston v. City Council of Charleston, 2 Pet. 449 190 West Virginia ex rel. Dyer v. Sims, 341 U.S. 22 128 Wheeler v. United States, 159 U.S. 523 741 Whitehead v. Stith, 268 Ky. 703 742 Widmar v. Vincent, 454 U.S. 263 584, 607, 613 Wildenhaus’s Case, 120 U.S. 1 554 Wiley & Sons, Inc. v. Livingston, 376 U.S. 543 40 Wilko v. Swan, 346 U.S. 427 224-226, 228-234, 236-238, 243-254, 256-259, 261-264, 268, 269 Willcuts v. Bunn, 282 U.S. 216 190 Williams v. Briscoe, 641 F. 2d 274 379 Williams v. Missouri Bd. of Probation and Parole, 661 F. 2d 697 379 Williams v. State, 492 So. 2d 1308, 500 So. 2d 501 435 Williamson v. Secretary of Health and Human Services, 796 F. 2d 146 180 Williamson County Regional Planning Comm’n v. Hamilton Bank, 473 U.S. 172 310, 311, 314, 320, 331, 333, 335, 338, 340 Wilson v. Garcia, 471 U.S. 261 660-663, 670, 671, 677-679, 689 Page Wilson v. Secretary of Health and Human Services, 796 F. 2d 36 145, 156 Wimmer v. Celebrezze, 355 F. 2d 289 171 Winsett v. McGinnes, 617 F. 2d 996 380 Winters v. New York, 333 U.S. 507 459 Wisconsin v. Constantineau, 400 U.S. 433 467, 468 Wisconsin v. Illinois, 281 U.S. 179 134 Wisconsin v. Yoder, 406 U.S. 205 617 Witherspoon v. Illinois, 391 U.S. 510 504 Witters v. Washington Dept, of Services for Blind, 474 U.S. 481 613, 614 Wolfe v. E. F. Hutton & Co., 800 F. 2d 1032 225,247,249,256 Wolff v. McDonnell, 418 U.S. 539 84, 384 Wolman v. Walter, 433 U.S. 229 613, 616 Woods v. Nierstheimer, 328 U.S. 211 339 Woodson v. North Carolina, 428 U.S. 280 504, 509, 519 Wyoming v. Colorado, 309 U.S. 572 132 Yamamoto v. United States Parole Comm’n, 794 F. 2d 1295 434 Younger v. Harris, 401 U.S. 37 465 Youngman v. Nevada Irrigation Dist., 70 Cal. 2d 240 395 Zablocki v. Redhail, 434 U.S. 374 95 Zant v. Stephens, 462 U.S. 862 502 Zim’s Foodliner, Inc. v. NLRB, 495 F. 2d 1131 46 Zwickler v. Koota, 389 U.S. 241 468 CASES ADJUDGED IN THE SUPREME COURT OF THE UNITED STATES AT OCTOBER TERM, 1986 FORT HALIFAX PACKING CO., INC. v. COYNE, DIRECTOR, BUREAU OF LABOR STANDARDS OF MAINE, et al. APPEAL FROM THE SUPREME JUDICIAL COURT OF MAINE No. 86-341. Argued March 24, 1987—Decided June 1, 1987 After appellant closed its poultry packaging and processing plant and laid off most of the employees who worked there, the Director of Maine’s Bureau of Labor Standards filed suit to enforce the provisions of a state statute requiring employers, in the event of a plant closing, to provide a one-time severance payment to employees not covered by an express contract providing for severance pay. The State Superior Court granted the Director summary judgment, holding appellant liable under the statute, and the State Supreme Court affirmed, rejecting appellant’s contentions that the state statute was pre-empted by the Employee Retirement Income Security Act of 1974 (ERISA) and by the National Labor Relations Act (NLRA). Held: 1. The Maine severance pay statute is not pre-empted by ERISA, since it does not “relate to any employee benefit plan” under that statute’s pre-emption provision, 29 U. S. C. § 1144(a). Appellant’s contention that any state law pertaining to a type of employee benefit listed in ERISA, such as severance pay, necessarily regulates an employee benefit plan, and is therefore pre-empted, fails in light of the plain meaning and underlying purpose of § 1144(a) and the overall objectives of ERISA itself. Pp. 7-19. (a) Section 1144(a) does not refer to state laws relating simply to “employee benefits,” but expressly states that state laws are superseded 1 2 OCTOBER TERM, 1986 Syllabus 482 U. S. insofar as they “relate to any employee benefit plan” (emphasis added). In fact, ERISA uses the words “benefit” and “plan” separately throughout the statute, and nowhere treats them as equivalent. Given the basic difference between the two concepts, Congress’ choice of language is significant in its pre-emption of only the latter, which cannot be read out of ERISA. In order to be pre-empted, a state statute must have some connection with, or reference to, a plan. Pp. 7-8. (b) Pre-emption of the Maine statute would not further the purpose of ERISA pre-emption, which is to allow plans to adopt a uniform scheme for coordinating complex administrative activities, unaffected by conflicting regulatory requirements in differing States. The Maine statute neither establishes, nor requires an employer to maintain, a plan that would embody a set of administrative practices vulnerable to the burden imposed by a patchwork, multistate regulatory scheme. In fact, the theoretical possibility of a one-time, lump-sum severance payment triggered by a single event requires no administrative scheme whatsoever to meet the employer’s statutory obligation. Pp. 8-15. (c) Similarly, the Maine statute does not implicate the regulatory concerns of ERISA itself, which was enacted to ensure administrative integrity in the operation of plans by preventing potential fiduciary abuse. The Maine statute neither establishes a plan nor generates any administrative activity capable of being abused. Pp. 15-16. (d) Appellant’s contention that failure to pre-empt the Maine statute will allow employers to circumvent ERISA, by persuading States to require types of plans the employers would otherwise have established on their own, has no force with respect to a state statute that, as here, does not establish a plan, generates no ERISA-covered program activity, presents no risk that otherwise applicable federal requirements will be evaded by an employer or dislodged by a State, and creates no prospect that an employer will face difficulty in operating a unified administrative benefit payment scheme. Holland n. Burlington Industries, Inc., 772 F. 2d 1140, summarily aff’d, 477 U. S. 901, and Gilbert v. Burlington Industries, Inc., 765 F. 2d 320, summarily aff’d, 477 U. S. 901, distinguished. Pp. 16-19. (e) Where, as here, a state statute creates no danger of conflict with a federal statute, there is no reason to disable it from, attempting to address uniquely local social and economic problems. P. 19. 2. The Maine severance pay statute is not pre-empted by the NLRA. Appellant’s argument that the statute’s establishment of a minimum labor standard impermissibly intrudes upon the collective-bargaining process was rejected in Metropolitan Life Ins. Co. v. Massachusetts, 471 U. S. 724, and is without merit here. Although the statute does FORT HALIFAX PACKING CO. v. COYNE 3 1 Opinion of the Court give employees something for which they might otherwise have had to bargain, that is true of any state law that substantively regulates employment conditions. Moreover, appellant’s argument that this case is distinguishable from Metropolitan Life because the statutory obligation at issue here is optional, in that it applies only in the absence of an agreement between employer and employees, is not persuasive, since, in fact, the parties’ freedom to devise their own severance pay arrangements strengthens the case that the statute works no intrusion on collective bargaining. Thus, the statute is a valid and unexceptional exercise of the State’s police power, and is compatible with the NLRA. Pp. 19-22. 510 A. 2d 1054, affirmed. Brennan, J., delivered the opinion of the Court, in which Marshall, Blackmun, Powell, and Stevens, JJ., joined. White, J., filed a dissenting opinion, in which Rehnquist, C. J., and O’Connor and Scalia, JJ., joined, post, p. 23. John C. Yavis, Jr., argued the cause for appellant. With him on the briefs were Thomas M. Cloherty and Barry J. Waters. Thomas D. Warren, Assistant Attorney General of Maine, argued the cause for appellees. With him on the brief for appellee Coyne was James E. Tierney, Attorney General.* Justice Brennan delivered the opinion of the Court. In this case we must decide whether a Maine statute requiring employers to provide a one-time severance payment to employees in the event of a plant closing, Me. Rev. Stat. *Briefs of amici curiae urging reversal were filed for the United States by Solicitor General Fried, Deputy Solicitor General Ayer, Christopher J. Wright, George R. Salem, and Allen H. Feldman; and for the Chamber of Commerce of the United States by Richard G. Moon, Linda D. McGill, John H. Rich III, and Stephen A. Bokat. Briefs of amici curiae urging affirmance were filed for the American Federation of Labor and Congress of Industrial Organizations by Marsha S. Berzon and Laurence Gold; and for the Employment Law Center et al. by Joan M. Graff, Robert Barnes, John M. True, Patricia A. Shiu, and James Eggleston. 4 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. Ann., Tit. 26, §625-B (Supp. 1986-1987),1 is pre-empted by either the Employee Retirement Income Security Act of 1974, 88 Stat. 832, as amended, 29 U. S. C. §§ 1001-1381 (ERISA), or the National Labor Relations Act, 49 Stat. 452, as amended, 29 U. S. C. §§157-158 (NLRA). The statute was upheld by the Maine Superior Court, Civ. Action No. CV81-516 (Oct. 29,1982), and by the Maine Supreme Judicial Court, 510 A. 2d 1054 (1986). We noted probable jurisdiction, 479 U. S. 947 (1986), and now affirm. I In 1972, Fort Halifax Packing Company (Fort Halifax or Company) purchased a poultry packaging and processing plant that had operated in Winslow, Maine, for almost two decades. The Company continued to operate the plant for almost another decade, until, on May 23, 1981, it discontinued operations at the plant and laid off all its employees except several maintenance and clerical workers. At the time 1 The statute provides in pertinent part: “2. Severance pay. Any employer who relocates or terminates a covered establishment shall be liable to his employees for severance pay at the rate of one week’s pay for each year of employment by the employee in that establishment. The severance pay to eligible employees shall be in addition to any final wage payment to the employee and shall be paid within one regular pay period after the employee’s last full day of work, notwithstanding any other provisions of law. “3. Mitigation of severance pay liability. There shall be no liability for severance pay to an eligible employee if: “A. Relocation or termination of a covered establishment is necessitated by a physical calamity; “B. The employee is covered by an express contract providing for severance pay; “C. That employee accepts employment at the new location; or “D. That employee has been employed by the employer for less than 3 years.” Section 625-B(l)(A) defines “covered establishment” as a facility that employs 100 or more persons, while § 625-B(l)(F) defines “relocation” as the removal of all or substantially all operations at least 100 miles away from their original location. FORT HALIFAX PACKING CO. v. COYNE 5 1 Opinion of the Court of closing, over 100 employees were on the payroll. Forty-five had worked in the plant for over 10 years, 19 for over 20 years, and 2 for 29 years. Plaintiff’s Supplementary Response to Employee List, Exhibit A (June 3, 1983). Following the closing, the Company met with state officials and with representatives of Local 385 of the Amalgamated Meat Cutters & Butcher Workmen of North America, which represented many of the employees who had worked in the plant. While Fort Halifax initially suggested that reopening the plant might be feasible if the union agreed to certain concessions in the form of amendments to the collective-bargaining agreement, ultimately the Company decided against resuming operations and to close the plant. On October 30, 1981, 11 employees filed suit in Superior Court seeking severance pay pursuant to Me. Rev. Stat. Ann., Tit. 26, §625-B (Supp. 1986-1987). This statute, which is set forth in n. 1, supra, provides that any employer that terminates operations at a plant with 100 or more employees, or relocates those operations more than 100 miles away, must provide one week’s pay for each year of employment to all employees who have worked in the plant at least three years. The employer has no such liability if the employee accepts employment at the new location, or if the employee is covered by a contract that deals with the issue of severance pay. §§ 625-B(2), (3). Under authority granted by the statute, the Maine Director of the Bureau of Labor Standards also commenced an action to enforce the provisions of the state law, which action superseded the suit filed by the employees.2 2 Section 625-B(5) of the Maine statute provides in relevant part: “5. Suits by the director. The director is authorized to supervise the payment of the unpaid severance pay owing to any employee under this section. The director may bring an action in any court of competent jurisdiction to recover the amount of any unpaid severance pay. The right provided by subsection 4 to bring an action by or on behalf of any employee, and of any employee to become a party plaintiff to any such action, shall terminate upon the filing of a complaint by the director in an action under 6 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. The Superior Court, ruling on cross-motions for summary judgment, granted the Director’s motion, holding that Fort Halifax is liable for severance pay under the statute. Civ. Action No. CV81-516 (Oct. 29, 1982). The Maine Supreme Judicial Court affirmed. 510 A. 2d 1054 (1986). The court rejected the Company’s contention that the plant-closing statute was pre-empted by ERISA, holding that ERISA preempted only benefit plans created by employers or employee organizations. Id., at 1059. It observed that the severance pay liability in this case results from the operation of the state statute, rather than from the operation of an employer-created benefit plan. Ibid. Therefore, reasoned the court, “[i]nasmuch as § 625-B does not implicate a plan created by an employer or employee organization, it cannot be said to be preempted by ERISA.” Ibid. The court also rejected the argument that the state provision was pre-empted by the NLRA because it regulated conduct covered by either § 7 or § 8 of that statute. It found that the Maine statute applies equally to union and nonunion employees, and reflects “the state’s substantial interest in protecting Maine citizens from the economic dislocation that accompanies large-scale plant closings.” Id., at 1062. As a result, the court found that eligible employees were entitled to severance pay due to the closure of the plant at Winslow.3 We hold that the Maine statute is not pre-empted by ERISA, not for the reason offered by the Maine Supreme Judicial Court, but because the statute neither establishes, nor requires an employer to maintain, an employee welfare benefit “plan” under that federal statute.4 We hold further that this subsection, unless the action is dismissed without prejudice by the director. ...” 3 Ninety-three employees of the plant are eligible for lump-sum payments ranging from $490 to $11,500. The total amount due is about $256,600. Affidavit of Xavier J. Dietrich, Exhibit A (Aug. 13, 1984). 4 Because we hold that the obligation created by the Maine statute does not involve a plan, we do not address the State’s alternative argument that, even if the law does establish a plan, it is not pre-empted by virtue of FORT HALIFAX PACKING CO. v. COYNE 7 1 Opinion of the Court the Maine law is not pre-empted by the NLRA, since it establishes a minimum labor standard that does not intrude upon the collective-bargaining process. As a result, we affirm the judgment of the Maine Supreme Judicial Court that the Maine statute is not pre-empted by either ERISA or the NLRA. II Appellant’s basic argument is that any state law pertaining to a type of employee benefit listed in ERISA necessarily regulates an employee benefit plan, and therefore must be pre-empted. Because severance benefits are included in ERISA, see 29 U. S. C. § 1002(l)(B), appellant argues that ERISA pre-empts the Maine statute.6 In effect, appellant argues that ERISA forecloses virtually all state legislation regarding employee benefits. This contention fails, however, in light of the plain language of ERISA’s pre-emption provision, the underlying purpose of that provision, and the overall objectives of ERISA itself. A The first answer to appellant’s argument is found in the express language of the statute. ERISA’s pre-emption provision does not refer to state laws relating to “employee benefits,” but to state laws relating to “employee benefit plans”'. the exemption for plans “maintained solely for the purpose of complying with applicable . . . unemployment compensation or disability insurance laws.” 29 U. S. C. § 1003(b)(3). 5 Section 1002(l)(B) defines an employee welfare benefit plan as a plan that pays, inter alia, benefits described in 29 U. S. C. § 186(c). The latter section includes, inter alia, money paid by an employer to a trust fund to pay for severance benefits. Section 1002(l)(B) has been construed to include severance benefits paid out of general assets, as well as out of a trust fund. See Holland v. Burlington Industries, Inc., 772 F. 2d 1140 (CA4 1985), summarily aff’d, 477 U. S. 901 (1986); Gilbert v. Burlington Industries, Inc., 765 F. 2d 320 (CA21985), summarily aff’d, 477 U. S. 901 (1986); Scott v. Gulf Oil Corp., 754 F. 2d 1499 (CA9 1985); 29 CFR § 2510.3-l(a)(3) (1986). See also discussion, infra, at 17-19. 8 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. “[T]he provisions of this subchapter . . . shall supersede any and all State laws insofar as they may now or hereafter relate to any employee benefit plan described in § 1003(a) of this title and not exempt under § 1003(b) of this title.” 29 U. S. C. § 1144(a) (emphasis added). We have held that the words “relate to” should be construed expansively: “[a] law ‘relates to’ an employee benefit plan, in the normal sense of the phrase, if it has a connection with or reference to such a plan.” Shaw v. Delta Airlines, Inc., 463 U. S. 85, 96-97 (1983). Nothing in our case law, however, supports appellant’s position that the word “plan” should in effect be read out of the statute. Indeed, Shaw itself speaks of a state law’s connection with or reference to a plan. Ibid. The words “benefit” and “plan” are used separately throughout ERISA, and nowhere in the statute are they treated as the equivalent of one another. Given the basic difference between a “benefit” and a “plan,” Congress’ choice of language is significant in its pre-emption of only the latter. Thus, as a first matter, the language of the ERISA presents a formidable obstacle to appellant’s argument. The reason for Congress’ decision to legislate with respect to plans rather than to benefits becomes plain upon examination of the purpose of both the pre-emption section and the regulatory scheme as a whole. B The second answer to appellant’s argument is that preemption of the Maine statute would not further the purpose of ERISA pre-emption. In analyzing whether ERISA’s preemption section is applicable to the Maine law, “as in any preemption analysis, ‘the purpose of Congress is the ultimate touchstone.’” Metropolitan Life Ins. Co. v. Massachusetts, 471 U. S. 724, 747 (1985) (quoting Malone n. White Motor Corp., 435 U. S. 497, 504 (1978)). Attention to purpose is particularly necessary in this case because the terms “employee benefit plan” and “plan” are defined only tautologically in the statute, each being described as “an employee welfare FORT HALIFAX PACKING CO. v. COYNE 9 1 Opinion of the Court benefit plan or employee pension benefit plan or a plan which is both an employee welfare benefit plan and an employee pension benefit plan.” 29 U. S. C. §1002(3). Statements by ERISA’s sponsors in the House and Senate clearly disclose the problem that the pre-emption provision was intended to address. In the House, Representative Dent stated that “with the preemption of the field [of employee benefit plans], we round out the protection afforded participants by eliminating the threat of conflicting and inconsistent State and local regulation.” 120 Cong. Rec. 29197 (1974). Similarly, Senator Williams declared: “It should be stressed that with the narrow exceptions specified in the bill, the substantive and enforcement provisions of the conference substitute are intended to preempt the field for Federal regulations, thus eliminating the threat of conflicting or inconsistent State and local regulation of employee benefit plans.” Id., at 29933. These statements reflect recognition of the administrative realities of employee benefit plans. An employer that makes a commitment systematically to pay certain benefits undertakes a host of obligations, such as determining the eligibility of claimants, calculating benefit levels, making disbursements, monitoring the availability of funds for benefit payments, and keeping appropriate records in order to comply with applicable reporting requirements. The most efficient way to meet these responsibilities is to establish a uniform administrative scheme, which provides a set of standard procedures to guide processing of claims and disbursement of benefits. Such a system is difficult to achieve, however, if a benefit plan is subject to differing regulatory requirements in differing States. A plan would be required to keep certain records in some States but not in others; to make certain benefits available in some States but not in others; to process claims in a certain way in some States but not in others; and to comply with certain fiduciary standards in some States but not in others. 10 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. We have not hesitated to enforce ERISA’s pre-emption provision where state law created the prospect that an employer’s administrative scheme would be subject to conflicting requirements. In Alessi v. Raybestos-Manhattan, Inc., 451 U. S. 504 (1981), for instance, we struck down a New Jersey statute that prohibited offsetting worker compensation payments against pension benefits. Since such a practice is permissible under federal law and the law of other States, the effect of the statute was to force the employer either to structure all its benefit payments in accordance with New Jersey law, or to adopt different payment formulae for employees inside and outside the State. The employer therefore was required to accommodate conflicting regulatory schemes in devising and operating a system for processing claims and paying benefits—precisely the burden that ERISA preemption was intended to avoid. This point was emphasized in Shaw, supra, where we said with respect to another form of State regulation: “Obligating the employer to satisfy the varied and perhaps conflicting requirements of particular state fair employment laws . . . would make administration of a nationwide plan more difficult.” 463 U. S., at 105, n. 25. Such a situation would produce considerable inefficiencies, which the employer might choose to offset by lowering benefit levels. As the Court in Shaw indicated, “ERISA’s comprehensive preemption of state law was meant to minimize this sort of interference with the administration of employee benefit plans,” ibid., so that employers would not have to “administer their plans differently in each State in which they have employees.” Id., at 105 (footnote omitted). This concern about the effect of state regulation on the administration of benefit programs is reflected in Shaw’s holding that only disability programs administered separately from other benefit plans fall within ERISA’s pre-emption exemption for plans maintained “for the purpose of complying with . . . disability insurance laws.” 29 U. S. C. § 1003(b)(3). FORT HALIFAX PACKING CO. v. COYNE 11 1 Opinion of the Court To permit the exemption to apply to disability benefits paid under a multibenefit plan was held to be inconsistent with the purpose of ERISA’s pre-emption provision: “An employer with employees in several States would find its plan subject to a different jurisdictional pattern of regulation in each State, depending on what benefits the State mandated under disability, workmen’s compensation, and unemployment compensation laws. The administrative impracticality of permitting mutually exclusive pockets of federal and state jurisdiction within a plan is apparent.” 463 U. S., at 107-108. It is thus clear that ERISA’s pre-emption provision was prompted by recognition that employers establishing and maintaining employee benefit plans are faced with the task of coordinating complex administrative activities. A patchwork scheme of regulation would introduce considerable inefficiencies in benefit program operation, which might lead those employers with existing plans to reduce benefits, and those without such plans to refrain from adopting them. Pre-emption ensures that the administrative practices of a benefit plan will be governed by only a single set of regulations. See, e. g., H. R. Rep. No. 93-533, p. 12 (1973) (“[A] fiduciary standard embodied in Federal legislation is considered desirable because it will bring a measure of uniformity in an area where decisions under the same set of facts may differ from state to state”). The purposes of ERISA’s pre-emption provision make clear that the Maine statute in no way raises the types of concerns that prompted pre-emption. Congress intended pre-emption to afford employers the advantages of a uniform set of administrative procedures governed by a single set of regulations. This concern only arises, however, with respect to benefits whose provision by nature requires an ongoing administrative program to meet the employer’s obligation. It is for this reason that Congress pre-empted state laws relating to plans, rather than simply to benefits. Only a plan embodies a set of 12 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. administrative practices vulnerable to the burden that would be imposed by a patchwork scheme of regulation. The Maine statute neither establishes, nor requires an employer to maintain, an employee benefit plan. The requirement of a one-time, lump-sum payment triggered by a single event requires no administrative scheme whatsoever to meet the employer’s obligation. The employer assumes no responsibility to pay benefits on a regular basis, and thus faces no periodic demands on its assets that create a need for financial coordination and control. Rather, the employer’s obligation is predicated on the occurrence of a single contingency that may never materialize. The employer may well never have to pay the severance benefits. To the extent that the obligation to do so arises, satisfaction of that duty involves only making a single set of payments to employees at the time the plant closes. To do little more than write a check hardly constitutes the operation of a benefit plan.6 Once this single event is over, the employer has no further responsibility. The theoretical possibility of a one-time obligation in the future simply creates no need for an ongoing administrative program for processing claims and paying benefits. This point is underscored by comparing the consequences of the Maine statute with those produced by a state statute requiring the establishment of a benefit plan. In Standard Oil Co. of California v. Agsalud, 633 F. 2d 760 (CA9 1980), summarily aff’d, 454 U. S. 801 (1981), for instance, Hawaii had required that employers provide employees with a comprehensive health care plan. The Hawaii law was struck 6 See Martori Bros. Distributors v. James-Massengale, 781 F. 2d 1349, 1358 (CA9) (“It is difficult to see how the making of one-time lump sum payments could constitute the establishment of a plan”), amended on other grounds, 791 F. 2d 799, cert, denied, 479 U. S. 949 (1986). Cf. Donovan v. Dillingham, 688 F. 2d 1367, 1373 (CA11 1982) (“A decision to extend benefits is not the establishment of a plan or program”). FORT HALIFAX PACKING CO. v. COYNE 13 1 Opinion of the Court down, for it posed two types of problems.7 First, the employer in that case already had in place a health care plan governed by ERISA, which did not comply in all respects with the Hawaii Act. If the employer sought to achieve administrative efficiencies by integrating the Hawaii plan into its existing plan, different components of its single plan would be subject to different requirements. If it established a separate plan to administer the program directed by Hawaii, it would lose the benefits of maintaining a single administrative scheme. Second, if Hawaii could demand the operation of a particular benefit plan, so could other States, which would require that the employer coordinate perhaps dozens of programs. Agsalud thus illustrates that whether a State requires an existing plan to pay certain benefits, or whether it requires the establishment of a separate plan where none existed before, the problem is the same. Faced with the difficulty or impossibility of structuring administrative practices according to a set of uniform guidelines, an employer may decide to reduce benefits or simply not to pay them at all.8 7 In 1983, Congress amended ERISA to exempt from pre-emption certain provisions of the Hawaii Act in place before the enactment of ERISA, Haw. Rev. Stat. §§393-1 through 393-48 (1976 and Supp. 1984). 29 U. S. C. § 1144(b)(5). The amendment did not exempt from pre-emption those portions of the law dealing with reporting, disclosure, and fiduciary requirements. 8 The dissent draws support for its position from the the court’s rejection in Agsalud of the argument that only state laws relating to plan administration, as opposed to plan benefits, are pre-empted by ERISA. Post, at 26. The court’s position, however, no more than acknowledges what we have said in our discussion, supra: state laws requiring the payment of benefits also “relate to a[n] employee benefit plan” if they attempt to dictate what benefits shall be paid under a plan. To hold otherwise would create the prospect that plan administration would be subject to differing requirements regarding benefit eligibility and benefit levels — precisely the type of conflict that ERISA’s pre-emption provision was intended to prevent. 14 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. By contrast, the Maine law does not put the employer to the choice of either: (1) integrating a state-mandated ongoing benefit plan with an existing plan or (2) establishing a separate plan to process and pay benefits under the plan required by the State. This is because there is no state-mandated benefit plan to administer. In this case, for instance, Fort Halifax found no need to respond to passage of the Maine statute by setting up an administrative scheme to meet its contingent statutory obligation, any more than it would find it necessary to set up an ongoing scheme to deal with the obligations it might face in the event that some day it might go bankrupt. The Company makes no contention that its statutory duty has in any way hindered its ability to operate its retirement plan in uniform fashion, a plan that pays retirement, death, and permanent and total disability benefits on an ongoing basis. App. 40. The obligation imposed by the Maine statute thus differs radically in impact from a requirement that an employer pay ongoing benefits on a continuous basis. The Maine statute therefore creates no impediment to an employer’s adoption of a uniform benefit administration scheme. Neither the possibility of a one-time payment in the future, nor the act of making such a payment, in any way creates the potential for the type of conflicting regulation of benefit plans that ERISA pre-emption was intended to prevent.9 As a result, pre-emption of the Maine law would not 9 Appellant notes that death benefits sometimes involve a one-time payment to beneficiaries, and that ERISA nonetheless defines an employee welfare benefit plan to include a program that pays such benefits. 29 U. S. C. § 1002(1). Thus, it contends, the fact that the Maine statute requires a single payment does not mean that the statute does not establish a plan. This argument, however, misunderstands what it is that makes a plan a plan. While death benefits may represent a one-time payment from the perspective of the beneficiaries, the employer clearly foresees the need to make regular payments to survivors on an ongoing basis. The ongoing, predictable nature of this obligation therefore creates the need for an administrative scheme to process claims and pay out benefits, whether those FORT HALIFAX PACKING CO. v. COYNE 15 1 Opinion of the Court serve the purpose for which ERISA’s pre-emption provision was enacted. C The third answer to appellant’s argument is that the Maine statute not only fails to implicate the concerns of ERISA’s pre-emption provision, it fails to implicate the regulatory concerns of ERISA itself. The congressional declaration of policy, codified at 29 U. S. C. § 1001, states that ERISA was enacted because Congress found it desirable that “disclosure be made and safeguards be provided with respect to the establishment, operation, and administration of [employee benefit] plans.” § 1001(a). Representative Dent, the House sponsor of the legislation, represented that ERISA’s fiduciary standards “will prevent abuses of the special responsibilities borne by those dealing with plans.” 120 Cong. Rec. 29197 (1974). Senator Williams, the Senate sponsor, stated that these standards would safeguard employees from “such abuses as self-dealing, imprudent investing, and misappropriation of plan funds.” Id., at 29932. The focus of the statute thus is on the administrative integrity of benefit plans—which presumes that some type of administrative activity is taking place. See, e. g., H. R. Rep. No. 94-1785, p. 46 (1977) (“In electing deliberately to preclude state authority over these plans, Congress acted to insure uniformity of regulation with respect to their activities”) (emphasis added); 120 Cong. Rec. 29197 (1974) (remarks of Rep. Dent) (disclosure and reporting requirements “will enable both participants and the Federal Government to monitor the plans’ operations”) (emphasis added); id., at 29935 (remarks of Sen. Javits) (disclosure benefits are received by beneficiaries in a lump sum or on a periodic basis. This is borne out by the fact that death benefits are included in appellant’s retirement plan, with instructions on how eligibility is to be determined, benefit levels calculated, and disbursements made. App. 54-56. By contrast, appellant’s statutory obligation did not prompt the establishment of any payment program, since there were no ongoing benefits to be paid. 16 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. meant to provide employees information “covering in detail the fiscal operations of their plan”) (emphasis added). The foregoing makes clear both why ERISA is concerned with regulating benefit “plans” and why the Maine statute does not establish one. Only “plans” involve administrative activity potentially subject to employer abuse. The obligation imposed by Maine generates no such activity. There is no occasion to determine whether a “plan” is “operated” in the interest of its beneficiaries, because nothing is “operated.” No financial transactions take place that would be listed in an annual report, and no further information regarding the terms of the severance pay obligation is needed because the statute itself makes these terms clear. It would make no sense for pre-emption to clear the way for exclusive federal regulation, for there would be nothing to regulate. Under such circumstances, pre-emption would in no way serve the overall purpose of ERISA. D Appellant contends that failure to pre-empt the Maine law will create the opportunity for employers to circumvent ERISA’s regulatory requirements by persuading a State to require the type of benefit plan that the employer otherwise would establish on its own. That may be so under the rationale offered by the State Supreme Judicial Court, but that is not the rationale on which we rely today. The Maine Supreme Judicial Court rested its decision on the premise that ERISA only pre-empts state regulation of pre-existing benefit plans established by the employer, and not state-mandated benefit plans. We agree that such an approach would afford employers a readily available means of evading ERISA’s regulatory scope, thereby depriving employees of the protections of that statute. In addition, it would permit States to circumvent ERISA’s pre-emption provision, by allowing them to require directly what they are forbidden to regulate. In contrast, our analysis of the pur- FORT HALIFAX PACKING CO. v. COYNE 17 1 Opinion of the Court pose of ERISA pre-emption makes clear why the mere fact that a plan is required by a State is insufficient to fend off pre-emption. The requirements imposed by a State’s establishment of a benefit plan would pose a formidable barrier to the development of a uniform set of administrative practices. As Standard Oil Co. of California v. Agsalud, 633 F. 2d 760 (CA9 1980), illustrates, an employer would be put to the choice of operating separate ongoing benefit plans or a single plan subject to different regulatory requirements, and would face the prospect that numerous other States would impose their own distinct requirements—a result squarely inconsistent with the goal of ERISA pre-emption. Appellant’s arguments are thus well taken insofar as they are addressed to the reasoning of the court below. We have demonstrated, supra, however, they have no force with respect to a state statute that, as here, does not establish a plan. Such a statute generates no program activity that normally would be subject to ERISA regulation. Enforcement of the Maine statute presents no risk either that an employer will evade or that a State will dislodge otherwise applicable federal regulatory requirements. Nor is there any prospect that an employer will face difficulty in operating a unified administrative scheme for paying benefits. The rationale on which we rely thus does not create the dangers that appellant contends will result from upholding the Maine law. Appellant also argues that its contention that the severance obligation under the Maine statute is an ERISA plan is supported by Holland n. Burlington Industries, Inc., 772 F. 2d 1140 (CA4 1985), summarily aff’d, 477 U. S. 901 (1986), and Gilbert v. Burlington Industries, Inc., 765 F. 2d 320 (CA2 1985), summarily aff’d, 477 U. S. 901 (1986). We disagree. Those cases hold that a plan that pays severance benefits out of general assets is an ERISA plan. That holding is completely consistent with our analysis above. There was no question in the Burlington cases, as there is in this 18 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. case, whether the employer had a “plan”;10 there was a “plan” and the only issue was whether the type of benefits paid by that plan are among those covered by ERISA. The precise question was simply whether severance benefits paid by a plan out of general assets, rather than out of a trust fund, should be regarded as employee welfare benefits under 29 U. S. C. §1002.11 The courts’ conclusion that they should be so regarded took into account ERISA’s central focus on administrative integrity: if an employer has an administrative scheme for paying benefits, it should not be able to evade the requirements of the statute merely by paying those benefits out of general assets. Some severence benefit obligations by their nature necessitate an ongoing administrative scheme, but others do not. Those that do not, such as the obligation imposed in this case, simply do not involve a state law that “relate[s] to” an employee benefit “plan.” 29 U. S. C. § 1144(a).12 10 The employer had made a commitment to pay severance benefits to employees as each person left employment. This commitment created the need for an administrative scheme to pay these benefits on an ongoing basis, and the company had distributed both a Policy Manual and Employees’ Handbook that provided details on matters such as eligibility, benefit levels, and payment schedules. 772 F. 2d, at 1143-1144, and n. 1; 765 F. 2d, at 323. The fact that the employer had not complied with the requirements of ERISA in operating this scheme therefore does not, as the dissent contends, post, at 25-26, mean that no such program for paying benefits was in existence. 11 The question arose because § 1002(l)(B) provides that an employee welfare benefit plan includes a plan that provides benefits described in 29 U. S. C. § 186(c). The latter section lists, inter alia, money paid by an employer to a trust fund for severance benefits. 12 Thus, if a State required a benefit whose regularity of payment necessarily required an ongoing benefit program, it could not evade pre-emption by the simple expedient of somehow formally characterizing the obligation as a one-time, lump-sum payment triggered by the occurrence of a certain contingency. It is therefore not the case, as the dissent argues, post, at 23, that a State could dictate the payment of numerous employee benefits “by simply characterizing them as non-‘administrative.”’ Ibid. FORT HALIFAX PACKING CO. v. COYNE 19 1 Opinion of the Court The Burlington cases therefore do not support appellant’s argument. E ERISA pre-emption analysis “must be guided by respect for the separate spheres of governmental authority preserved in our federalist system.” Alessi v. Raybestos-Manhattan, Inc., 451 U. S., at 522. The argument that ERISA pre-empts state laws relating to certain employee benefits, rather than to employee benefit plans, is refuted by the express language of the statute, the purposes of the preemption provision, and the regulatory focus of ERISA as a whole. If a State creates no prospect of conflict with a federal statute, there is no warrant for disabling it from attempting to address uniquely local social and economic problems.13 Since the Maine severance payment statute raises no danger of such conflict, we hold that the statute is not preempted by ERISA. Ill Appellant also contends that Maine’s statute is pre-empted by the NLRA. In so arguing, the Company relies on the strand of NLRA pre-emption analysis that prohibits States from “imposing additional restrictions on economic weapons of self-help.” Golden State Transit Corp. v. City of Los Angeles, 475 U. S. 608, 614 (1986).14 Restriction on state activity in this area rests on the theory that pre-emption is necessary to further Congress’ intent that “the conduct involved 13 During the decade between 1971 and 1981, a total of 107 plants were closed in Maine, resulting in the direct loss of 21,215 jobs. Leighton, Plant Closings in Maine: Law and Reality, in Key Issues, No. 27, Plant Closing Legislation 1 (A. Aboud ed., 1984). Taking into account the multiplier effects of these job losses on the local communities, it is estimated that the total number of jobs lost in Maine during this period was 49,219. Id., at 3. These losses were concentrated in the poorer counties of the State and in the lower wage industries, resulting in a significant burden on local public and private social service agencies. Id., at 4. 14 The National Labor Relations Act contains no express pre-emption provision. 20 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. be unregulated because [it should be] left ‘to be controlled by the free play of economic forces.”’ Machinists n. Wisconsin Employment Relations Comm’n, 427 U. S. 132, 140 (1976) (quoting NLRB v. Nash-Finch Co., 404 U. S. 138, 144 (1971)). Appellant concedes that, unlike cases in which state laws have been struck down under this doctrine, Maine has not directly regulated any economic activity of either of the parties. See, e. g., Machinists, supra (State enjoined union members from continuing to refuse to work overtime); Garner v. Teamsters, 346 U. S. 485 (1953) (State enjoined union picketing). Nor has the State sought directly to force a party to forgo the use of one of its economic weapons. See, e. g., Golden State Transit, supra (City Council conditioned taxicab franchise renewal on settlement of strike). Nonetheless, appellant maintains that the Maine law intrudes on the bargaining activities of the parties because the prospect of a statutory obligation undercuts an employer’s ability to withstand a union’s demand for severance pay. This argument—that a State’s establishment of minimum substantive labor standards undercuts collective bargaining—was considered and rejected in Metropolitan Life Ins. Co. v. Massachusetts, 471 U. S. 724 (1985). That case involved a state law requiring that minimum mental health benefits be provided under certain health insurance policies. Appellants there presented the same argument that appellant makes in this case: “[B]ecause Congress intended to leave the choice of terms in collective-bargaining agreements to the free play of economic forces, . . . mandated-benefit laws should be pre-empted by the NLRA.” Id., at 748. The Court held, however, that the NLRA is concerned with ensuring an equitable bargaining process, not with the substantive terms that may emerge from such bargaining. “The evil Congress was addressing thus was entirely unrelated to local or federal regulation establishing minimum terms of employment.” Id., at 754. Such regulation provides protec- FORT HALIFAX PACKING CO. v. COYNE 21 1 Opinion of the Court tions to individual union and nonunion workers alike, and thus “neither encourage[s] nor discourage[s] the collectivebargaining processes that are the subject of the NLRA.” Id., at 755. Furthermore, pre-emption should not be lightly inferred in this area, since the establishment of labor standards falls within the traditional police power of the State. As a result, held the Court: “When a state law establishes a minimal employment standard not inconsistent with the general legislative goals of the NLRA, it conflicts with none of the purposes of the Act.” Id., at 757. It is true that the Maine statute gives employees something for which they otherwise might have to bargain. That is true, however, with regard to any state law that substantively regulates employment conditions. Both employers and employees come to the bargaining table with rights under state law that form a “ ‘backdrop’ ” for their negotiations. Ibid, (quoting Taggart n. Weinacker^, Inc., 397 U. S. 223, 228 (1970) (concurring opinion)). Absent a collective-bargaining agreement, for instance, state common law generally permits an employer to run the workplace as it wishes. The employer enjoys this authority without having to bargain for it. The parties may enter negotiations designed to alter this state of affairs, but, if impasse is reached, the employer may rely on pre-existing state law to justify its authority to make employment decisions; that same state law defines the rights and duties of employees. Similarly, Maine provides that employer and employees may negotiate with the intention of establishing severance pay terms. If impasse is reached, however, preexisting state law determines the right of employees to a certain level of severance pay and the duty of the employer to provide it. Thus, the mere fact that a state statute pertains to matters over which the parties are free to bargain cannot support a claim of pre-emption, for “there is nothing in the NLRA . . . which expressly forecloses all state regulatory power with respect to those issues . . . that may be the 22 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. subject of collective bargaining.” Malone n. White Motor Corp., 435 U. S. 497, 504-505 (1978). Appellant maintains that this case is distinguishable from Metropolitan Life. It points out that, unlike Metropolitan Life, the statutory obligation at issue here is optional, since it applies only in the absence of an agreement between employer and employees. Therefore, the Company argues, the Maine law cannot be regarded as establishing a genuine minimum labor standard. The fact that the parties are free to devise their own severance pay arrangements, however, strengthens the case that the statute works no intrusion on collective bargaining. Maine has sought to balance the desirability of a particular substantive labor standard against the right of self-determination regarding the terms and conditions of employment. If a statute that permits no collective bargaining on a subject escapes NLRA pre-emption, see Metropolitan Life, surely one that permits such bargaining cannot be pre-empted.15 We therefore find that Maine’s severance payment law is “a valid and unexceptional exercise of the [State’s] police power.” Metropolitan Life, 471 U. S., at 758. Since “Congress developed the framework for self-organization and collective bargaining of the NLRA within the larger body of state law promoting public health and safety,” id., at 756, the Maine statute is not pre-empted by the NLRA.16 15 Appellant also contends that, unlike the statute in Metropolitan Life, the Maine law does not fall equally upon union and nonunion employees. Nonunion employers, it argues, are free unilaterally to escape their statutory obligation by establishing severance payment levels, while unionized employers must engage in collective bargaining in order to achieve the same result. Any difference in the ease of establishing alternative severance payment obligations, however, flows not from the statute, but from the basic fact that a nonunion employer is freer to set employment terms than is a unionized employer. 16 We also find no support for an argument of pre-emption under the rule established in San Diego Building Trades Council n. Garmon, 359 U. S. 236 (1959), since the Maine statute does not purport to regulate any con FORT HALIFAX PACKING CO. u COYNE 23 1 White, J., dissenting IV We hold that the Maine severance pay statute is not preempted by ERISA, since it does not “relate to any employee benefit plan” under that statute. 29 U. S. C. § 1144(a). We hold further that the law is not pre-empted by the NLRA, since its establishment of a minimum labor standard does not impermissibly intrude upon the collective-bargaining process. The judgment of the Maine Supreme Judicial Court is therefore Affirmed. Justice White, with whom The Chief Justice, Justice O’Connor, and Justice Scalia join, dissenting. The Court rejects appellant’s pre-emption challenge to Maine’s severance pay statute by reasoning that the statute does not create a “plan” under ERISA because it does not require an “administrative scheme” to administer the payment of severance benefits. By making pre-emption turn on the existence of an “administrative scheme,” the Court creates a loophole in ERISA’s pre-emption statute, 29 U. S. C. § 1144, which will undermine Congress’ decision to make employeebenefit plans a matter of exclusive federal regulation. The Court’s rule requiring an established “administrative scheme” as a prerequisite for ERISA pre-emption will allow States to effectively dictate a wide array of employee benefits that must be provided by employers by simply characterizing them as non-“administrative.” The Court has also chosen to ignore completely what precedent exists as to what constitutes a “plan” under ERISA. I dissent because it is incredible to believe that Congress intended that the broad preemption provision contained in ERISA would depend upon the extent to which an employer exercised administrative foresight in preparing for the eventual payment of employee benefits. duct subject to regulation by the National Labor Relations Board. See Metropolitan Life Ins. Co. v. Massachusetts, 471 U. S., at 748-749. 24 OCTOBER TERM, 1986 White, J., dissenting 482 U. S. ERISA pre-empts “any and all State laws insofar as they may now or hereafter relate to any employee benefit plan . . . .” 29 U. S. C. §1144. Congress defined an “employee welfare benefit plan” as “any plan, fund, or program which was heretofore or is hereafter established or maintained by an employer or an employee organization” and which provides certain benefits, including severance pay. 29 U. S. C. §1002(1). See Gilbert v. Burlington Industries, Inc., 765 F. 2d 320, 325 (CA2 1985), summarily aff’d, 477 U. S. 901 (1986). A state law “which requires employers to pay em-. ployees specific benefits clearly ‘relate[s] to’ benefit plans” as contemplated by ERISA’s pre-emption provision. Shaw v. Delta Air Lines, Inc., 463 U. S. 85, 97 (1983). I would have thought this to be the end of the pre-emption inquiry. Here, the Maine statute clearly creates an employee benefit plan, and having created an ERISA plan, the statute plainly “relates to” such a plan. The Maine Supreme Judicial Court, in effect, acknowledged as much, but held that Maine’s statute was not pre-empted by ERISA because it was created by the state legislature instead of by a private employer. Apparently recognizing the flaw inherent in this reasoning, the majority nevertheless struggles to achieve its desired result by asserting that the statute does not create a “plan” because it does not require an employer to establish an administrative scheme. I cannot accept this conclusion. First, § 1002(1) establishes no requirement that a “plan” meet any specific formalities or that there be some policy manual or employee handbook to effectuate it. Cf. ante, at 14-15, n. 9. In reading such a requirement into § 1002(1), the majority ignores the obvious: when a Maine employer is called upon to discharge its legislatively mandated duty under the severance pay statute, the funds from which it pays the benefits do not materialize out of thin air. The Maine Legislature has presumed, as it is so entitled, that employers will comply with the dictates of the statute’s requirements. That an employer’s liability is contingent upon an FORT HALIFAX PACKING CO. v. COYNE 25 1 White, J., dissenting event that may never happen does not make the plan that the legislature has imposed upon employers any less of a plan. And that there may be imprudent employers who either are unaware of the severance pay statute or order their business affairs as if the statute’s obligations do not exist—and it is upon the behavior of this class of employers that the majority seemingly relies in concluding that the severance pay statute does not embody an “administrative scheme”—in no way supports the remarkable conclusion that the statutory obligations do not constitute a plan for the payment of severance benefits. Second, in concluding that Maine’s statute does not establish a “plan” as contemplated by ERISA, the Court overrules, sub silentio, recent decisions of this Court. Gilbert v. Burlington Industries, Inc., supra, involved an employer’s policy to pay severance benefits to employees who were involuntarily terminated. The employer had no separate fund from which to make severance pay payments, and, of particular note, there was virtually no “administrative scheme” to effectuate the program: “The granting or denial of severance pay was automatic upon termination. Plaintiffs [employees] allege that Burlington never sought to comply with ERISA respecting its severance pay policy. That is, they claim that: it never published or filed an annual report, a financial statement, a plan description or a statement of plan modifications; it did not designate a fiduciary for the plan or inform employees of their rights under ERISA and the plan; there was no established claims procedure; and, apart from the company’s ‘open door’ grievance policy, there was no established appeals procedure.” Gilbert, 765 F. 2d, at 323. The employees and numerous amici claimed that “a promise or agreement to pay severance benefits, without more, does not constitute a welfare benefit plan within the meaning of ERISA.” Id., at 324. The Second Circuit rejected this contention, id., at 325, and we summarily affirmed, 477 U. S. 901 (1986). See 26 OCTOBER TERM, 1986 White, J., dissenting 482 U. S. also Holland n. Burlington Industries, Inc., 772 F. 2d 1140 (CA4 1985), summarily aff’d, 477 U. S. 901 (1986). The Court characterizes Standard Oil Co. of California v. Agsalud, 633 F. 2d 760, 766 (CA9 1980), summarily aff’d, 454 U. S. 801 (1981), as holding that ERISA pre-empted Hawaii’s health care statute because it impaired employers’ ability to “structur[e] [their] administrative practices according to a set of uniform guidelines.” Ante, at 13. But that case involved more than administrative uniformity. Indeed, in Agsalud, the Ninth Circuit expressly rejected the argument that ERISA was concerned only with the administration of benefit plans, not state statutes which require employers to provide particular employee benefits: “Appellants in the district court argued that since ERISA was concerned primarily with the administration of benefit plans, its provisions were not intended to prevent the operation of laws like the Hawaii Act pertaining principally to benefits rather than administration. There is, however, nothing in the statute to support such a distinction between the state laws relating to benefits as opposed to administration.” 633 F. 2d, at 765. The Ninth Circuit held that the Hawaii Act “directly and expressly regulates employers and the type of benefits they provide employees. It must ‘relate to’ employee benefit plans within the meaning of ERISA’s broad pre-emption provision . . . .” Id., at 766. Representatives of the State of Hawaii appealed to this Court, No. 80-1841, claiming, inter alia, that the State’s police power permits it to require employers to provide certain employee benefits, and that Hawaii’s statute “in no way conflicts with any substantive provision in ERISA, since that statute requires no benefits at all.” Juris. Statement, O. T. 1981, No. 80-1841, p. 7. We disagreed and summarily affirmed. 454 U. S. 801 (1981). The Court’s “administrative-scheme” rationale provides States with a means of circumventing congressional intent, clearly expressed in §1144, to pre-empt all state laws that relate to employee benefit plans. For that reason, I dissent. FALL RIVER DYEING & FINISHING CORP. v. NLRB 27 Syllabus FALL RIVER DYEING & FINISHING CORP. v. NATIONAL LABOR RELATIONS BOARD CERTIORARI TO THE UNITED STATES COURT OF APPEALS FOR THE FIRST CIRCUIT No. 85-1208. Argued March 2, 1987—Decided June 1, 1987 This case involves interpretation of the ruling in NLRB v. Bums International Security Services, Inc., 406 U. S. 272, that the new employer, succeeding to another’s business, had an obligation to bargain with the union representing the predecessor’s employees. Sterlingwale Corp., which had operated a textile dyeing and finishing plant, laid off all of its production employees in February 1982 and finally went out of business in late summer. During this period, one of its former officers and the president of one of its major customers formed petitioner company, intending to engage in one aspect of Sterlingwale’s business and to take advantage of its assets and its work force. Petitioner acquired Sterlingwale’s plant, real property, equipment, and some of its remaining inventory, and began operating out of Sterlingwale’s former facilities and hiring employees in September 1982, with an initial hiring goal of one full shift of workers. In October 1982, the union that had represented Sterlingwale’s production and maintenance employees for almost 30 years requested petitioner to recognize it as the bargaining agent for petitioner’s employees and to begin collective bargaining. Petitioner refused the request. At that time, a majority of petitioner’s employees were ex-Sterlingwale employees, as also was true in mid-January 1983, when petitioner met its initial hiring goal of one shift of workers. By mid-April 1983, petitioner had reached two shifts, and, for the first time, ex-Sterlingwale employees were in the minority. The same working conditions existed as under Sterlingwale, and over half of petitioner’s business came from ex-Sterlingwale customers. In November 1982, the union filed an unfair labor practice charge with the National Labor Relations Board, alleging that in refusing to bargain petitioner violated §§ 8(a)(1) and (5) of the National Labor Relations Act (NLRA). An Administrative Law Judge (ALJ) concluded that (1) petitioner was a “successor” to Sterlingwale, (2) the proper date for determining whether the majority of petitioner’s employees were ex-Sterlingwale employees (necessary to require petitioner to bargain with the union) was not midApril, when petitioner had two shifts working, but mid-January when petitioner had obtained a “representative complement” of employees, (3) the union’s October 1982 demand for bargaining (necessary to trigger pe 28 OCTOBER TERM, 1986 Syllabus 482 U. S. titioner’s obligation), although premature, was “of a continuing nature” and was still in effect in mid-January, and (4) petitioner thus committed an unfair labor practice in refusing to bargain. The Board affirmed the ALJ’s decision, and the Court of Appeals enforced the Board’s order. Held: 1. A “successor” employer’s obligation to bargain is not limited to the situation (as in Bums) where the union in question only recently was certified before the transition in employers. Where, as here, a union certified for more than one year has a rebuttable presumption of majority status, that status continues despite the change in employers. Although the new employer is not bound by the substantive provisions of the predecessor’s bargaining agreement, it has an obligation to bargain with the union so long as it is in fact a successor of the old employer and the majority of its employees were employed by its predecessor. Pp. 36-41. 2. Petitioner was a “successor” to Sterlingwale. The Board’s approach in determining this question, approved in Bums, is based upon the totality of the circumstances and requires that the Board focus on whether there is “substantial continuity” between the enterprises, with particular emphasis on the retained employees’ perspective as to whether their job situations are essentially unaltered. The Board’s determination that there was “substantial continuity” here and that petitioner was Sterlingwale’s successor is supported by substantial evidence in the record. It is not dispositive that there was a 7-month hiatus between Sterlingwale’s demise and petitioner’s start-up, or that employees were hired through newspaper advertisements rather than through Sterlingwale’s employment records. Pp. 42-46. 3. The Board’s “substantial and representative complement” rule— which fixes the moment when the determination is to be made as to whether a majority of the successor’s employees are former employees of the predecessor, a moment that triggers the successor’s bargaining obligation—is reasonable in the successorship context. Petitioner’s proposal that majority status be determined instead at the “full complement” stage so that all the employees would have a voice in the selection of their bargaining representative fails to consider the employees’ significant interest in being represented as soon as possible. Nor does the Board’s rule place an unreasonable burden on the employer. The application of the Board’s rule to the facts of this case, moreover, is supported by substantial record evidence. Pp. 46-52. 4. The Board’s “continuing demand” rule—whereby a union’s premature demand for bargaining continues in effect until the successor acquires a “substantial and representative complement” of employees that triggers its obligation to bargain—also is reasonable in the successorship FALL RIVER DYEING & FINISHING CORP. v. NLRB 29 27 Opinion of the Court context. The rule places a minimal burden on the successor and makes sense in light of the union’s position. It would make no sense to require the union repeatedly to renew its bargaining demand in the hope of having it correspond with the “substantial and representative complement” date, when, with little trouble, the employer can regard a previous demand as a continuing one. Pp. 52-54. 775 F. 2d 425, affirmed. Blackmun, J., delivered the opinion of the Court, in which Brennan, Marshall, Stevens, and Scalia, JJ., joined, and in Parts I and III of which White, J., joined. Powell, J., filed a dissenting opinion, in which Rehnquist, C. J., and O’Connor, J., joined, post, p. 54. Ira Drogin argued the cause and filed briefs for petitioner. Deputy Solicitor General Cohen argued the cause for respondent. With him on the brief were Solicitor General Fried, Christopher J. Wright, Norton J. Come, Linda Sher, and Robert C. Bell, Jr. * Justice Blackmun delivered the opinion of the Court.! In this case we are confronted with the issue whether the National Labor Relations Board’s decision is consistent with NLRB v. Burns International Security Services, Inc., 406 U. S. 272 (1972). In Bums, this Court ruled that the new employer, succeeding to the business of another, had an obligation to bargain with the union representing the predecessor’s employees. Id., at 278-279. We first must decide whether Bums is limited to a situation where the union only recently was certified before the transition in employers, or whether that decision also applies where the union is entitled to a presumption of majority support. Our inquiry then pro *Briefs of amici curiae urging reversal were filed for the Chamber of Commerce of the United States by Peter G. Nash and Stephen A. Bokat; and for the Legal Foundation of America by Jean Fleming Powers and David Crump. Marsha S. Berzon, D. Bruce Shine, and Laurence Gold filed a brief for the American Federation of Labor and Congress of Industrial Organizations et al. as amici curiae urging affirmance. Nustice White joins only Parts I and III of this opinion. 30 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. ceeds to three questions that concern rules the Labor Board has developed in the successorship context. First, we must determine whether there is substantial record evidence to support the Board’s conclusion that petitioner was a “successor” to Sterlingwale Corp., its business predecessor. Second, we must decide whether the Board’s “substantial and representative complement” rule, designed to identify the date when a successor’s obligation to bargain with the predecessor’s employees’ union arises, is consistent with Bums, is reasonable, and was applied properly in this case. Finally, we must examine the Board’s “continuing demand” principle to the effect that, if a union has presented to a successor a premature demand for bargaining, this demand continues in effect until the successor acquires the “substantial and representative complement” of employees that triggers its obligation to bargain. I For over 30 years before 1982, Sterlingwale operated a textile dyeing and finishing plant in Fall River, Mass. Its business consisted basically of two types of dyeing, called, respectively, “converting” and “commission.” Under the converting process, which in 1981 accounted for 60% to 70% of its business, see App. 149, Sterlingwale bought unfinished fabrics for its own account, dyed and finished them, and then sold them to apparel manufacturers. Id., at 123. In commission dyeing, which accounted for the remainder of its business, Sterlingwale dyed and finished fabrics owned by customers according to their specifications. Id., at 124. The financing and marketing aspects of converting and commission dyeing are different. Converting requires capital to purchase fabrics and a sales force to promote the finished products. Id., at 123. The production process, however, is the same for both converting and commission dyeing. Id., at 98. In the late 1970’s the textile-dyeing business, including Sterlingwale’s, began to suffer from adverse economic condi- FALL RIVER DYEING & FINISHING CORP. v. NLRB 31 27 Opinion of the Court tions and foreign competition. After 1979, business at Sterlingwale took a serious turn for the worse because of the loss of its export market, id., at 127-128, and the company reduced the number of its employees, id., at 192-195. Finally, in February 1982, Sterlingwale laid off all its production employees, primarily because it no longer had the capital to continue the converting business. Id., at 77-78, 104,130-132. It retained a skeleton crew of workers and supervisors to ship out the goods remaining on order and to maintain the corporation’s building and machinery. Id., at 147-148. In the months following the layoff, Leonard Ansin, Sterlingwale’s president, liquidated the inventory of the corporation and, at the same time, looked for a business partner with whom he could “resurrect the business.” Id., at 114-115, 146-147. Ansin felt that he owed it to the community and to the employees to keep Sterlingwale in operation. Id., at 103-104. For almost as long as Sterlingwale had been in existence, its production and maintenance employees had been represented by the United Textile Workers of America, AFL-CIO, Local 292 (Union). Id., at 60-61. The most recent collective-bargaining agreement before Sterlingwale’s demise had been negotiated in 1978 and was due to expire in 1981. By an agreement dated October 1980, however, in response to the financial difficulties suffered by Sterlingwale, the Union agreed to amend the 1978 agreement to extend its expiration date by one year, until April 1, 1982, without any wage increase and with an agreement to improve labor productivity. Id., at 353-355. In the months following the final February 1982 layoff, the Union met with company officials over problems involving this job action, and, in particular, Sterlingwale’s failure to pay premiums on group-health insurance. Id., at 66-67, 86,131. In addition, during meetings with Ansin, Union officials told him of their concern with Sterlingwale’s future and their interest in helping to keep the 32 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. company operating or in meeting with prospective buyers. Id., at 67-68, 86, 146-147. In late summer 1982, however, Sterlingwale finally went out of business. It made an assignment for the benefit of its creditors, id., at 115,147, primarily Ansin’s mother, who was an officer of the corporation and holder of a first mortgage on most of Sterlingwale’s real property, id., at 113, and the Massachusetts Capital Resource Corporation (MCRC), which held a security interest on Sterlingwale’s machinery and equipment, id., at 113-114. Ansin also hired a professional liquidator to dispose of the company’s remaining assets, mostly its inventory, at auction. Id., at 115. During this same period, a former Sterlingwale employee and officer, Herbert Chace, and Arthur Friedman, president of one of Sterlingwale’s major customers, Marcamy Sales Corporation (Marcamy), formed petitioner Fall River Dyeing & Finishing Corp. Chace, who had resigned from Sterlingwale in February 1982, had worked there for 27 years, had been vice president in charge of sales at the time of his departure, and had participated in collective bargaining with the Union during his tenure at Sterlingwale. Id., at 189, 232. Chace and Friedman formed petitioner with the intention of engaging strictly in the commission-dyeing business and of taking advantage of the availability of Sterlingwale’s assets and work force. Id., at 203-204, 223-224. Accordingly, Friedman had Marcamy acquire from MCRC and Ansin’s mother Sterlingwale’s plant, real property, and equipment, id., at 238-272, and convey them to petitioner, id., at 278-289? Petitioner also obtained some of Sterlingwale’s remaining inventory at the liquidator’s auction. Id., at 200-202, 290-293. Chace became petitioner’s vice president in charge of operations and Friedman became its president. Id., at 190, 232. In September 1982, petitioner began operating out of Sterlingwale’s former facilities and began hiring employees. 1 Petitioner did not acquire one of the three buildings formerly used by Sterlingwale, App. 200-201, and closed one that it did acquire, id., at 195. FALL RIVER DYEING & FINISHING CORP. v. NLRB 33 27 Opinion of the Court Id., at 206-207. It advertised for workers and supervisors in a local newspaper, id., at 197-198, and Chace personally got in touch with several prospective supervisors, id., at 197. Petitioner hired 12 supervisors, of whom 8 had been supervisors with Sterlingwale and 3 had been production employees there. Id., at 196, 220-222. In its hiring decisions for production employees, petitioner took into consideration recommendations from these supervisors and a prospective employee’s former employment with Sterlingwale. Id., at 223-224. Petitioner’s initial hiring goal was to attain one full shift of workers, which meant from 55 to 60 employees. Id., at 208. Petitioner planned to “see how business would be” after this initial goal had been met and, if business permitted, to expand to two shifts. Ibid. The employees who were hired first spent approximately four to six weeks in start-up operations and an additional month in experimental production. Id., at 156-157, 207, 226-227. By letter dated October 19, 1982, the Union requested petitioner to recognize it as the bargaining agent for petitioner’s employees and to begin collective bargaining. Id., at 360. Petitioner refused the request, stating that, in its view, the request had “no legal basis.” Id., at 362. At that time, 18 of petitioner’s 21 employees were former employees of Sterlingwale. See 272 N. L. R. B. 839, 840 (1984). By November of that year, petitioner had employees in a complete range of jobs, had its production process in operation, and was handling customer orders, App. 225-226; by midJanuary 1983, it had attained its initial goal of one shift of workers, id., at 225, 227. Of the 55 workers in this initial shift, a number that represented over half the workers petitioner would eventually hire, 36 were former Sterlingwale employees. Tr. of Oral Arg. 28. Petitioner continued to expand its work force, and by mid-April 1983, it had reached two full shifts. For the first time, ex-Sterlingwale employees were in the minority but just barely so (52 or 53 out of 107 employees). App. 294-302; Tr. of Oral Arg. 28. 34 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. Although petitioner engaged exclusively in commission dyeing, the employees experienced the same conditions they had when they were working for Sterlingwale. The production process was unchanged and the employees worked on the same machines, in the same building, with the same job classifications, under virtually the same supervisors. App. 152-156, 205-206. Over half the volume of petitioner’s business came from former Sterlingwale customers, and, in particular, Marcamy. Id., at 314-316. On November 1, 1982, the Union filed an unfair labor practice charge with the Board, alleging that in its refusal to bargain petitioner had violated §§ 8(a)(1) and (5) of the National Labor Relations Act (NLRA), 49 Stat. 452, as amended, 29 U. S. C. §§ 158(a)(1) and (5).2 After a hearing, the Administrative Law Judge (ALJ) decided that, on the facts of the case, petitioner was a successor to Sterlingwale. 272 N. L. R. B., at 840. He observed that petitioner therefore would have an obligation to bargain with the Union if the majority of petitioner’s employees were former employees of Sterlingwale. He noted that the proper date for making this determination was not mid-April, when petitioner first had two shifts working, but mid-January, when petitioner had attained a “representative complement” of employees. Ibid. The ALJ acknowledged that a demand for bargaining from the Union was necessary to trigger petitioner’s obligation to bargain, but noted that the Union’s demand of October 1982, although premature, was “of a continuing nature.” Ibid. 2 These read in pertinent part: “§ 158. Unfair labor practices “(a) Unfair labor practices by employer “It shall be an unfair labor practice for an employer— “(1) to interfere with, restrain, or coerce employees in the exercise of the rights guaranteed in section 157 of this title; “(5) to refuse to bargain collectively with the representatives of his employees, subject to the provisions of section 159(a) of this title.” FALL RIVER DYEING & FINISHING CORP. v. NLRB 35 27 Opinion of the Court Thus, in the view of the ALJ, petitioner’s duty to bargain arose in mid-January because former Sterlingwale employees then were in the majority and because the Union’s October demand was still in effect. Petitioner thus committed an unfair labor practice in refusing to bargain. In a brief decision and order, the Board, with one member dissenting, affirmed this decision. Id., at 839.3 The Court of Appeals for the First Circuit, also by a divided vote, enforced the order. 775 F. 2d 425 (1985). The court first found, id., at 428-430, that the Board’s determination that petitioner was Sterlingwale’s successor was consistent with Bums and was “supported by substantial evidence in the record.” 775 F. 2d, at 430. The court observed: “The differences between [petitioner’s] business and Sterlingwale’s are not sufficiently significant to require a finding that the continuity of the enterprise, viewed from the employees’ standpoint, was broken.” Ibid. The court then noted that the Board’s longstanding “substantial and representative complement” standard, id., at 431, which the ALJ applied in this case, is an attempt to establish a method for determining when a successor has to bargain with the predecessor’s union in a situation where, at the moment of the transition between the old and new enterprises, it is not clear when the new employer will reach a “full complement of employees.” Id., at 430-431. According to the court, the Board’s determination that petitioner had “employed a substantial and representative complement of its workforce in mid-January” was reasonable. Id., at 431. Finally, the court found that the Board’s rule treating a premature union demand for bargaining as a continuing demand also was reasonable and “practical” and entitled to deference. Id., at 432-433.4 3 In the view of the dissenting member, the Union’s complaint should have been dismissed because the Union failed to renew its bargaining request after petitioner properly denied it. 272 N. L. R. B. 839 (1984). 4 The dissenting judge argued that the Board’s “substantial and representative complement” rule wTas contrary to this Court’s decision in NLRB 36 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. Because of the importance of the successorship issue in labor law, and because of our interest in the rules developed by the Board for successorship cases, we granted certiorari. 476 U. S. 1139 (1986). II Fifteen years ago in NLRB v. Bums International Security Services, Inc., 406 U. S. 272 (1972), this Court first dealt with the issue of a successor employer’s obligation to bargain with a union that had represented the employees of its predecessor. In Bums, about four months before the employer transition, the security-guard employees of Wackenhut Corp, had chosen a particular union as their bargaining representative and that union had negotiated a collective-bargaining agreement with Wackenhut. Wackenhut, however, lost its service contract on certain airport property to Burns. Burns proceeded to hire 27 of the Wackenhut guards for its 42-guard operation at the airport. Burns told its guards that, as a condition of their employment, they must join the union with which Burns already had collective-bargaining agreements at other locations. When the union that had represented the Wackenhut employees brought unfair labor practice charges against Burns, this Court agreed with the Board’s determination that Burns had an obligation to bargain with this union. We observed: “In an election held but a few months before, the union had been designated bargaining agent for the employees in the unit and a majority of these employees had been hired by Burns for work in the identical unit. It is undisputed that Burns knew all the relevant facts in this regard and was aware of the certification and of the v. Burns International Security Services, Inc., 406 U. S. 272 (1972), that petitioner was not a successor of Sterlingwale, and that, in light of the premature bargaining demand of the Union, which petitioner properly rejected, petitioner had a “good faith doubt” about the Union’s majority status that relieved it of any obligation to bargain. 775 F. 2d, at 434-441. FALL RIVER DYEING & FINISHING CORP. v. NLRB 37 27 Opinion of the Court existence of a collective-bargaining contract. In these circumstances, it was not unreasonable for the Board to conclude that the union certified to represent all employees in the unit still represented a majority of the employees and that Bums could not reasonably have entertained a good-faith doubt about that fact. Bums’ obligation to bargain with the union over terms and conditions of employment stemmed from its hiring of Wack-enhut’s employees and from the recent election and Board certification.” Id., at 278-279. Although our reasoning in Bums was tied to the facts presented there, see id., at 274, we suggested that our analysis would be equally applicable even if a union with which a successor had to bargain had not been certified just before the transition in employers. We cited with approval, id., at 279 and 281, Board and Court of Appeals decisions where it “ha[d] been consistently held that a mere change of employers or of ownership in the employing industry is not such an ‘unusual circumstance’ as to affect the force of the Board’s certification within the normal operative period if a majority of employees after the change of ownership or management were employed by the preceding employer.” Id., at 279. Several of these cases involved successorship situations where the union in question had not been certified only a short time before the transition date. See, e. g., NLRB v. Auto Ventshade, Inc., 276 F. 2d 303, 305 (CA5 1960); Tom-A-Hawk Transit, Inc. v. NLRB, 419 F. 2d 1025, 1026 (CA7 1969). Moreover, in defining “the force of the Board’s certification within the normal operative period,” 406 U. S., at 279, we referred in Bums to two presumptions regarding a union’s majority status following certification. See id., at 279, n. 3. First, after a union has been certified by the Board as a bargaining-unit representative, it usually is entitled to a conclusive presumption of majority status for one year following the certification. See ibid., citing Brooks v. NLRB, 38 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. 348 U. S. 96, 98-99 (1954); see also 29 U. S. C. § 159(c)(3) (“No election shall be directed in any bargaining unit or any subdivision within which in the preceding twelve-month period, a valid election shall have been held”). Second, after this period, the union is entitled to a rebuttable presumption of majority support. 406 U. S., at 279, n. 3, citing Celanese Corp, of America, 95 N. L. R. B. 664, 672 (1951). These presumptions are based not so much on an absolute certainty that the union’s majority status will not erode following certification, as on a particular policy decision. The overriding policy of the NLRA is “industrial peace.” Brooks v. NLRB, 348 U. S., at 103. The presumptions of majority support further this policy by “promot[ing] stability in collective-bargaining relationships, without impairing the free choice of employees.” Terrell Machine Co., 173 N. L. R. B. 1480 (1969), enf’d, 427 F. 2d 1088 (CA4), cert, denied, 398 U. S. 929 (1970). In essence, they enable a union to concentrate on obtaining and fairly administering a collectivebargaining agreement without worrying that, unless it produces immediate results, it will lose majority support and will be decertified. See Brooks v. NLRB, 348 U. S., at 100. The presumptions also remove any temptation on the part of the employer to avoid good-faith bargaining in the hope that, by delaying, it will undermine the union’s support among the employees. See ibid.; see also R. Gorman, Labor Law 53 (1976).6 The upshot of the presumptions is to permit unions 6 Because the Chamber of Commerce as amicus curiae overlooks or ignores our acceptance of the presumptions in Burns as well as their significance, it can contend that Burns “turned on the particular circumstances in that case—the recent union election and certification which arguably provided a factual basis for presuming that a majority of Bums’ employees wanted to be represented by the union,” Brief for Chamber of Commerce of United States as Amicus Curiae 17, and that Bums requires “that there must be some rational factual basis for presumptions of majority union support among a successor’s workforce. ” Id., at 18-19. This misunderstanding of the nature of the presumptions leads to the Chamber’s proposal that, in a situation where the successor employer arrives at a “full complement” FALL RIVER DYEING & FINISHING CORP. v. NLRB 39 27 Opinion of the Court to develop stable bargaining relationships with employers, which will enable the unions to pursue the goals of their members, and this pursuit, in turn, will further industrial peace. The rationale behind the presumptions is particularly pertinent in the successorship situation and so it is understandable that the Court in Burns referred to them. During a transition between employers, a union is in a peculiarly vulnerable position. It has no formal and established bargaining relationship with the new employer, is uncertain about the new employer’s plans, and cannot be sure if or when the new employer must bargain with it. While being concerned with the future of its members with the new employer, the union also must protect whatever rights still exist for its members under the collective-bargaining agreement with the predecessor employer.6 Accordingly, during this unsettling transition period, the union needs the presumptions of majority status to which it is entitled to safeguard its members’ rights and to develop a relationship with the successor. The position of the employees also supports the application of the presumptions in the successorship situation. If the employees find themselves in a new enterprise that substantially resembles the old, but without their chosen bargaining representative, they may well feel that their choice of a union of employees only gradually, the union should be forced to petition for a Board election to establish again its majority support. Id., at 26. Acceptance of the Chamber’s views, which essentially advocate a rejection of the presumptions as they are presently understood, logically would require such an election whenever any doubts existed about a union’s majority status, regardless of whether the employer remained the same. 6 The difficulty a union faces during an employer-transition period is graphically exhibited by the facts of this case. The Union was confronted with the layoff. App. 64. Although officials at Sterlingwale were willing to meet with it, the Union unsuccessfully attempted to have Sterlingwale honor its commitments under the collective-bargaining agreement, particularly those dealing with health benefits. Id., at 78-86. Moreover, despite the Union’s desire to participate in the transition between employers, it was left entirely in the dark about petitioner’s acquisition. Id., at 68-69. 40 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. is subject to the vagaries of an enterprise’s transformation. This feeling is not conducive to industrial peace. In addition, after being hired by a new company following a layoff from the old, employees initially will be concerned primarily with maintaining their new jobs. In fact, they might be inclined to shun support for their former union, especially if they believe that such support will jeopardize their jobs with the successor or if they are inclined to blame the union for their layoff and problems associated with it.7 Without the presumptions of majority support and with the wide variety of corporate transformations possible, an employer could use a successor enterprise as a way of getting rid of a labor contract and of exploiting the employees’ hesitant attitude towards the union to eliminate its continuing presence. In addition to recognizing the traditional presumptions of union majority status, however, the Court in Bums was careful to safeguard “ ‘the rightful prerogative of owners independently to rearrange their businesses.’” Golden State Bottling Co. v. NLRB, 414 U. S. 168, 182 (1973), quoting John Wiley & Sons, Inc. v. Livingston, 376 U. S. 543, 549 (1964). We observed in Bums that, although the successor has an obligation to bargain with the union, it “is ordinarily free to set initial terms on which it will hire the employees of a predecessor,” 406 U. S., at 294, and it is not bound by the substantive provisions of the predecessor’s collectivebargaining agreement. Id., at 284. We further explained that the successor is under no obligation to hire the employees of its predecessor, subject, of course, to the restriction that it not discriminate against union employees in its hiring. Id., at 280, and n. 5; see also Howard Johnson Co. v. Hotel Employees, 417 U. S. 249, 262, and n. 8 (1974). Thus, to 7 In fact, it appears that the dissatisfaction with the Union felt by some former Sterlingwale employees who were hired by petitioner was due to the Union’s inability to obtain benefits, such as payment for health insurance, severance pay, and vacation pay, from the failing Sterlingwale. App. 168, 174, 179-180. See also n. 18, infra. FALL RIVER DYEING & FINISHING CORP. v. NLRB 41 27 Opinion of the Court a substantial extent the applicability of Bums rests in the hands of the successor. If the new employer makes a conscious decision to maintain generally the same business and to hire a majority of its employees from the predecessor, then the bargaining obligation of § 8(a)(5) is activated. This makes sense when one considers that the employer intends to take advantage of the trained work force of its predecessor.8 Accordingly, in Bums we acknowledged the interest of the successor in its freedom to structure its business and the interest of the employees in continued representation by the union. We now hold that a successor’s obligation to bargain is not limited to a situation where the union in question has been recently certified. Where, as here, the union has a rebuttable presumption of majority status, this status continues despite the change in employers. And the new employer has an obligation to bargain with that union so long as the new employer is in fact a successor of the old employer and the majority of its employees were employed by its predecessor.9 8 If, during negotiations, a successor questions a union’s continuing majority status, the successor “may lawfully withdraw from negotiation at any time following recognition if it can show that the union had in fact lost its majority status at the time of the refusal to bargain or that the refusal to bargain was grounded on a good-faith doubt based on objective factors that the union continued to command majority support.” Harley-Davidson Transp. Co., 273 N. L. R. B. 1531 (1985). The AL J made no express finding on the issue of petitioner’s good-faith doubt. Moreover, an employer, unsure of a union’s continued majority support, may petition the Board for another election. See NLRB v. Financial Institution Employees, 475 U. S. 192, 198 (1986); Brooks v. NLRB, 348 U. S. 96, 101 (1954). Petitioner did not request an election. ’Last Term, we struck down a recently adopted Board rule requiring that nonunion employees must be permitted to vote in a certified union’s decision to affiliate with another union. Under that rule, if the union did not permit such voting, the Board would not amend the union’s certification or compel the employer to bargain with the reorganized union. See NLRB v. Financial Institution Employees, 475 U. S., at 201. This rule was in direct conflict with a previous Board position whereby the affiliation 42 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. Ill We turn now to the three rules, as well as to their application to the facts of this case, that the Board has adopted for the successorship situation. The Board, of course, is given considerable authority to interpret the provisions of the NLRA. See NLRB v. Financial Institution Employees, 475 U. S. 192, 202 (1986). If the Board adopts a rule that is rational and consistent with the Act, see ibid., then the rule is entitled to deference from the courts. Moreover, if the Board’s application of such a rational rule is supported by substantial evidence on the record, courts should enforce the Board’s order. See Beth Israel Hospital n. NLRB, 437 U. S. 483, 501 (1978); Universal Camera Corp. v. NLRB, 340 U. S. 474, 488 (1951). These principles also guide our review of the Board’s action in a successorship case. See, e. g., Golden State Bottling Co. v. NLRB, 414 U. S., at 181. was permitted so long as the members of the union voted for it and there was substantial continuity between the new and the old unions. Id., at 199-200. In rejecting the new rule, we observed that “ ‘[t]he industrial stability sought by the Act would unnecessarily be disrupted if every union organizational adjustment were to result in displacement of the employerbargaining representative relationship. ’ ” Id., at 202-203, quoting Canton Sign Co., 174 N. L. R. B. 906, 909 (1969), enf. denied on other grounds, 457 F. 2d 832 (CA6 1972). We observed: “In many cases, a majority of employees will continue to support the union despite any changes precipitated by affiliation.” 475 U. S., at 203. In our view, “[t]he Act assumes that stable bargaining relationships are best maintained by allowing an affiliated union to continue representing a bargaining unit unless the Board finds that the affiliation raises a question of representation. The Board’s rule contravenes this assumption, since an employer may invoke a perceived procedural defect to cease bargaining even though the union succeeds the organization the employees chose, the employees have made no effort to decertify the union, and the employer presents no evidence to challenge the union’s majority status.” Id., at 209. As explained earlier, this concern about stable bargaining relations and the presumption of a union’s majority status are equally applicable in the instant case. FALL RIVER DYEING & FINISHING CORP. v. NLRB 43 27 Opinion of the Court A In Bums we approved the approach taken by the Board and accepted by courts with respect to determining whether a new company was indeed the successor to the old. 406 U. S., at 280-281, and n. 4. This approach, which is primarily factual in nature and is based upon the totality of the circumstances of a given situation, requires that the Board focus on whether the new company has “acquired substantial assets of its predecessor and continued, without interruption or substantial change, the predecessor’s business operations.” Golden State Bottling Co. v. NLRB, 414 U. S., at 184. Hence, the focus is on whether there is “substantial continuity” between the enterprises. Under this approach, the Board examines a number of factors: whether the business of both employers is essentially the same; whether the employees of the new company are doing the same jobs in the same working conditions under the same supervisors; and whether the new entity has the same production process, produces the same products, and basically has the same body of customers. See Bums, 406 U. S., at 280, n. 4; Aircraft Magnesium, Division of Grico Corp., 265 N. L. R. B. 1344, 1345 (1982), enf’d, 730 F. 2d 767 (CA9 1984); Premium Foods, Inc., 260 N. L. R. B. 708, 714 (1982), enf’d, 709 F. 2d 623 (CA9 1983). In conducting the analysis, the Board keeps in mind the question whether “those employees who have been retained will understandably view their job situations as essentially unaltered.” See Golden State Bottling Co., 414 U. S., at 184; NLRB v. Jeffries Lithograph Co., 752 F. 2d 459, 464 (CA9 1985). This emphasis on the employees’ perspective furthers the Act’s policy of industrial peace. If the employees find themselves in essentially the same jobs after the employer transition and if their legitimate expectations in continued representation by their union are thwarted, their 44 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. dissatisfaction may lead to labor unrest. See Golden State Bottling Co., 414 U. S., at 184. Although petitioner does not challenge the Board’s “substantial continuity” approach, it does contest the application of the rule to the facts of this case. Essentially for the reasons given by the Court of Appeals, 775 F. 2d, at 430, however, we find that the Board’s determination that there was “substantial continuity” between Sterlingwale and petitioner and that petitioner was Sterlingwale’s successor is supported by substantial evidence in the record. Petitioner acquired most of Sterlingwale’s real property, its machinery and equipment, and much of its inventory and materials.10 It introduced no new product line. Of particular significance is the fact that, from the perspective of the employees, their jobs did not change. Although petitioner abandoned converting dyeing in exclusive favor of commission dyeing, this change did not alter the essential nature of the employees’ jobs, because both types of dyeing involved the same production process. The job classifications of petitioner were the same as those of Sterlingwale; petitioner’s employees worked on the same machines under the direction of supervisors most of whom were former supervisors of Sterlingwale. The record, in fact, is clear that petitioner acquired Sterlingwale’s assets with the express purpose of taking advantage of its predecessor’s work force. 10 Petitioner makes much of the fact that it purchased the assets of Sterlingwale on the “open market.” Brief for Petitioner 17. Petitioner, however, overlooks the fact that it was formed with the express purpose of acquiring Sterlingwale’s assets, a purpose it accomplished by having its parent company acquire some of Sterlingwale’s major assets and then transferring them to petitioner. So long as there are other indicia of “substantial continuity,” the way in which a successor obtains the predecessor’s assets is generally not determinative of the “substantial continuity” question. See Howard Johnson Co. n. Hotel Employees, 417 U. S. 249, 257 (1974); Golden State Bottling Co. v. NLRB, 414 U. S. 168, 182, n. 5 (1973); see also R. Gorman, Labor Law 122 (1976). FALL RIVER DYEING & FINISHING CORP. v. NLRB 45 27 Opinion of the Court We do not find determinative of the successorship question the fact that there was a 7-month hiatus between Sterlingwale’s demise and petitioner’s start-up. Petitioner argues that this hiatus, coupled with the fact that its employees were hired through newspaper advertisements—not through Sterlingwale employment records, which were not transferred to it—resolves in its favor the “substantial continuity” question. See Brief for Petitioner 16-17, 20-22; see also 775 F. 2d, at 439 (dissenting opinion). Yet such a hiatus is only one factor in the “substantial continuity” calculus and thus is relevant only when there are other indicia of discontinuity. See NLRB v. Band-Age, Inc., 534 F. 2d, 1, 5 (CAI), cert, denied, 429 U. S. 921 (1976). Conversely, if other factors indicate a continuity between the enterprises, and the hiatus is a normal start-up period, the “totality of the circumstances” will suggest that these circumstances present a successorship situation. See NLRB v. Daneker Clock Co., 516 F. 2d 315, 316 (CA4 1975); C. G. Conn, Ltd., 197 N. L. R. B. 442, 446-447 (1972), enf’d, 474 F. 2d 1344 (CA5 1973). For the reasons given above, this is a case where the other factors suggest “substantial continuity” between the companies despite the 7-month hiatus. Here, moreover, the extent of the hiatus between the demise of Sterlingwale and the start-up of petitioner is somewhat less than certain. After the February layoff, Sterlingwale retained a skeleton crew of supervisors and employees that continued to ship goods to customers and to maintain the plant. In addition, until the assignment for the benefit of the creditors late in the summer, Ansin was seeking to resurrect the business or to find a buyer for Sterlingwale. The Union was aware of these efforts. Viewed from the employees’ perspective, therefore, the hiatus may have been much less than seven months. Although petitioner hired the employees through advertisements, it often relied on recommendations from supervisors, themselves formerly employed by Sterlingwale, and intended 46 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. the advertisements to reach the former Sterlingwale work force.11 Accordingly, we hold that, under settled law, petitioner was a successor to Sterlingwale. We thus must consider if and when petitioner’s duty to bargain arose. B In Bums, the Court determined that the successor had an obligation to bargain with the union because a majority of its employees had been employed by Wackenhut. 406 U. S., at 278-279. The “triggering” fact for the bargaining obligation was this composition of the successor’s work force.12 The 11 Similarly, in light of the general continuity between Sterlingwale and petitioner from the perspective of the employees, we do not find determinative the differences between the two enterprises cited by petitioner. Petitioner’s change in marketing and sales, Brief for Petitioner 20, appears to have had no effect on the employer-employee relationship. That petitioner did not assume Sterlingwale’s liabilities or trade name, id., at 16, also is not sufficient to outweigh the other factors. See NLRB v. Band-Age, Inc., 534 F. 2d 1, 5 (CAI), cert, denied, 429 U. S. 921 (1976); Zim’s Foodliner, Inc. v. NLRB, 495 F. 2d 1131, 1133-1134 (CA7), cert, denied, 419 U. S. 838 (1974). Moreover, the mere reduction in petitioner’s size, in comparison to that of Sterlingwale, see Brief for Petitioner 17-18, does not change the nature of the company so as to defeat the employees’ expectations in continued representation by their Union. See NLRB v. Middle-boro Fire Apparatus, Inc., 590 F. 2d 4, 8 (CAI 1978). 12 After Burns, there was some initial confusion concerning this Court’s holding. It was unclear if work force continuity would turn on whether a majority of the successor’s employees were those of the predecessor or on whether the successor had hired a majority of the predecessor’s employees. Compare 406 U. S., at 281 (“[A] majority of the employees hired by the new employer are represented by a recently certified bargaining agent”), with id., at 278 (“[T]he union had been designated bargaining agent for the employees in the unit and a majority of these employees had been hired by Bums”). See also Howard Johnson Co. v. Hotel Employees, 417 U. S., at 263 (“[S]uccessor employer hires a majority of the predecessor’s employees”); Golden State Bottling Co. v. NLRB, 414 U. S., at 184, n. 6 (same). The Board, with the approval of the Courts of Appeals, has adopted the former interpretation. See Spruce Up Corp., 209 N. L. R. B. 194, 196 (1974), enf’d, 529 F. 2d 516 (CA4 1975); United Main FALL RIVER DYEING & FINISHING CORP. v. NLRB 47 27 Opinion of the Court Court, however, did not have to consider the question when the successor’s obligation to bargain arose: Wackenhut’s contract expired on June 30 and Bums began its services with a majority of former Wackenhut guards on July 1. See id., at 275. In other situations, as in the present case, there is a start-up period by the new employer while it gradually builds its operations and hires employees. In these situations, the Board, with the approval of the Courts of Appeals, has adopted the “substantial and representative complement” rule for fixing the moment when the determination as to the composition of the successor’s work force is to be made.13 If, at this particular moment, a majority of the successor’s employees had been employed by its predecessor, then the successor has an obligation to bargain with the union that represented these employees.14 tenance & Mfg. Co., 214 N. L. R. B. 529, 532-534 (1974); Saks & Co. v. NLRB, 634 F. 2d 681, 684-686, and nn. 2 and 3 (CA2 1980) (and cases cited therein); see also Note, Appropriate Standards of Successor Employer Obligations under Wiley, Howard Johnson, and Bums, 25 Wayne L. Rev. 1279, 1299 (1979). This issue is not presented by the instant case. 13See, e. g., Indianapolis Mack Sales & Service, Inc., 272 N. L. R. B. 690, 694-696 (1984), enf. denied on other grounds, 802 F. 2d 280 (CA7 1986); NLRB v. Jeffries Lithograph Co., 752 F. 2d 459, 467 (CA9 1985); Aircraft Magnesium, a Division of Grico Corp., 265 N. L. R. B. 1344, 1345 (1982), enf’d, 730 F. 2d 767 (CA9 1984); Hudson River Aggregates, Inc., 246 N. L. R. B. 192, 197-198 (1979), enf’d, 639 F. 2d 865, 870 (CA2 1981). 14 Petitioner argues that Bums requires that the majority determination be made only when the successor has attained a “full complement” of employees. Brief for Petitioner 22, 29-31; see also Brief for Chamber of Commerce of United States as Amicus Curiae 20-21. Petitioner and the amicus particularly rely for this argument on one reference in Bums to a “full complement”: “Although a successor employer is ordinarily free to set initial terms on which it will hire the employees of a predecessor, there will be instances in which it is perfectly clear that the new employer plans to retain all of the employees in the unit and in which it will be appropriate to have him initially consult with the employees’ bargaining representative before he fixes terms. In other situations, however, it may not be clear until the succes- 48 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. This rule represents an effort to balance “‘the objective of insuring maximum employee participation in the selection of a bargaining agent against the goal of permitting employees to be represented as quickly as possible.’” 775 F. 2d, at 430-431, quoting NLRB v. Pre-Engineered Building Products, Inc., 603 F. 2d 134, 136 (CAIO 1979).15 In deciding sor employer has hired his full complement of employees that he has a duty to bargain with a union, since it will not be evident until then that the bargaining representative represents a majority of the employees in the unit as required by §9(a) of the Act, 29 U. S. C. § 159(a).” 406 U. S., at 294-295. This remark, however, was made after the Court had resolved the successorship issue and when it was examining whether a successor would have to bargain with the union before setting the initial terms and conditions of employment. In particular, in using the term “full complement,” the Court was distinguishing the exceptional situation, alluded to in the prior sentence, in which a successor should consult with the union before setting these terms arid conditions, from the standard situation in which a successor could set its own terms free of the union’s involvement. The Court was not defining “full complement” with respect to fixing the moment when the successor would have to bargain with the union. Burns therefore lends no support to an interpretation of that term to mean that the successor’s bargaining obligation arises only when it has hired all the employees it intends to employ. 16 The “substantial and representative complement” rule originated in the context of the initial representation election when, faced with an expanding or contracting work force, the Board had to determine the appropriate time for an election. See, e. g., Clement-Blythe Companies, 182 N. L. R. B. 502 (1970), enf’d, 77 LRRM 2373 (CA4 1971). The rationale for the rule was as follows: “The Board must often balance what are sometimes conflicting desiderata, the insurance of maximum employee participation in the selection of a bargaining agent, and permitting employees who wish to be represented as immediate representation as possible. Thus, it would unduly frustrate existing employees’ choice to delay selection of a bargaining representative for months or years until the very last employee is on board. Conversely, it would be pointless to hold an election for very few employees when in a relatively short period the employee complement is expected to multiply many times.” 182 N. L. R. B., at 502. Similar reasoning applies in the successorship context. On the one hand, there is a concern to allow as many employees as possible of the successor FALL RIVER DYEING & FINISHING CORP. v. NLRB 49 27 Opinion of the Court when a “substantial and representative complement” exists in a particular employer transition, the Board examines a number of factors. It studies “whether the job classifications designated for the operation were filled or substantially filled and whether the operation was in normal or substantially normal production.” See Premium Foods, Inc. v. NLRB, 709 F. 2d 623, 628 (CA9 1983). In addition, it takes into consideration “the size of the complement on that date and the time expected to elapse before a substantially larger complement would be at work ... as well as the relative certainty of the employer’s expected expansion.” Ibid. Petitioner contends that the Board’s “representative complement” rule is unreasonable, given that it injures the representation rights of many of the successor’s employees and that it places significant burdens upon the successor, which is unsure whether and when the bargaining obligation will arise. Brief for Petitioner 24-31; see also Brief for Chamber of Commerce of United States as Amicus Curiae 21-25. According to petitioner, if majority status is determined at the “full complement” stage, all the employees will have a voice in the selection of their bargaining representative, and this will reveal if the union truly has the support of most of the successor’s employees. This approach, however, focuses only on the interest in having a bargaining representative selected by the majority of the employees. It fails to take into account the significant interest of employees in being represented as soon as possible. The latter interest is especially heightened in a situation where many of the successor’s employees, who were formerly represented by a union, find themselves after the employer transition in essentially the same enterprise, but without their bargaining representative. Having the new employer refuse to bargain with the chosen representative of these employees “disrupts the employees’ morale, deters their organizational activities, and to participate in the selection of the union. On the other hand, the previous choice of a union by those employees of the successor who had worked for the predecessor should not be frustrated. 50 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. discourages their membership in unions.” Franks Bros. Co. v. NLRB, 321 U. S. 702, 704 (1944). Accordingly, petitioner’s “full complement” proposal must fail.16 Nor do we believe that this “substantial and representative complement” rule places an unreasonable burden on the employer. It is true that, if an employer refuses to bargain with the employees once the representative complement has been attained, it risks violating § 8(a)(5). Furthermore, if an employer recognizes the union before this complement has been reached, this recognition could constitute a violation of § 8(a)(2), which makes it an unfair labor practice for an employer to support a labor organization. 29 U. S. C. § 158(a)(2). And, unlike the initial election situation, see n. 15, supra, here the employer, not the Board, applies this rule. We conclude, however, that in this situation the successor is in the best position to follow a rule the criteria of which are straightforward.17 The employer generally will know with tolerable certainty when all its job classifications have been filled or substantially filled, when it has hired a majority of the employees it intends to hire, and when it has begun normal production. Moreover, the “full complement” standard advocated by petitioner is not necessarily easier for a succes- 16 Long ago, in Brooks v. NLRB, 348 U. S. 96 (1954), this Court observed: “To allow employers to rely on employees’ rights in refusing to bargain with the formally designated union is not conducive to that end [of industrial peace], it is inimical to it.” Id., at 103. Moreover, the employees are not powerless to reject a union that they believe no longer commands their support. See NLRB v. Financial Institution Employees, 475 U. S., at 198. 17 The distinction between the successorship situation and the initial election context, where the Board itself applies the “substantial and representative complement” rule, lies partly in the fact that, in the latter case, the Board is involved in supervising the selection of a bargaining representative for the bargaining unit. See Gorman, Labor Law, at 46-49. In contrast, where, as in this case, a union already has been selected and is entitled to a presumption of majority support, the Board’s involvement is more limited. FALL RIVER DYEING & FINISHING CORP. v. NLRB 51 27 Opinion of the Court sor to apply than is the “substantial and representative complement.” In fact, given the expansionist dreams of many new entrepreneurs, it might well be more difficult for a successor to identify the moment when the “full complement” has been attained, which is when the business will reach the limits of the new employer’s initial hopes, than it would be for this same employer to acknowledge the time when its business has begun normal production—the moment identified by the “substantial and representative complement” rule.18 18 In addition, even if an employer were to err as to the “substantial and representative complement” date and thus were to recognize the union prematurely, its good-faith violation of § 8(a)(2) would be subject only to a remedial order. See Garment Workers v. NLRB, 366 U. S. 731, 740 (1961). Similarly, we assume that if the employer were to refuse to recognize a union on the basis of its reasonable good-faith belief that it had not yet hired a “substantial and representative complement,” the Board would likewise enter a remedial order, see ibid., with no collateral consequences such as a decertification bar. Finally, if the employer has a good-faith doubt about the union’s continuing majority status, it has several remedies available to it. See n. 8, supra. Petitioner in its brief offers, as support for its position that its employees, once they reached “full complement,” were opposed to the Union, certain employee petitions signed three days before the Board hearing on May 2, 1983. Brief for Petitioner 25; App. 364-367. We approve the Board’s and Court of Appeals’ treatment of these petitions. The ALJ ruled that such petitions were not relevant to petitioner’s good-faith doubt about the Union’s majority status at the mid-January date when petitioner’s bargaining obligation arose. Id., at 177-178. The Court of Appeals observed that “once it has been determined that an employer has unlawfully withheld recognition of an employees’ bargaining representative, the employer cannot defend against a remedial bargaining by pointing to an intervening loss of employee support for the union when such loss of support is a foreseeable consequence of the employer’s unfair labor practice.” 775 F. 2d, at 433. That petitioner’s refusal to bargain with the Union undermined the employees’ support for the Union and thus led to the petitions is suggested by evidence in the record. An employee testified that the petitions were signed out of employees’ fear that the Board proceeding might delay an expected wage raise. App. 183. Thus, the very refusal to bargain on petitioner’s part that led to the unfair labor practice hearing produced the petitions on which petitioner would rely. 52 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. We therefore hold that the Board’s “substantial and representative complement” rule is reasonable in the successorship context. Moreover, its application to the facts of this case is supported by substantial record evidence. The Court of Appeals observed that by mid-January petitioner “had hired employees in virtually all job classifications, had hired at least fifty percent of those it would ultimately employ in the majority of those classifications, and it employed a majority of the employees it would eventually employ when it reached full complement.” 775 F. 2d, at 431-432. At that time petitioner had begun normal production. Although petitioner intended to expand to two shifts, and, in fact, reached this goal by mid-April, that expansion was contingent expressly upon the growth of the business. Accordingly, as found by the Board and approved by the Court of Appeals, midJanuary was the period when petitioner reached its “substantial and representative complement.” Because at that time the majority of petitioner’s employees were former Sterlingwale employees, petitioner had an obligation to bargain with the Union then. C We also hold that the Board’s “continuing demand” rule is reasonable in the successorship situation. The successor’s duty to bargain at the “substantial and representative complement” date is triggered only when the union has made a bargaining demand. Under the “continuing demand” rule, when a union has made a premature demand that has been rejected by the employer, this demand remains in force until the moment when the employer attains the “substantial and representative complement.” See, e. g., Aircraft Magnesium, 265 N. L. R. B., at 1345, n. 9; Spruce Up Corp., 209 N. L. R. B. 194, 197 (1974), enf’d, 529 F. 2d 516 (CA4 1975). Such a rule, particularly when considered along with the “substantial and representative complement” rule, places a minimal burden on the successor and makes sense in light of the union’s position. Once the employer has concluded that FALL RIVER DYEING & FINISHING CORP. v. NLRB 53 27 Opinion of the Court it has reached the appropriate complement, then, in order to determine whether its duty to bargain will be triggered, it has only to see whether the union already has made a demand for bargaining. Because the union has no established relationship with the successor and because it is unaware of the successor’s plans for its operations and hiring, it is likely that, in many cases, a union’s bargaining demand will be premature. It makes no sense to require the union repeatedly to renew its bargaining demand in the hope of having it correspond with the “substantial and representative complement” date, when, with little trouble, the employer can regard a previous demand as a continuing one.19 The reasonableness of the “continuing demand” rule is demonstrated by the facts of this case. Although the Union had asked Ansin to inform it about his plans for Sterlingwale so that it could become involved in the employer transition, the Union learned about this transition only after it had become a fait accompli. Without having any established relationship with petitioner, it therefore is not surprising that the Union’s October bargaining demand was premature. The Union, however, made clear after this demand that, in its view, petitioner had a bargaining obligation: the Union filed an unfair labor practice charge in November. Petitioner responded by denying that it had any duty to bargain. Rather than being a successor confused about when a bargaining obligation might arise, petitioner took an initial posi 19 In contrast, in the situation where a union has not yet been recognized as a representative of a bargaining unit, the rationale for the “continuing demand” rule is not so compelling as it is in the successorship context where a union is entitled to a presumption of majority status and where the employer simply has to determine whether, at the appropriate date, the predecessor’s employees are in the majority. In the initial recognition context the union, not the employer, is in the best position to have access to the relevant information—whether the union has the majority support of the employees. 54 OCTOBER TERM, 1986 Powell, J., dissenting 482 U. S. tion—and stuck with it—that it never would have any bargaining obligation with the Union.20 The judgment of the Court of Appeals is affirmed. It is so ordered. Justice Powell, with whom The Chief Justice and Justice O’Connor join, dissenting. Today the Court holds that petitioner Fall River Dyeing & Finishing Corp, violated §§ 8(a)(1) and (a)(5) of the National Labor Relations Act, 29 U. S. C. §§ 158(a)(1), (5) (NLRA), by refusing to bargain with a union that claims to represent its workers. The Court agrees with the National Labor Relations Board (NLRB or Board) that this duty to bargain arose because petitioner is a “successor” to Sterlingwale Corp., a defunct entity that had engaged in a similar line of business. The Court also agrees that the duty to bargain arose when petitioner had brought its first shift into full operation. The theory is that petitioner then had hired a “substantial and representative complement” of its work force. In my view, the Court has misconstrued the successorship doctrine and misapplied the substantial complement test. Accordingly, I dissent.1 “Although the unfair labor practice charge was filed and the complaint issued before mid-January when petitioner’s obligation arose and the violation occurred, an unfair labor practice proceeding may be based on actions following the filing of a complaint. See Curtiss-Wright Corp., Wright Aeronautical Div. n. NLRB, 347 F. 2d 61, 73-74 (CA3 1965). 1 As a preliminary matter, the Court holds that if one company is a successor to another, it has an obligation to bargain with the prior company’s union even though that union had not been certified recently by the workers. Ante, at 41. As the Court notes, the finding of successorship in NLRB n. Bums International Security Services, Inc., 406 U. S. 272 (1972), was based partly on the fact that the union had been certified almost immediately before the employees were hired by the new company. Id., at 278. Although the Court concludes that the successorship doctrine is not limited to such cases, it certainly would be reasonable to assume that the more remote the certification, the weaker the presumption should be that the union retains majority support. In any event, I do not reach the FALL RIVER DYEING & FINISHING CORP. v. NLRB 55 27 Powell, J., dissenting I A Although the Court describes the background of this case in great detail, it gives insufficient consideration to a number of critical facts. On February 12,1982, a financially troubled Sterlingwale ceased operations and indefinitely laid off its production workers, retaining only a skeleton crew to ship out the remaining orders and liquidate the inventory. The collective-bargaining agreement (CBA) between the union and Sterlingwale was allowed to expire in April, and the company ceased paying the workers’ life and health insurance premiums. Attempts to obtain new financing to keep the business afloat were unsuccessful. Sterlingwale commenced its liquidation by making an assignment for the benefit of creditors, and then hired a professional liquidator to sell the remaining assets at a public auction. By mid to late summer of 1982, all business activity had ceased, and the company permanently closed its doors. Petitioner Fall River Dyeing & Finishing Corp, was incorporated at the end of August 1982. It bought most of Sterlingwale’s machinery, furniture, and fixtures. It also bought a portion of the Sterlingwale inventory at the public auction.2 Three weeks later it began recruiting new employees by placing ads in the local newspaper. Petitioner hired some former Sterlingwale workers, although by no means all or even a large percentage of those who had been laid off.3 When making its hiring decisions, petitioner took issue because I think it is clear that petitioner is not a successor to Sterlingwale. 2 At the auction, that apparently took place in October 1982, petitioner bought $13,000 worth of inventory out of the $30,000 worth that was sold. App. 200-201. 3 More than 150 workers were laid off in February 1982; by mid-January 1983, petitioner had hired perhaps 36 of them. See App. 92-93; 775 F. 2d 425, 428 (CAI 1985). The record does not reveal how many of the laid-off Sterlingwale workers applied for positions with the new company, al 56 OCTOBER TERM, 1986 Powell, J., dissenting 482 U. S. into account the applicant’s experience with either Sterlingwale or other finishing plants, App. 223; although the former Sterlingwale supervisors who had been hired were consulted as to the former Sterlingwale workers who applied, there is no finding that these workers as a group received a hiring preference. Once the new company began operations in November 1982, it performed commission finishing work exclusively, rather than the converting finishing that had accounted for 60%-70% of Sterlingwale’s business.4 B Of course, a decision by the NLRB that one company is a successor of another is entitled to deference, and its conclusions will be upheld if they are based on substantial record evidence. See Golden State Bottling Co. v. NLRB, 414 U. S. 168, 181 (1973). The critical question in determining successorship is whether there is “substantial continuity” between the two businesses. Aircraft Magnesium, Division of Grico Corp., 265 N. L. R. B. 1344, 1345 (1982), enf’d, 730 F. 2d 767 (CA9 1984). See also NLRB v. Bums International Security Services, Inc., 406 U. S. 272, 279-281 (1972). Here the Board concluded that there was sufficient though petitioner’s vice president testified that he received what “seemed like thousands” of applications in response to the ads. App. 198. 4 The Court finds little significance in this switch from converting to commission work, since the change was thought to have no direct effect on the employer-employee relationship. See ante, at 46, n. 11. This difference alone would not be determinative, but it hardly is irrelevant. The change meant that unlike petitioner, Sterlingwale did not have to maintain a sales force or retail outlet to sell its cloth, nor did it have to allocate capital for purchasing material. These facts are pertinent to the question whether there is substantial continuity between the two enterprises. The Board in the past has recognized the significance of similar considerations that have an indirect impact on the workers. See, e. g., Gladding Corp., 192 N. L. R. B. 200, 202 (1971) (change in suppliers); Radiant Fashions, Inc., 202 N. L. R. B. 938, 940 (1973) (substantial change in identity of customers). FALL RIVER DYEING & FINISHING CORP. v. NLRB 57 27 Powell, J., dissenting continuity between petitioner and Sterlingwale, primarily because the workers did the same finishing work on the same equipment for petitioner as they had for their former employer. See 272 N. L. R. B. 839, 840 (1984) (decision of Administrative Law Judge (AL J)). In reaching this conclusion, however, the Board, and now the Court, give virtually no weight to the evidence of discontinuity, that I think is overwhelming. In this case the undisputed evidence shows that petitioner is a completely separate entity from Sterlingwale. There was a clear break between the time Sterlingwale ceased normal business operations in February 1982 and when petitioner came into existence at the end of August.5 In addition, it is apparent that there was no direct contractual or other business relationship between petitioner and Sterlingwale. See App. 205. Although petitioner bought some of Sterlingwale’s inventory, it did so by outbidding several other buyers on the open market. Also, the purchases at the public sale involved only tangible assets. Petitioner did not buy Sterlingwale’s trade name or goodwill, nor did it assume any of its liabilities. And while over half of petitioner’s business (measured in dollars) came from former Sterlingwale customers, apparently this was due to the new company’s skill in marketing its services. There was no sale or transfer of customer lists, and given the 9-month interval between the time that Sterlingwale ended production and petitioner commenced its operations in November, the natural conclusion is that the new business attracted customers through its own efforts. No other explanation was offered. Cf. Lincoln Private Police, Inc., 189 N. L. R. B. 717, 719 (1971) (finding it relevant to the successorship question that, while the new 5 The Court dismisses the effect of this 7-month hiatus, stating that such a break is important only if there are “other indicia of discontinuity.” Ante, at 45 (citing NLRB v. Band-Age, Inc., 534 F. 2d 1, 5 (CAI), cert, denied, 429 U. S. 921 (1976)). Of course, as noted in the text, there are a number of other “indicia of discontinuity” in this case. 58 OCTOBER TERM, 1986 Powell, J., dissenting 482 U. S. business acquired many of the former company’s clients, “it did so by means of independent solicitation”). Any one of these facts standing alone may be insufficient to defeat a finding of successorship, but together they persuasively demonstrate that the Board’s finding of “substantial continuity” was incorrect.6 The Court nevertheless is unpersuaded. It views these distinctions as not directly affecting the employees’ expectations about their job status or the status of the union as their representative, even though the CBA with the defunct corporation had long since expired. See Golden State Bottling Co. v. NLRB, supra, at 184 (emphasizing the importance of the workers’ perception that their job situation continues “essentially unaltered”). Yet even from the employees’ perspective, there was little objective evidence that the jobs with petitioner were simply a continuation of those at Sterlingwale. When all of the production employees were laid off indefinitely in February 1982, there could have been little hope—and certainly no reasonable expectation—that Sterlingwale would ever reopen. Nor was it reasonable for the employees to expect that Sterlingwale’s failed textile operations would be resumed by a corporation not then in existence. The CBA had expired in April with no serious effort to renegotiate it, and with several of the employees’ benefits left unpaid. The possibility of further employment with Sterlingwale then disappeared entirely in August 1982 when 6 The case before us bears a substantial resemblance to Radiant Fashions, Inc., supra. In that case, the alleged successor was engaged in a business similar to that of its predecessor, at the same location, with the same equipment, the same supervisory personnel, and a reduced but similar work force. The Board nevertheless ruled that the company was not a successor. It based its conclusion on four factors: (i) there was a “lengthy” hiatus of 2Vz to 3 months between the time the first company shut down and the second company began production; (ii) the second company bought only the assets of the first business, rather than an ongoing enterprise; (iii) the second company served a different market; and (iv) there was no transfer of customers as a result of the sale. 202 N. L. R. B., at 940-941. FALL RIVER DYEING & FINISHING CORP. v. NLRB 59 27 Powell, J., dissenting the company liquidated its remaining assets. Cf. Textile Workers n. Darlington Manufacturing Co., 380 U. S. 263, 274 (1965) (the “closing of an entire business . . . ends the employer-employee relationship”). After petitioner was organized, it advertised for workers in the newspaper, a move that hardly could have suggested to the old workers that they would be reinstated to their former positions. The sum of these facts inevitably would have had a negative “effect on the employees’ expectations of rehire.” See Aircraft Magnesium, 265 N. L. R. B., at 1346. See also Radiant Fashions, Inc., 202 N. L. R. B. 938, 940 (1973). The former employees engaged by petitioner found that the new plant was smaller, and that there would be fewer workers, fewer shifts, and more hours per shift than at their prior job. Moreover, as petitioner did not acquire Sterlingwale’s personnel records, the benefits of having a favorable work record presumably were lost to these employees. In deferring to the NLRB’s decision, the Court today extends the successorship doctrine in a manner that could not have been anticipated by either the employer or the employees. I would hold that the successorship doctrine has no application when the break in continuity between enterprises is as complete and extensive as it was here. II Even if the evidence of genuine continuity were substantial, I could not agree with the Court’s decision. As we have noted in the past, if the presumption of majority support for a union is to survive a change in ownership, it must be shown that there is both a continuity of conditions and a continuity of work force. Howard Johnson Co. v. Hotel Employees, 417 U. S. 249, 263 (1974). This means that unless a majority of the new company’s workers had been employed by the former company, there is no justification for assuming that the new employees wish to be represented by the former union, or by any union at all. See Spruce Up Corp., 209 60 OCTOBER TERM, 1986 Powell, J., dissenting 482 U. S. N. L. R. B. 194, 196 (1974), enf’d, 529 F. 2d 516 (CA4 1975); 209 N. L. R. B., at 200 (member Kennedy, concurring in part and dissenting in part); Saks & Co. v. NLRB, 634 F. 2d 681, 685-686 (CA2 1980). Indeed, the rule hardly could be otherwise. It would be contrary to the basic principles of the NLRA simply to presume in these cases that a majority of workers supports a union when more than half of them have never been members, and when there has been no election. The Court acknowledges that when petitioner completed the employment of its anticipated work force in April 1983, less than 50% of its employees formerly had worked for Sterlingwale. It nevertheless finds that the new company violated its duty to bargain, because at an earlier date chosen by the Board, a majority of the work force formerly had worked for Sterlingwale. The NLRB concluded that even though petitioner was still in the process of hiring employees, by the middle of January it had hired a “substantial and representative complement,” when its first shift was adequately staffed and most job categories had been filled. In my view, the Board’s decision to measure the composition of petitioner’s work force in mid-January is unsupportable. The substantial and representative complement test can serve a useful role when the hiring process is sporadic, or the future expansion of the work force is speculative. But as the Court recognized in NLRB v. Bums International Security Services, Inc., in some cases “it may not be clear until the successor employer has hired his full complement of employees that he has a duty to bargain with a union, since it will not be evident until then that the bargaining representative represents a majority of the employees in the unit. ” 406 U. S., at 295. Indeed, where it is feasible to wait and examine the full complement—as it was here—it clearly is fairer to both employer and employees to do so. The substantial complement test provides no more than an estimate of the percentage of employees from the old company that eventually will be part of FALL RIVER DYEING & FINISHING CORP. v. NLRB 61 27 Powell, J., dissenting the new business, and thus often will be an imperfect measure of continuing union support. The risks of relying on such an estimate are obvious. If the “substantial complement” examined by the Board at a particular time contains a disproportionate number of workers from the old company, the result either might be that the full work force is deprived of union representation that a majority favors, or is required to accept representation that a majority does not want. Accordingly, unless the delay or uncertainty of future expansion would frustrate the employees’ legitimate interest in early representation—a situation not shown to exist here—there is every reason to wait until the full anticipated work force has been employed. In this case the date chosen by the NLRB for measuring the substantial complement standard is unsupportable, and the Court’s affirmance of this choice, curious. In prior decisions, courts and the Board have looked not only to the number of workers hired and positions filled on a particular date, but also to “the time expected to elapse before a substantially larger complement would be at work ... as well as the relative certainty of the employer’s expected expansion.” Premium Foods, Inc. v. NLRB, 709 F. 2d 623, 628 (CA9 1983). See also St. John of God Hospital, Inc., 260 N. L. R. B. 905 (1982). Here the anticipated expansion was both imminent and reasonably definite. The record shows that in January petitioner both expected to, and in fact subsequently did, hire a significant number of new employees to staff its second shift. Although the Court finds that the growth of the work force was “contingent” on business conditions, neither the AL J nor the NLRB made such a finding.7 In fact, they both noted that by January 15, the second shift already had begun limited operations. See 272 N. L. R. B., at 839, n. 1 (“In 7 The evidence shows that in the textile industry, two shifts are necessary for proper finishing work. See 775 F. 2d 425, 428 (CAI 1985). See also App. 227. Thus, it was clear in mid-January that petitioner would need more employees in the immediate future. 62 OCTOBER TERM, 1986 Powell, J., dissenting 482 U. S. mid-January [petitioner] had one shift in full operation and had started a second shift”); id., at 840. In fact, less than three months after the duty to bargain allegedly arose, petitioner had nearly doubled the size of its mid-January work force by hiring the remaining 50-odd workers it needed to reach full production. This expansion was not unexpected; instead, it closely tracked petitioner’s original forecast for growth during its first few months in business.8 Thus there was no reasonable basis for selecting mid-January as the time that petitioner should have known that it should commence bargaining.9 As the Court notes, the substantial complement rule reflects the need to balance “the objective of insuring maximum employee participation in the selection of a bargaining agent against the goal of permitting employees to be represented as quickly as possible.” Ante, at 48 (citations and internal 8 Petitioner’s vice president of operations testified: “We planned to have a full one shift operation of 55 to 60 employees. And after we reached that goal, and then we’d see how business would be, and then we’d had [sic] planned that by the end of March, April, we should be in a full two shift operation and up to our expected production.” App. 208. There is no evidence that business conditions during this period were such that the company considered changing its hiring goal. 9 The NLRB’s reliance on the substantial complement standard is particularly puzzling on these facts, since the evidence shows that the “substantial” complement examined by the Board was not truly “representative” of the work force. When the unfair labor practice hearing was held on May 2, 1983, petitioner already had hired a full complement of workers. At that point the company employed 106-109 workers, less than half of whom were former Sterlingwale employees. Rather than rely on this accurate measure of the composition of the work force, the AL J looked back to the middle of January, and concluded that petitioner should have acted differently because it appeared at the time that most of the workers who eventually would be represented by the union would be ex-Sterlingwale employees. In other words, the Board ruled that petitioner violated the NLRA because it failed to make the same estimate in January that the AL J made in May—an estimate that already had proved to be erroneous at the time that the AL J made it. FALL RIVER DYEING & FINISHING CORP. v. NLRB 63 27 Powell, J., dissenting quotation marks omitted). The decision today “balances” these interests by overprotecting the latter and ignoring the former. In an effort to ensure that some employees will not be deprived of representation for even a short time, the Court requires petitioner to recognize a union that has never been elected or accepted by a majority of its workers. For the reasons stated, I think that the Court’s decision is unfair both to petitioner, who hardly could have anticipated the date chosen by the Board, and to most of petitioner’s employees, who were denied the opportunity to choose their union. I dissent. 64 OCTOBER TERM, 1986 Syllabus 482 U. S. UNITED STATES v. HOHRI et al. CERTIORARI TO THE UNITED STATES COURT OF APPEALS FOR THE DISTRICT OF COLUMBIA CIRCUIT No. 86-510. Argued April 20, 1987—Decided June 1, 1987 Title 28 U. S. C. § 1295(a)(2) gives the United States Court of Appeals for the Federal Circuit exclusive appellate jurisdiction over a variety of cases involving the Federal Government in which the District Court’s jurisdiction was based, “in whole or in part,” on 28 U. S. C. § 1346(a)(2), the Little Tucker Act, “except that jurisdiction of an appeal in a case brought in a district court under [the Federal Tort Claims Act (FTCA)] . . . shall be governed by” provisions vesting jurisdiction in the regional Federal Courts of Appeals. Respondents, a Japanese-American organization and individuals, brought suit in District Court seeking damages and declaratory relief for the tangible and intangible injuries suffered when, during World War II, the Federal Government removed approximately 120,000 Japanese-Americans from their homes and placed them in internment camps. Jurisdiction was based on the Little Tucker Act and the FTCA. The District Court concluded that all claims were barred, but the Court of Appeals reversed the dismissal of certain Little Tucker Act claims. The court held that it, rather than the Federal Circuit, had jurisdiction over the appeal. Although noting that § 1295(a)(2) generally grants the Federal Circuit exclusive jurisdiction of appeals in cases involving nontax Little Tucker Act claims, the court concluded that Congress did not intend the Federal Circuit to hear such appeals when they also included FTCA claims. Held: The Federal Circuit rather than the appropriate regional court of appeals has jurisdiction over an appeal from a district court’s decision of a “mixed” case raising both a nontax Little Tucker Act claim and an FTCA claim. Pp. 68-76. (a) Section 1295(a)(2) clearly establishes that the Federal Circuit has exclusive appellate jurisdiction of a case raising only a nontax Little Tucker Act claim, and that the appropriate regional court of appeals has exclusive appellate jurisdiction of a case raising only an FTCA claim. However, § 1295(a)(2)’s language does not clearly address a “mixed” case and is thus inherently ambiguous on this point. Pp. 68-69. (b) Given this ambiguity, the more plausible reading of § 1295(a)(2) is the Solicitor General’s view that the section’s “except” clause merely describes claims that do not suffice to create Federal Circuit jurisdiction, and that, thus, such claims must be heard in that court if they are joined UNITED STATES v. HOHRI 65 64 Syllabus with claims that fall within its exclusive jurisdiction. The proximity of the except clause to the “granting” clause at the beginning of § 1295(a)(2) suggests that the except clause’s failure to repeat the granting clause’s “in whole or in part” phrase in characterizing FTCA claims was not accidental. Moreover, the except clause’s description of the excepted tax cases by reference to the basis of “the claim” suggests that the clause was directed at cases raising one rather than multiple claims. Respondents’ contention that the except clause indicates not only that FTCA claims fail to create Federal Circuit jurisdiction, but also that the presence of such a claim renders inapplicable that court’s otherwise exclusive jurisdiction over nontax Little Tucker Act claims, is not persuasive. Although it has some force, respondents’ argument, which ultimately is based on a comparison of the language of the except clauses in §§ 1295(a)(2) and 1295(a)(1), is more attenuated than the Solicitor General’s view that rests simply on the variation between § 1295(a)(2)’s own clauses. Pp. 69-71. (c) Given the comprehensive statutory framework, under which the Federal Circuit has exclusive jurisdiction over every appeal from a Tucker Act or nontax Little Tucker Act claim, and the legislative history’s strong expressions of the need for judicial uniformity in this area, it seems likely that Congress would have rendered explicit any intended exceptions. Pp. 71-73. (d) Also unpersuasive is respondents’ argument that a congressional intent to deprive the Federal Circuit in “mixed” cases of its exclusive jurisdiction over nontax Little Tucker Act claims is evidenced by a congressional Report statement that FTCA appeals, because they frequently involve application of state law, would continue to be brought in the regional courts of appeals. When viewed as a whole, the legislative history establishes that Congress intended for centralized determination of nontax Little Tucker Act claims to predominate over regional adjudication of FTCA claims. Pp. 73-76. 251 U. S. App. D. C. 145, 782 F. 2d 227, vacated and remanded. Powell, J., delivered the opinion of the Court, in which all other Members joined, except Scalia, J., who took no part in the consideration or decision of the case. Blackmun, J., filed a concurring opinion, post, p. 76. Solicitor General Fried argued the cause for the United States. With him on the briefs were Assistant Attorney General Willard, Deputy Solicitor General Ayer, Roy T. Englert, Jr., Barbara L. Herwig, and Jay S. Bybee. 66 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. Benjamin L. Zelenko argued the cause for respondents. With him on the brief were Wallace M. Cohen, B. Michael Rauh, Ellen Godbey Carson, and Martin Shulman.* Justice Powell delivered the opinion of the Court. In this case we must decide which court—the Court of Appeals for the Federal Circuit or the appropriate regional Court of Appeals—has jurisdiction over an appeal from a Federal District Court’s decision of a case raising both a nontax claim under the Little Tucker Act and a claim under the Federal Tort Claims Act (FTCA). I During World War II, the Government of the United States removed approximately 120,000 Japanese-Americans from their homes and placed them in internment camps. Respondents are an organization of Japanese-Americans and 19 individuals—former internees and their representatives. They filed this action in the United States District Court for the District of Columbia, seeking damages and declaratory relief for the tangible and intangible injuries suffered because of this incident. Jurisdiction was based on, inter alia, the Little Tucker Act, 28 U. S. C. § 1346(a)(2),1 and the FTCA, * Briefs of amici curiae urging affirmance were filed for the State of California et al. by John K. Van de Kamp, Attorney General of California, Warren Price III, Attorney General of Hawaii, Corinne K. A. Watanabe, First Deputy Attorney General, and Steven S. Michaels, Deputy Attorney General; and for the American Friends Service Committee et al. by David Kairys. Briefs of amici curiae were filed for the American Civil Liberties Union et al. by Fred Okrand, Paul L. Hoffman, and Douglas E. Mirell; and for Fred Korematsu et al. by Dale Minami, Peggy Nagae, and Ruti G. Teitel. Jurisdiction in district courts under the Little Tucker Act is limited to nontort claims not exceeding $10,000. 28 U. S. C. § 1346(a)(2). See 14 C. Wright, A. Miller, & E. Cooper, Federal Practice and Procedure § 3657, pp. 284-288 (2d ed. 1985). This decentralized jurisdiction was designed to “ ‘give all persons having claims for comparatively small amounts the right to bring suits in the districts where they and their witnesses reside without subjecting them to the expense and annoyance of litigating in Washing UNITED STATES v. HOHRI 67 64 Opinion of the Court 28 U. S. C. § 1346(b). The District Court concluded that all claims were barred either by sovereign immunity or the applicable statute of limitations. 586 F. Supp. 769 (1984). Respondents appealed to the Court of Appeals for the District of Columbia Circuit. That court reversed the District Court’s dismissal of certain claims under the Little Tucker Act. 251 U. S. App. D. C. 145, 782 F. 2d 227 (1986). First, the court concluded that it, rather than the Court of Appeals for the Federal Circuit, had jurisdiction over the appeal. It noted that 28 U. S. C. § 1295(a)(2) generally grants the Federal Circuit exclusive jurisdiction of appeals in cases involving nontax claims under the Little Tucker Act. But it concluded that Congress did not intend the Federal Circuit to hear appeals of such cases when they also included FTCA claims. Id., at 157-158, 782 F. 2d, at 239-241. On the merits, the court concluded that the statute of limitations did not begin to run on certain of respondents’ Little Tucker Act claims until 1980, when Congress created the Commission on Wartime Relocation and Internment of Civilians. Id., at 171, 782 F. 2d, at 253. Chief Judge Markey, sitting by designation pursuant to 28 U. S. C. § 291(b), filed a dissent, disagreeing with the court’s jurisdictional analysis as well as its decision as to the statute of limitations. Id., at 174-175, 782 F. 2d, at 256-263. A petition for rehearing en banc was denied by a 6-to-5 vote. 253 U. S. App. D. C. 233, 793 F. 2d 304 (1986). Judge Bork, joined by four other judges, filed a dissent from denial of the petition, in which he disagreed with both of the court’s conclusions. Id., at 233-234, 793 F. 2d, at 304-313. ton,’” Id., at 274 (quoting United States v. King, 119 F. Supp. 398, 403 (Alaska 1954)). With minor exceptions, the Tucker Act grants the United States Claims Court jurisdiction of similar claims without regard to the amount of the claim. 28 U. S. C. § 1491(a)(1). Thus, Tucker Act claims for more than $10,000 can be brought only in the United States Claims Court. Claims for less than $10,000 generally can be brought either in a federal district court or in the United States Claims Court. 68 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. Because of the potentially broad impact of the Court of Appeals’ decision and because of the importance of the jurisdictional question, we granted the Government’s petition for a writ of certiorari. 479 U. S. 960 (1986). We conclude that the Court of Appeals did not have jurisdiction and therefore do not address the merits of its decision.2 II In 1982, Congress passed the Federal Courts Improvement Act, creating the United States Court of Appeals for the Federal Circuit. Among other things, the Act grants the Federal Circuit exclusive appellate jurisdiction over a variety of cases involving the Federal Government. 28 U. S. C. § 1295(a)(2). Specifically, the Act provides: “The United States Court of Appeals for the Federal Circuit shall have exclusive jurisdiction— “(2) of an appeal from a final decision of a district court of the United States ... if the jurisdiction of that court was based, in whole or in part, on section 1346 of this title, except that jurisdiction of an appeal in a case brought in a district court under section 1346(a)(1), 1346(b), 1346(e), or 1346(f) of this title or under section 1346(a)(2) when the claim is [related to federal taxes] shall be governed by sections 1291, 1292, and 1294 of this title.” This section establishes two undisputed propositions relevant to this case. First, the Federal Circuit has exclusive appellate jurisdiction of a case raising only a nontax claim under the Little Tucker Act, § 1346(a)(2). Second, the appropriate regional Court of Appeals—in this case, the Court of Appeals 2 Respondents also filed a petition for a writ of certiorari, No. 86-298, seeking review of other aspects of the Court of Appeals’ judgment. Because our resolution of No. 86-510 requires us to vacate the entire judgment of the Court of Appeals, we grant the petition in No. 86-298 and remand the entire case to the Court of Appeals. UNITED STATES v. HOHRI 69 64 Opinion of the Court for the District of Columbia Circuit—has exclusive appellate jurisdiction under §§ 1291, 1292, and 1294 of a case raising only a claim under the FTCA, § 1346(b). This case presents claims under both the Little Tucker Act and the FTCA, a situation not specifically addressed by § 1295(a)(2). Resolution of this problem turns on interpretation of the second clause of this subsection, the so-called “except clause.” The Solicitor General contends that the except clause merely describes claims that do not suffice to create jurisdiction in the Federal Circuit. Thus, he argues, appeals of FTCA claims must be heard in the Federal Circuit if, as in this case, they are joined with claims that fall within its exclusive jurisdiction. By contrast, respondents contend that the except clause indicates not only that FTCA claims fail to create jurisdiction in the Federal Circuit, but also that the presence of an FTCA claim renders inapplicable the Federal Circuit’s otherwise exclusive jurisdiction over nontax Little Tucker Act claims.3 A As always, the “‘starting point in every case involving construction of a statute is the language itself.’” Kelly n. Robinson, 479 U. S. 36, 43 (1986) (quoting Blue Chip Stamps v. Manor Drug Stores, 421 U. S. 723, 756 (1975) (Powell, J., concurring)). Unfortunately, as we have noted, see 3 Neither the parties nor any judge of the Court of Appeals suggested bifurcating the case so that the Little Tucker Act claims would be transferred to the Federal Circuit and the FTCA claims would remain in the Court of Appeals for the District of Columbia Circuit. We agree that bifurcation is inappropriate. The language of § 1295(a)(2) discusses jurisdiction over an appeal “in a case,” not over an appeal from decision of “a claim.” This strongly suggests that appeals of different parts of a single case should not go to different courts. Also, at least when a case has not been bifurcated in the district court, a bifurcated appeal of the different legal claims raised in any one case would result in an inefficient commitment of the limited resources of the federal appellate courts. Cf. Atari, Inc. v. JS & A Group, Inc., 747 F. 2d 1422, 1438-1440 (CA Fed. 1984) (en banc) (rejecting bifurcated appeals in patent cases). 70 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. United States v. Mottaz, 476 U. S. 834, 848-849, n. 11 (1986), the language of this statute does not clearly address a “mixed” case that presents both nontax Little Tucker Act claims and FTCA claims. Congress could have expressed the Solicitor General’s interpretation more clearly by adding the word “solely” to the except clause, and thus provided that an appeal from a case brought solely under § 1346(b) should be to the regional court of appeals. Or, if Congress had intended the broader meaning of the except clause urged by respondents, it could have added a phrase akin to “in whole or in part” to the except clause, thus providing that an appeal of a case brought in whole or in part under § 1346(b) should be to the regional court of appeals. Because Congress employed neither of these alternatives, we are left with the task of determining the more plausible interpretation of the language Congress did include in § 1295(a)(2). In our view, the Solicitor General’s reading of the statute is more natural. Although Congress included the phrase “in whole or in part” in the granting clause at the beginning of § 1295(a)(2), it did not repeat this phrase in the except clause later in the same paragraph. The proximity of the clauses suggests that the variation in wording was not accidental. Also, in one instance the statute describes the excepted cases by reference to the basis of “the claim.” See § 1295(a)(2) (providing for appeals to regional courts of appeals “in a case brought in a district court. . . under section 1346(a)(2) when the claim is [related to federal taxes]” (emphasis added)). This suggests that the except clause was directed at cases raising only one claim; it strains the language to apply the except clause to cases raising multiple claims, some within and some not within the except clause. Respondents rely heavily on the wording of the preceding paragraph of the statute, § 1295(a)(1). That section provides exclusive appellate jurisdiction in the Federal Circuit “of an appeal from a final decision of a district court of the United States ... if the jurisdiction of that UNITED STATES v. HOHRI 71 64 Opinion of the Court court was based, in whole or in part, on section 1338 of this title, except that a case involving a claim arising under any Act of Congress relating to copyrights or trademarks and no other claims under section 1338(a) shall be [appealed to the appropriate regional court of appeals].” Like subsection (a)(2), subsection (a)(1) grants the Federal Circuit exclusive jurisdiction over cases arising under one section of the Judicial Code, but has an except clause governing certain types of cases under that section. Respondents note that the (a)(1) except clause specifically deals with mixed cases. On its face it applies only to cases raising the excepted claims “and no other claims.” They argue that the failure to include the words “and no other claims” in the subsection (a)(2) except clause indicates that Congress intended a different scope for the two except clauses. Thus, in their view, the Solicitor General’s interpretation requires us to read into (a)(2) the words that Congress intentionally omitted. Although respondents’ argument has some force, ultimately we are not persuaded. Neither of the proffered readings can remove the ambiguity inherent in this statute. Respondents’ textual argument—based on a comparison of the language of § 1295(a)(1) with the language of §1295 (a)(2)—is more attenuated than the Solicitor General’s textual argument, that rests on the variation between the clauses of subsection (a)(2) itself. While a more carefully drafted statute would have avoided both of these problems, we find it difficult to assume that the variation within the same subsection was inadvertent. B Because the statute is ambiguous, congressional intent is particularly relevant to our decision. A motivating concern of Congress in creating the Federal Circuit was the “special need for nationwide uniformity” in certain areas of the law. 72 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. S. Rep. No. 97-275, p. 2 (1981) (hereinafter 1981 Senate Report); S. Rep. No. 96-304, p. 8 (1979) (hereinafter 1979 Senate Report). The Senate Reports explained: “[T]here are areas of the law in which the appellate courts reach inconsistent decisions on the same issue, or in which—although the rule of law may be fairly clear—courts apply the law unevenly when faced with the facts of individual cases.” 1981 Senate Report, at 3; 1979 Senate Report, at 9. The Federal Circuit was designed to provide “a prompt, definitive answer to legal questions” in these areas. 1981 Senate Report, at 1; 1979 Senate Report, at 1. Nontort claims against the Federal Government present one of the principal areas in which Congress sought such uniformity. Thus, Congress decided to confer jurisdiction on the Federal Circuit in “all federal contract appeals in which the United States is a defendant.” H. R. Rep. No. 97-312, p. 18 (1981) (hereinafter 1981 House Report); H. R. Rep. No. 96-1300, p. 16 (1980) (hereinafter 1980 House Report).4 For the most part, the statute unambiguously effectuates this goal. Tucker Act claims for more than $10,000 may be brought only in the United States Claims Court. 28 U. S. C. § 1491(a)(1). Decisions of the United States Claims Court are appealable only to the Federal Circuit, not the regional courts of appeals. § 1295(a)(3). Claims for less than $10,000 (i. e., Little Tucker Act claims) may be brought either in a federal district court or in the United States Claims Court. § 1346(a)(2). These claims, so long as they are not related to federal taxes, also are appealable only to the Federal Circuit. §§ 1295(a)(2), (3). A conspicuous fea 4 The Little Tucker Act, of course, covers not only contract claims, but also other claims for money damages “founded either upon the Constitution, or any Act of Congress, or any regulation of an executive department, ... or for liquidated or unliquidated damages in cases not sounding in tort.” 28 U. S. C. § 1346(a)(2). See Eastport S.S. Corp. v. United States, 178 Ct. Cl. 599, 605-606, 372 F. 2d 1002, 1007-1008 (1967) (discussing noncontractual liability under the Tucker Act). UNITED STATES v. HOHRI 73 64 Opinion of the Court ture of these judicial arrangements is the creation of exclusive Federal Circuit jurisdiction over every appeal from a Tucker Act or nontax Little Tucker Act claim. Given this comprehensive framework and the strong expressions of the need for uniformity in the area, one would expect any exception intended by Congress to have been made explicit, rather than left to inferences drawn from loose language. C Despite the language of the statute and the evident congressional desire for uniform adjudication of Little Tucker Act claims, the Court of Appeals inferred an exception to exclusive Federal Circuit jurisdiction in cases that include FTC A claims. In supporting the Court of Appeals’ judgment, respondents rely on the statement, thrice repeated in the congressional Reports, that “[b]ecause cases brought under the Federal Tort Claims Act frequently involve the application of State law, those appeals will continue to be brought to the regional courts of appeals.” 1981 Senate Report, at 20; 1981 House Report, at 42; 1980 House Report, at 34. Respondents argue that this statement evidences a congressional intent to deprive the Federal Circuit of its otherwise exclusive appellate jurisdiction over nontax Little Tucker Act claims whenever an FTCA claim is presented in the same case. We find this argument unpersuasive when viewed in the context of the legislative history as a whole. First, the congressional Reports indicate only that Congress saw no affirmative need for national uniformity in FTCA cases, not that the perceived need for regional adjudication of FTCA claims outweighed the strong and oft-noted intent of Congress that only the Federal Circuit should have jurisdiction of appeals in nontax Little Tucker Act cases. Second, Congress specifically rejected the idea that patent and Tucker Act appeals should be decided by a “specialized 74 482 U. S. OCTOBER TERM, 1986 Opinion of the Court court” incapable of deciding more general legal issues.6 The Federal Circuit decides questions arising under the Federal Constitution and statutes whenever such questions arise in cases within the Federal Circuit’s jurisdiction. There is no reason to believe that Congress intended to exempt the relatively common tort questions presented by the average FTCA cases from the Federal Circuit’s already-broad docket.6 Third, if Congress thought the presence of statelaw issues was sufficient to override the need for centralization of nontax Little Tucker Act claims, the except clause logically should have included all cases raising state-law The Senate Reports described “such an approach as being inconsistent with the imperative of avoiding undue specialization within the Federal judicial system.” 1981 Senate Report, at 6; 1979 Senate Report, at 13. They also proffered a broad conception of the Federal Circuit’s jurisdiction: “[I]t will have a varied docket spanning a broad range of legal issues and types of cases. It will handle all patent appeals, plus government claims case[s] and all other appellate matters that are now considered by the [Court of Customs and Patent Appeals] or the Court of Claims—cases which contain a wide variety of issues. “This rich docket assures that the work of the proposed court will be broad and diverse and not narrowly specialized. The judges will have no lack of exposure to a broad variety of legal problems. Moreover, the subject matter of the new court will be sufficiently mixed to prevent any special interest from dominating it.” 1981 Senate Report, at 6; 1979 Senate Report, at 13. See also 1981 House Report, at 19; 1980 House Report, at 17. 6 There may have been a concern that Federal Circuit judges would not be familiar with questions of state tort law. But this problem is mitigated considerably by the fact that these cases are tried before local federal district judges, who are likely to be familiar with the applicable state law. Indeed, a district judge’s determination of a state-law question usually is reviewed with great deference. E. g., Railroad Comm’n v. Pullman Co., 312 U. S. 496, 499 (1941) (review by this Court); Alabama Elec. Cooperative, Inc. v. First National Bank, 684 F. 2d 789, 792 (CA11 1982) (review by a Court of Appeals). But see In re McLinn, 739 F. 2d 1395, 1400 (CA9 1984) (en banc) (de novo review by a Court of Appeals). It is certainly not clear that a panel of the Federal Circuit would be less competent to review such determinations than a panel of a regional court of appeals. UNITED STATES v. HOHRI 75 64 Opinion of the Court issues, not just cases under the FTCA.7 But Congress did not preclude the Federal Circuit from deciding all statelaw questions, or even all FTCA claims. For example, even under respondents’ reading of the statute, the Federal Circuit can hear FTCA claims whenever they are joined with patent claims under § 1338. See § 1295(a)(1). That Congress allowed the Federal Circuit to hear FTCA claims in this context refutes respondents’ contention that Congress demanded regional adjudication of all FTCA claims. For these reasons, we conclude that Congress intended for centralized determination of nontax Little Tucker Act claims to predominate over regional adjudication of FTCA claims. We hold that a mixed case, presenting both a nontax Little 7 The legislative development of the other claims in the except clause also supports our interpretation. As enacted, the except clause lists five types of claims. Its original version listed only two types of claims, FTCA claims and claims under § 1346(a)(1). S. 677, 96th Cong., 1st Sess., § 735(a) (1979). Under that bill, the Federal Circuit would have had exclusive jurisdiction (in addition to the jurisdiction Congress eventually gave it) over claims under §§ 1346(e) and (f), as well as tax-related Little Tucker Act claims. Subsequent amendments added these three types of claims to the except clause. Under respondents’ theory, the original bill must have reflected a view that there was an affirmative need for centralization of these claims; subsequent amendments reflected a complete reversal of viewpoint to a decision that there was an affirmative harm in centralized adjudication of these claims. Thus, respondents implicitly argue that Congress originally thought that all of these claims should have been appealed to the Federal Circuit, but eventually determined not only that these claims should be appealed to the regional circuits, but also that the need for regional appeals of those claims would render inapplicable the Federal Circuit’s otherwise exclusive jurisdiction over nontax Little Tucker Act claims. We think such a reversal of intent would have been reflected somewhere in the legislative history. Our interpretation does not posit such a sharp and undocumented swing of viewpoint. In our view, Congress originally thought that appeals of these claims should be centralized. Subsequently, Congress decided that they could be adjudicated adequately in the regional courts of appeals. Nothing suggests, however, that Congress thought regional adjudication was so important as to bar centralized adjudication of mixed cases. 76 OCTOBER TERM, 1986 Blackmun, J., concurring 482 U. S. Tucker Act claim and an FTCA claim, may be appealed only to the Federal Circuit. Ill We vacate the judgment of the Court of Appeals, and remand the case to that court, with instructions to transfer the case to the Federal Circuit. See 28 U. S. C. § 1631. It is so ordered. Justice Scalia took no part in the consideration or decision of this case. Justice Blackmun, concurring. I join the Court’s opinion and its judgment. I do so, however, with less than full assurance and satisfaction. There are three reasons for my concern. The first is the consequent element of further delay in the decision on the merits in a case that has roots already more than four decades old. The issue on the merits probably will be back in this Court once again months or years hence. The second is that the statute the Court is forced to construe in this case is not a model of legislative craftsmanship. Surely, Congress is able to make its intent more evident than in the language it has utilized here. It is to be hoped that Congress will look at the problem it has created and will set forth in precise terms its conclusion as to jurisdiction of federal appellate courts in mixed-claims cases of this kind. My third reason is an administrative one. I am somewhat surprised and concerned over the fact that the Chief Judge of the Federal Circuit was designated to sit on this appeal. The jurisdictional issue, on which the case presently goes off, involves the jurisdiction of his own court as against that of the District of Columbia Circuit. In concluding to dissent, as he had every right to do—and as the Court today vindicates —the Chief Judge was forced to take a position favoring his own court’s jurisdiction. The “appearance” is troubling. I wonder why what must have been a measure of embarrass- UNITED STATES v. HOHRI 77 64 Blackmun, J., concurring ment for the Chief Judge was not avoided by refraining to assign him, or any other judge from the “opposite” court, to sit on this case. Unless the designation was purposeful (in order to have a panel with views of judges of both courts), one must observe that the Court of Appeals for the District of Columbia Circuit had a complement of other judges from which to fill the third seat on the three-judge panel. 78 OCTOBER TERM, 1986 Syllabus 482 U. S. TURNER et al. v. SAFLEY et al. CERTIORARI TO THE UNITED STATES COURT OF APPEALS FOR THE EIGHTH CIRCUIT No. 85-1384. Argued January 13, 1987—Decided June 1, 1987 Respondent inmates brought a class action challenging two regulations promulgated by the Missouri Division of Corrections. The first permits correspondence between immediate family members who are inmates at different institutions within the Division’s jurisdiction, and between inmates “concerning legal matters,” but allows other inmate correspondence only if each inmate’s classification/treatment team deems it in the best interests of the parties. The second regulation permits an inmate to marry only with the prison superintendent’s permission, which can be given only when there are “compelling reasons” to do so. Testimony indicated that generally only a pregnancy or the birth of an illegitimate child would be considered “compelling.” The Federal District Court found both regulations unconstitutional, and the Court of Appeals affirmed. Held: 1. The lower courts erred in ruling that Procunier v. Martinez, 416 U. S. 396, and its progeny require the application of a strict scrutiny standard of review for resolving respondents’ constitutional complaints. Rather, those cases indicate that a lesser standard is appropriate whereby inquiry is made into whether a prison regulation that impinges on inmates’ constitutional rights is “reasonably related” to legitimate penological interests. In determining reasonableness, relevant factors include (a) whether there is a “valid, rational connection” between the regulation and a legitimate and neutral governmental interest put forward to justify it, which connection cannot be so remote as to render the regulation arbitrary or irrational; (b) whether there are alternative means of exercising the asserted constitutional right that remain open to inmates, which alternatives, if they exist, will require a measure of judicial deference to the corrections officials’ expertise; (c) whether and the extent to which accommodation of the asserted right will have an impact on prison staff, on inmates’ liberty, and on the allocation of limited prison resources, which impact, if substantial, will require particular deference to corrections officials; and (d) whether the regulation represents an “exaggerated response” to prison concerns, the existence of a ready alternative that fully accommodates the prisoner’s rights at de minimis TURNER v. SAFLEY 79 78 Syllabus costs to valid penological interests being evidence of unreasonableness. Pp. 84-91. 2. The Missouri inmate correspondence regulation is, on the record here, reasonable and facially valid. The regulation is logically related to the legitimate security concerns of prison officials, who testified that mail between prisons can be used to communicate escape plans, to arrange violent acts, and to foster prison gang activity. Moreover, the regulation does not deprive prisoners of all means of expression, but simply bars communication with a limited class of people—other inmates — with whom authorities have particular cause to be concerned. The regulation is entitled to deference on the basis of the significant impact of prison correspondence on the liberty and safety of other prisoners and prison personnel, in light of officials’ testimony that such correspondence facilitates the development of informal organizations that threaten safety and security at penal institutions. Nor is there an obvious, easy alternative to the regulation, since monitoring inmate correspondence clearly would impose more than a de minimis cost in terms of the burden on staff resources required to conduct item-by-item censorship, and would create an appreciable risk of missing dangerous communications. The regulation is content neutral and does not unconstitutionally abridge the First Amendment rights of prison inmates. Pp. 91-93. 3. The constitutional right of prisoners to marry is impermissibly burdened by the Missouri marriage regulation. Pp. 94-99. (a) Prisoners have a constitutionally protected right to marry under Zablocki v. Redhail, 434 U. S. 374. Although such a marriage is subject to substantial restrictions as a result of incarceration, sufficient important attributes of marriage remain to form a constitutionally protected relationship. Butler v. Wilson, 415 U'. S. 953, distinguished. Pp. 94-96. (b) The regulation is facially invalid under the reasonable relationship test. Although prison officials may regulate the time and circumstances under which a marriage takes place, and may require prior approval by the warden, the almost complete ban on marriages here is not, on the record, reasonably related to legitimate penological objectives. The contention that the regulation serves security concerns by preventing “love triangles” that may lead to violent inmate confrontations is without merit, since inmate rivalries are likely to develop with or without a formal marriage ceremony. Moreover, the regulation’s broad prohibition is not justified by the security of fellow inmates and prison staff, who are not affected where the inmate makes the private decision to marry a civilian. Rather, the regulation represents an exaggerated response to the claimed security objectives, since allowing marriages unless the warden finds a threat to security, order, or the public safety represents 80 OCTOBER TERM, 1986 Syllabus 482 U. S. an obvious, easy alternative that would accommodate the right to marry while imposing a de minimis burden. Nor is the regulation reasonably related to the articulated rehabilitation goal of fostering self-reliance by female prisoners. In requiring refusal of permission to marry to all inmates absent a compelling reason, the regulation sweeps much more broadly than is necessary, in light of officials’ testimony that male inmates’ marriages had generally caused them no problems and that they had no objections to prisoners marrying civilians. Pp. 96-99. 777 F. 2d 1307, affirmed in part, reversed in part, and remanded. O’Connor, J., delivered the opinion of the Court, in which Rehnquist, C. J., and White, Powell, and Scalia, JJ., joined, and in Part III-B of which Brennan, Marshall, Blackmun, and Stevens, JJ., joined. Stevens, J., filed an opinion concurring in part and dissenting in part, in which Brennan, Marshall, and Blackmun, JJ., joined, post, p. 100. Henry T. Herschel, Assistant Attorney General of Missouri, argued the cause for petitioners. With him on the briefs were William L. Webster, Attorney General, and Michael L. Boicourt. Floyd R. Finch, Jr., argued the cause and filed a brief for respondents. * *Briefs of amici curiae urging reversal were filed for the United States by Solicitor General Fried, Assistant Attorney General Trott, Deputy Solicitor General Cohen, and Roger Clegg; and for the State of Arkansas et al. by Thomas J. Miller, Attorney General of Iowa, Brent R. Appel, Deputy Attorney General, John Steven Clark, Attorney General of Arkansas, John K. Van de Kamp, Attorney General of California, Lacy Il-Thornburg, Attorney General of North Carolina, Nicholas Spaeth, Attorney General of North Dakota, T. Travis Medlock, Attorney General of South Carolina, Mark V. Meierhenry, Attorney General of South Dakota, Gerald L. Baliles, Attorney General of Virginia, and Robert M. Spire, Attorney General of Nebraska. Briefs of amici curiae urging affirmance were filed for the Correctional Association of New York by John H. Hall and Steven Klugman; for Prisoners’ Legal Services of New York, Inc., et al. by Robert Selcov; and for Guadalupe Guajardo, Jr., et al. by Harry M. Reasoner and Ann Lents. Jim Mattox, Attorney General of Texas, Mary F. Keller, Executive Assistant Attorney General, and F. Scott McCown and Michael F. Lynch, Assistant Attorneys General, filed a brief for the State of Texas as amicus curiae. TURNER v. SAFLEY 81 78 Opinion of the Court Justice O’Connor delivered the opinion of the Court. This case requires us to determine the constitutionality of regulations promulgated by the Missouri Division of Corrections relating to inmate marriages and inmate-to-inmate correspondence. The Court of Appeals for the Eighth Circuit, applying a strict scrutiny analysis, concluded that the regulations violate respondents’ constitutional rights. We hold that a lesser standard of scrutiny is appropriate in determining the constitutionality of the prison rules. Applying that standard, we uphold the validity of the correspondence regulation, but we conclude that the marriage restriction cannot be sustained. I Respondents brought this class action for injunctive relief and damages in the United States District Court for the Western District of Missouri. The regulations challenged in the complaint were in effect at all prisons within the jurisdiction of the Missouri Division of Corrections. This litigation focused, however, on practices at the Renz Correctional Institution (Renz), located in Cedar City, Missouri. The Renz prison population includes both male and female prisoners of varying security levels. Most of the female prisoners at Renz are classified as medium or maximum security inmates, while most of the male prisoners are classified as minimum security offenders. Renz is used on occasion to provide protective custody for inmates from other prisons in the Missouri system. The facility originally was built as a minimum security prison farm, and it still has a minimum security perimeter without guard towers or walls. Two regulations are at issue here. The first of the challenged regulations relates to correspondence between inmates at different institutions. It permits such correspondence “with immediate family members who are inmates in other correctional institutions,” and it permits correspondence between inmates “concerning legal matters.” Other correspondence between inmates, however, is permitted only 82 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. if “the classification/treatment team of each inmate deems it in the best interest of the parties involved.” App. 34. Trial testimony indicated that as a matter of practice, the determination whether to permit inmates to correspond was based on team members’ familiarity with the progress reports, conduct violations, and psychological reports in the inmates’ files rather than on individual review of each piece of mail. See 777 F. 2d 1307, 1308 (CA8 1985). At Renz, the District Court found that the rule “as practiced is that inmates may not write non-family inmates.” 586 F. Supp. 589, 591 (WD Mo. 1984). The challenged marriage regulation, which was promulgated while this litigation was pending, permits an inmate to marry only with the permission of the superintendent of the prison, and provides that such approval should be given only “when there are compelling reasons to do so.” App. 47. The term “compelling” is not defined, but prison officials testified at trial that generally only a pregnancy or the birth of an illegitimate child would be considered a compelling reason. See 586 F. Supp., at 592. Prior to the promulgation of this rule, the applicable regulation did not obligate Missouri Division of Corrections officials to assist an inmate who wanted to get married, but it also did not specifically authorize the superintendent of an institution to prohibit inmates from getting married. Ibid. The District Court certified respondents as a class pursuant to Federal Rule of Civil Procedure 23. The class certified by the District Court includes “persons who either are or may be confined to the Renz Correctional Center and who desire to correspond with inmates at other Missouri correctional facilities.” It also encompasses a broader group of persons “who desire to . . . marry inmates of Missouri correctional institutions and whose rights of . . . marriage have been or will be violated by employees of the Missouri Division of Corrections.” See App. 21-22. TURNER v. SAFLEY 83 78 Opinion of the Court The District Court issued a memorandum opinion and order finding both the correspondence and marriage regulations unconstitutional. The court, relying on Procunier v. Martinez, 416 U. S. 396, 413-414 (1974), applied a strict scrutiny standard. It held the marriage regulation to be an unconstitutional infringement upon the fundamental right to marry because it was far more restrictive than was either reasonable or essential for the protection of the State’s interests in security and rehabilitation. 586 F. Supp., at 594. The correspondence regulation also was unnecessarily broad, the court concluded, because prison officials could effectively cope with the security problems raised by inmate-to-inmate correspondence through less restrictive means, such as scanning the mail of potentially troublesome inmates. Id., at 596. The District Court also held that the correspondence regulation had been applied in an arbitrary and capricious manner. The Court of Appeals for the Eighth Circuit affirmed. 777 F. 2d 1307 (1985). The Court of Appeals held that the District Court properly used strict scrutiny in evaluating the constitutionality of the Missouri correspondence and marriage regulations. Under Procunier v. Martinez, supra, the correspondence regulation could be justified “only if it furthers an important or substantial governmental interest unrelated to the suppression of expression, and the limitation is no greater than necessary or essential to protect that interest.” 777 F. 2d, at 1310. The correspondence regulation did not satisfy this standard because it was not the least restrictive means of achieving the security goals of the regulation. In the Court of Appeals’ view, prison officials could meet the problem of inmate conspiracies by exercising their authority to open and read all prisoner mail. Id., at 1315-1316. The Court of Appeals also concluded that the marriage rule was not the least restrictive means of achieving the asserted goals of rehabilitation and security. The goal of rehabilitation could be met through alternatives such 84 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. as counseling, and violent “love triangles” were as likely to occur without a formal marriage ceremony as with one. Ibid. Absent evidence that the relationship was or would become abusive, the connection between an inmate’s marriage and the subsequent commission of a crime was simply too tenuous to justify denial of this constitutional right. Id., at 1315. We granted certiorari, 476 U. S. 1139 (1986). II We begin, as did the courts below, with our decision in Procunier n. Martinez, supra, which described the principles that necessarily frame our analysis of prisoners’ constitutional claims. The first of these principles is that federal courts must take cognizance of the valid constitutional claims of prison inmates. Id., at 405. Prison walls do not form a barrier separating prison inmates from the protections of the Constitution. Hence, for example, prisoners retain the constitutional right to petition the government for the redress of grievances, Johnson n. Avery, 393 U. S. 483 (1969); they are protected against invidious racial discrimination by the Equal Protection Clause of the Fourteenth Amendment, Lee n. Washington, 390 U. S. 333 (1968); and they enjoy the protections of due process, Wolff n. McDonnell, 418 U. S. 539 (1974); Haines v. Kerner, 404 U. S. 519 (1972). Because prisoners retain these rights, “[w]hen a prison regulation or practice offends a fundamental constitutional guarantee, federal courts will discharge their duty to protect constitutional rights.” Procunier v. Martinez, 416 U. S., at 405-406. A second principle identified in Martinez, however, is the recognition that “courts are ill equipped to deal with the increasingly urgent problems of prison administration and reform.” Id., at 405. As the Martinez Court acknowledged, “the problems of prisons in America are complex and intractable, and, more to the point, they are not readily susceptible of resolution by decree. ” Id., at 404-405. Running a prison TURNER v. SAFLEY 85 78 Opinion of the Court is an inordinately difficult undertaking that requires expertise, planning, and the commitment of resources, all of which are peculiarly within the province of the legislative and executive branches of government. Prison administration is, moreover, a task that has been committed to the responsibility of those branches, and separation of powers concerns counsel a policy of judicial restraint. Where a state penal system is involved, federal courts have, as we indicated in Martinez, additional reason to accord deference to the appropriate prison authorities. See id., at 405. Our task, then, as we stated in Martinez, is to formulate a standard of review for prisoners’ constitutional claims that is responsive both to the “policy of judicial restraint regarding prisoner complaints and [to] the need to protect constitutional rights.” Id., at 406. As the Court of Appeals acknowledged, Martinez did not itself resolve the question that it framed. Martinez involved mail censorship regulations proscribing statements that “unduly complain,” “magnify grievances,” or express “inflammatory political, racial, religious or other views.” Id., at 415. In that case, the Court determined that the proper standard of review for prison restrictions on correspondence between prisoners and members of the general public could be decided without resolving the “broad questions of ‘prisoners’ rights.’” Id., at 408. The Martinez Court based its ruling striking down the content-based regulation on the First Amendment rights of those who are not prisoners, stating that “[w]hatever the status of a prisoner’s claim to uncensored correspondence with an outsider, it is plain that the latter’s interest is grounded in the First Amendment’s guarantee of freedom of speech.” Id., at 408. Our holding therefore turned on the fact that the challenged regulation caused a “consequential restriction on the First and Fourteenth Amendment rights of those who are not prisoners.” Id., at 409 (emphasis added). We expressly reserved the question of the proper standard of 86 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. review to apply in cases “involving questions of ‘prisoners’ rights.’” Ibid. In four cases following Martinez, this Court has addressed such “questions of ‘prisoners’ rights.’” The first of these, Pell v. Procunier, 417 U. S. 817 (1974), decided the same Term as Martinez, involved a constitutional challenge to a prison regulation prohibiting face-to-face media interviews with individual inmates. The Court rejected the inmates’ First Amendment challenge to the ban on media interviews, noting that judgments regarding prison security “are peculiarly within the province and professional expertise of corrections officials, and, in the absence of substantial evidence in the record to indicate that the officials have exaggerated their response to these considerations, courts should ordinarily defer to their expert judgment in such matters.” 417 U. S., at 827. The next case to consider a claim of prisoners’ rights was Jones n. North Carolina Prisoners’ Union, 433 U. S. 119 (1977). There the Court considered prison regulations that prohibited meetings of a “prisoners’ labor union,” inmate solicitation of other inmates to join the union, and bulk mailings concerning the union from outside sources. Noting that the lower court in Jones had “got[ten] off on the wrong foot. . . by not giving appropriate deference to the decisions of prison administrators and appropriate recognition to the peculiar and restrictive circumstances of penal confinement,” id., at 125, the Court determined that the First and Fourteenth Amendment rights of prisoners were “barely implicated” by the prohibition on bulk mailings, see id., at 130, and that the regulation was “reasonable” under the circumstances. The prisoners’ constitutional challenge to the union meeting and solicitation restrictions was also rejected, because “[t]he ban on inmate solicitation and group meetings . . . was rationally related to the reasonable, indeed to the central, objectives of prison administration.” Id., at 129. TURNER v. SAFLEY 87 78 Opinion of the Court Bell v. Wolfish, 441 U. S. 520 (1979), concerned a First Amendment challenge to a Bureau of Prisons rule restricting inmates’ receipt of hardback books unless mailed directly from publishers, book clubs, or bookstores. The rule was upheld as a “rational response” to a clear security problem. Id., at 550. Because there was “no evidence” that officials had exaggerated their response to the security problem, the Court held that “the considered judgment of these experts must control in the absence of prohibitions far more sweeping than those involved here.” Id., at 551. And in Block v. Rutherford, 468 U. S. 576 (1984), a ban on contact visits was upheld on the ground that “responsible, experienced administrators have determined, in their sound discretion, that such visits will jeopardize the security of the facility,” and the regulation was “reasonably related” to these security concerns. Id., at 589, 586. In none of these four “prisoners’ rights” cases did the Court apply a standard of heightened scrutiny, but instead inquired whether a prison regulation that burdens fundamental rights is “reasonably related” to legitimate penological objectives, or whether it represents an “exaggerated response” to those concerns. The Court of Appeals in this case nevertheless concluded that Martinez provided the closest analogy for determining the appropriate standard of review for resolving respondents’ constitutional complaints. The Court of Appeals distinguished this Court’s decisions in Pell, Jones, Bell, and Block as variously involving “time, place, or manner” regulations, or regulations that restrict “presumptively dangerous” inmate activities. See 777 F. 2d, at 1310-1312. The Court of Appeals acknowledged that Martinez had expressly reserved the question of the appropriate standard of review based on inmates’ constitutional claims, but it nonetheless believed that the Martinez standard was the proper one to apply to respondents’ constitutional claims. We disagree with the Court of Appeals that the reasoning in our cases subsequent to Martinez can be so narrowly 88 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. cabined. In Pell, for example, it was found “relevant” to the reasonableness of a restriction on face-to-face visits between prisoners and news reporters that prisoners had other means of communicating with members of the general public. See 417 U. S., at 823-824. These alternative means of communication did not, however, make the prison regulation a “time, place, or manner” restriction in any ordinary sense of the term. As Pell acknowledged, the alternative methods of personal communication still available to prisoners would have been “unimpressive” if offered to justify a restriction on personal communication among members of the general public. Id., at 825. Nevertheless, they were relevant in determining the scope of the burden placed by the regulation on inmates’ First Amendment rights. Pell thus simply teaches that it is appropriate to consider the extent of this burden when “we [are] called upon to balance First Amendment rights against [legitimate] governmental interests.” Id., at 824. Nor, in our view, can the reasonableness standard adopted in Jones and Bell be construed as applying only to “presumptively dangerous” inmate activities. To begin with, the Court of Appeals did not indicate how it would identify such “presumptively dangerous” conduct, other than to conclude that the group meetings in Jones, and the receipt of hardback books in Bell, both fall into that category. See 777 F. 2d, at 1311-1312. The Court of Appeals found that correspondence between inmates did not come within this grouping because the court did “not think a letter presents the same sort of ‘obvious security problem’ as does a hardback book.” Id., at 1312. It is not readily apparent, however, why hardback books, which can be scanned for contraband by electronic devices and fluoroscopes, see Bell v. Wolfish, supra, at 574 (Marshall, J., dissenting), are qualitatively different in this respect from inmate correspondence, which can be written in codes not readily subject to detection; or why coordinated inmate activity within the same prison is categorically different TURNER v. SAFLEY 89 78 Opinion of the Court from inmate activity coordinated by mail among different prison institutions. The determination that an activity is “presumptively dangerous” appears simply to be a conclusion about the reasonableness of the prison restriction in light of the articulated security concerns. It therefore provides a tenuous basis for creating a hierarchy of standards of review. If Pell, Jones, and Bell have not already resolved the question posed in Martinez, we resolve it now: when a prison regulation impinges on inmates’ constitutional rights, the regulation is valid if it is reasonably related to legitimate penological interests. In our view, such a standard is necessary if “prison administrators . . . , and not the courts, [are] to make the difficult judgments concerning institutional operations.” Jones n. North Carolina Prisoners’ Union, 433 U. S., at 128. Subjecting the day-to-day judgments of prison officials to an inflexible strict scrutiny analysis would seriously hamper their ability to anticipate security problems and to adopt innovative solutions to the intractable problems of prison administration. The rule would also distort the decisionmaking process, for every administrative judgment would be subject to the possibility that some court somewhere would conclude that it had a less restrictive way of solving the problem at hand. Courts inevitably would become the primary arbiters of what constitutes the best solution to every administrative problem, thereby “unnecessarily perpetuat[ing] the involvement of the federal courts in affairs of prison administration.” Procunier v. Martinez, 416 U. S., at 407. As our opinions in Pell, Bell, and Jones show, several factors are relevant in determining the reasonableness of the regulation at issue. First, there must be a “valid, rational connection” between the prison regulation and the legitimate governmental interest put forward to justify it. Block v. Rutherford, supra, at 586. Thus, a regulation cannot be sustained where the logical connection between the regulation and the asserted goal is so remote as to render the policy 90 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. arbitrary or irrational. Moreover, the governmental objective must be a legitimate and neutral one. We have found it important to inquire whether prison regulations restricting inmates’ First Amendment rights operated in a neutral fashion, without regard to the content of the expression. See Pell v. Procunier, 417 U. S., at 828; Bell n. Wolfish, 441 U. S., at 551. A second factor relevant in determining the reasonableness of a prison restriction, as Pell shows, is whether there are alternative means of exercising the right that remain open to prison inmates. Where “other avenues” remain available for the exercise of the asserted right, see Jones v. North Carolina Prisoners’ Union, supra, at 131, courts should be particularly conscious of the “measure of judicial deference owed to corrections officials ... in gauging the validity of the regulation.” Pell v. Procunier, supra, at 827. A third consideration is the impact accommodation of the asserted constitutional right will have on guards and other inmates, and on the allocation of prison resources generally. In the necessarily closed environment of the correctional institution, few changes will have no ramifications on the liberty of others or on the use of the prison’s limited resources for preserving institutional order. When accommodation of an asserted right will have a significant “ripple effect” on fellow inmates or on prison staff, courts should be particularly deferential to the informed discretion of corrections officials. Cf. Jones v. North Carolina Prisoners’ Union, supra, at 132-133. Finally, the absence of ready alternatives is evidence of the reasonableness of a prison regulation. See Block v. Rutherford, 468 U. S., at 587. By the same token, the existence of obvious, easy alternatives may be evidence that the regulation is not reasonable, but is an “exaggerated response” to prison concerns. This is not a “least restrictive alternative” test: prison officials do not have to set up and then shoot down every conceivable alternative method of accommodat TURNER v. SAFLEY 91 78 Opinion of the Court ing the claimant’s constitutional complaint. See ibid. But if an inmate claimant can point to an alternative that fully accommodates the prisoner’s rights at de minimis cost to valid penological interests, a court may consider that as evidence that the regulation does not satisfy the reasonable relationship standard. Ill Applying our analysis to the Missouri rule barring inmate-to-inmate correspondence, we conclude that the record clearly demonstrates that the regulation was reasonably related to legitimate security interests. We find that the marriage restriction, however, does not satisfy the reasonable relationship standard, but rather constitutes an exaggerated response to petitioners’ rehabilitation and security concerns. A According to the testimony at trial, the Missouri correspondence provision was promulgated primarily for security reasons. Prison officials testified that mail between institutions can be used to communicate escape plans and to arrange assaults and other violent acts. 2 Tr. 76; 4 id., at 225-228. Witnesses stated that the Missouri Division of Corrections had a growing problem with prison gangs, and that restricting communications among gang members, both by transferring gang members to different institutions and by restricting their correspondence, was an important element in combating this problem. 2 id., at 75-77; 3 id., at 266-267; 4 id., at 226. Officials also testified that the use of Renz as a facility to provide protective custody for certain inmates could be compromised by permitting correspondence between inmates at Renz and inmates at other correctional institutions. 3 id., at 264-265. The prohibition on correspondence between institutions is logically connected to these legitimate security concerns. Undoubtedly, communication with other felons is a potential spur to criminal behavior: this sort of contact frequently is 92 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. prohibited even after an inmate has been released on parole. See, e. g., 28 CFR §2.40(a)(10) (1986) (federal parole conditioned on nonassociation with known criminals, unless permission is granted by the parole officer). In Missouri prisons, the danger of such coordinated criminal activity is exacerbated by the presence of prison gangs. The Missouri policy of separating and isolating gang members—a strategy that has been frequently used to control gang activity, see G. Camp & C. Camp, U. S. Dept, of Justice, Prison Gangs: Their Extent, Nature and Impact on Prisons 64-65 (1985)— logically is furthered by the restriction on prisoner-to-prisoner correspondence. Moreover, the correspondence regulation does not deprive prisoners of all means of expression. Rather, it bars communication only with a limited class of other people with whom prison officials have particular cause to be concerned—inmates at other institutions within the Missouri prison system. We also think that the Court of Appeals’ analysis overlooks the impact of respondents’ asserted right on other inmates and prison personnel. Prison officials have stated that in their expert opinion, correspondence between prison institutions facilitates the development of informal organizations that threaten the core functions of prison administration, maintaining safety and internai security. As a result, the correspondence rights asserted by respondents, like the organizational activities at issue in Jones n. North Carolina Prisoners’ Union, 433 U. S. 119 (1977), can be exercised only at the cost of significantly less liberty and safety for everyone else, guards and other prisoners alike. Indeed, the potential “ripple effect” is even broader here than in Jones, because exercise of the right affects the inmates and staff of more than one institution. Where exercise of a right requires this kind of tradeoff, we think that the choice made by corrections officials—which is, after all, a judgment “peculiarly within [their] province and professional expertise,” Pell n. Pro- TURNER v. SAFLEY 93 78 Opinion of the Court cunier, 417 U. S., at 827—should not be lightly set aside by the courts. Finally, there are no obvious, easy alternatives to the policy adopted by petitioners. Other well-run prison systems, including the Federal Bureau of Prisons, have concluded that substantially similar restrictions on inmate correspondence were necessary to protect institutional order and security. See, e. g., 28 CFR §540.17 (1986). As petitioners have shown, the only alternative proffered by the claimant prisoners, the monitoring of inmate correspondence, clearly would impose more than a de minimis cost on the pursuit of legitimate corrections goals. Prison officials testified that it would be impossible to read every piece of inmate-to-inmate correspondence, 3 Tr. 159, 4 id., at 42-43, and consequently there would be an appreciable risk of missing dangerous messages. In any event, prisoners could easily write in jargon or codes to prevent detection of their real messages. See Camp & Camp, supra, at 130 (noting “frequent” use of coded correspondence by gang members in federal prison); see also Brief for State of Texas as Amicus Curiae 7-9. The risk of missing dangerous communications, taken together with the sheer burden on staff resources required to conduct item-by-item censorship, see 3 Tr. 176, supports the judgment of prison officials that this alternative is not an adequate alternative to restricting correspondence. The prohibition on correspondence is reasonably related to valid corrections goals. The rule is content neutral, it logically advances the goals of institutional security and safety identified by Missouri prison officials, and it is not an exaggerated response to those objectives. On that basis, we conclude that the regulation does not unconstitutionally abridge the First Amendment rights of prison inmates.* *Suggesting that there is little difference between the “unnecessarily sweeping” standard applied by the District Court in reaching its judgment and the reasonableness standard described in Part II, see post, at 105, Justice Stevens complains that we have “ignore[d] the findings of fact that 94 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. B In support of the marriage regulation, petitioners first suggest that the rule does not deprive prisoners of a constitu- were made by the District Court,” post, at 102, n. 2, and have improperly “encroach[ed] into the factfinding domain of the District Court.” Post, at 101. The District Court’s inquiry as to whether the regulations were “needlessly broad” is not just semantically different from the standard we have articulated in Part II: it is the least restrictive alternative test of Procunier v. Martinez, 416 U. S. 396 (1974). As Martinez states, in a passage quoted by the District Court: “[T]he limitation of First Amendment freedoms must be no greater than is necessary or essential to the protection of the particular governmental interest involved. Thus a restriction on inmate correspondence . . . will . . . be invalid if its sweep is unnecessarily broad.” Id., at 413-414 (emphasis added). The District Court’s judgment that the correspondence regulation was “unnecessarily sweeping,” 586 F. Supp. 589, 596 (WD Mo. 1984), thus was a judgment based on application of an erroneous legal standard. The District Court’s findings of fact 7 and 13 likewise are predicated on application of the least restrictive means standard. Finding 7 is that the correspondence rule was applied without a letter-by-letter determination of harm, and without a showing that “there is no less restrictive alternative” available; finding 13 reiterates that the correspondence rule operated as a complete ban. See id., at 591-592. These findings are important only if petitioners have to show that the correspondence regulation satisfies a least restrictive alternative test: they are largely beside the point where the inquiry is simply whether the regulation is reasonably related to a legitimate governmental interest. Justice Stevens’ charge of appellate factfinding likewise suffers from the flawed premise that Part III-A answers the question Justice Stevens would pose, namely, whether the correspondence regulation satisfies strict scrutiny. Thus, our conclusion that there is a logical connection between security concerns identified by petitioners and a ban on inmate-to-inmate correspondence, see supra, at 91-92, becomes, in Justice Stevens’ hands, a searching examination of the record to determine whether there was sufficient proof that inmate correspondence had actually led to an escape plot, uprising, or gang violence at Renz. See post, at 106-109. Likewise, our conclusion that monitoring inmate correspondence “clearly would impose more than a de minimis cost on the pursuit of legitimate corrections goals,” supra, at 93, is described as a factual “finding” that it TURNER v. SAFLEY 95 78 Opinion of the Court tionally protected right. They concede that the decision to marry is a fundamental right under Zablocki v. Redhail, 434 U. S. 374 (1978), and Loving v. Virginia, 388 U. S. 1 (1967), but they imply that a different rule should obtain “in ... a prison forum.” See Brief for Petitioners 38, n. 6. Petitioners then argue that even if the regulation burdens inmates’ constitutional rights, the restriction should be tested under a reasonableness standard. They urge that the restriction is reasonably related to legitimate security and rehabilitation concerns. We disagree with petitioners that Zablocki does not apply to prison inmates. It is settled that a prison inmate “retains those [constitutional] rights that are not inconsistent with his status as a prisoner or with the legitimate penological objectives of the corrections system.” Pell v. Procunier, supra, at 822. The right to marry, like many other rights, is subject to substantial restrictions as a result of incarceration. Many important attributes of marriage remain, however, after taking into account the limitations imposed by prison life. First, inmate marriages, like others, are expressions of emotional support and public commitment. These elements would be “an insurmountable task” to read all correspondence sent to or received by the inmates at Renz. Post, at 110, 112. Nowhere, of course, do we make such a “finding,” nor is it necessary to do so unless one is applying a least restrictive means test. Finally, Justice Stevens complains that Renz’ ban on inmate correspondence cannot be reasonably related to legitimate corrections goals because it is more restrictive than the rule at other Missouri institutions. As our previous decisions make clear, however, the Constitution “does not mandate a ‘lowest common denominator’ security standard, whereby a practice permitted at one penal institution must be permitted at all institutions.” Bell v. Wolfish, 441 U. S. 520, 554 (1979). Renz raises different security concerns from other Missouri institutions, both because it houses medium and maximum security prisoners in a facility without walls or guard towers, and because it is used to house inmates in protective custody. Moreover, the Renz rule is consistent with the practice of other well-run institutions, including institutions in the federal system. See Brief for United States as Amicus Curiae 22-24. 96 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. are an important and significant aspect of the marital relationship. In addition, many religions recognize marriage as having spiritual significance; for some inmates and their spouses, therefore, the commitment of marriage may be an exercise of religious faith as well as an expression of personal dedication. Third, most inmates eventually will be released by parole or commutation, and therefore most inmate marriages are formed in the expectation that they ultimately will be fully consummated. Finally, marital status often is a precondition to the receipt of government benefits (e. g., Social Security benefits), property rights (e. g., tenancy by the entirety, inheritance rights), and other, less tangible benefits (e. g., legitimation of children born out of wedlock). These incidents of marriage, like the religious and personal aspects of the marriage commitment, are unaffected by the fact of confinement or the pursuit of legitimate corrections goals. Taken together, we conclude that these remaining elements are sufficient to form a constitutionally protected marital relationship in the prison context. Our decision in Butler v. Wilson, 415 U. S. 953 (1974), summarily affirming Johnson v. Rockefeller, 365 F. Supp. 377 (SDNY 1973), is not to the contrary. That case involved a prohibition on marriage only for inmates sentenced to life imprisonment; and, importantly, denial of the right was part of the punishment for crime. See id., at 381-382 (Lasker, J., concurring in part and dissenting in part) (asserted governmental interest of punishing crime sufficiently important to justify deprivation of right); see generally Mandel n. Bradley, 432 U. S. 173, 176 (1977) (“Because a summary affirmance is an affirmance of the judgment only, the rationale of the affirmance may not be gleaned solely from the opinion below”). The Missouri marriage regulation prohibits inmates from marrying unless the prison superintendent has approved the marriage after finding that there are compelling reasons for doing so. As noted previously, generally only pregnancy or birth of a child is considered a “compelling reason” to approve TURNER v. SAFLEY 97 78 Opinion of the Court a marriage. In determining whether this regulation impermissibly burdens the right to marry, we note initially that the regulation prohibits marriages between inmates and civilians, as well as marriages between inmates. See Brief for Petitioners 40. Although not urged by respondents, this implication of the interests of nonprisoners may support application of the Martinez standard, because the regulation may entail a “consequential restriction on the [constitutional] rights of those who are not prisoners.” See Procunier v. Martinez, 416 U. S., at 409. We need not reach this question, however, because even under the reasonable relationship test, the marriage regulation does not withstand scrutiny. Petitioners have identified both security and rehabilitation concerns in support of the marriage prohibition. The security concern emphasized by petitioners is that “love triangles” might lead to violent confrontations between inmates. See Brief for Petitioners 13, 36, 39. With respect to rehabilitation, prison officials testified that female prisoners often were subject to abuse at home or were overly dependent on male figures, and that this dependence or abuse was connected to the crimes they had committed. 3 Tr. 154-155. The superintendent at Renz, petitioner William Turner, testified that in his view, these women prisoners needed to concentrate on developing skills of self-reliance, 1 id., at 80-81, and that the prohibition on marriage furthered this rehabilitative goal. Petitioners emphasize that the prohibition on marriage should be understood in light of Superintendent Turner’s experience with several ill-advised marriage requests from female inmates. Brief for Petitioners 32-34. We conclude that on this record, the Missouri prison regulation, as written, is not reasonably related to these penological interests. No doubt legitimate security concerns may require placing reasonable restrictions upon an inmate’s right to marry, and may justify requiring approval of the superintendent. The Missouri regulation, however, represents an 98 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. exaggerated response to such security objectives. There are obvious, easy alternatives to the Missouri regulation that accommodate the right to marry while imposing a de minimis burden on the pursuit of security objectives. See, e. g., 28 CFR §551.10 (1986) (marriage by inmates in federal prison generally permitted, but not if warden finds that it presents a threat to security or order of institution, or to public safety). We are aware of no place in the record where prison officials testified that such ready alternatives would not fully satisfy their security concerns. Moreover, with respect to the security concern emphasized in petitioners’ brief—the creation of “love triangles”—petitioners have pointed to nothing in the record suggesting that the marriage regulation was viewed as preventing such entanglements. Common sense likewise suggests that there is no logical connection between the marriage restriction and the formation of love triangles: surely in prisons housing both male and female prisoners, inmate rivalries are as likely to develop without a formal marriage ceremony as with one. Finally, this is not an instance where the “ripple effect” on the security of fellow inmates and prison staff justifies a broad restriction on inmates’ rights —indeed, where the inmate wishes to marry a civilian, the decision to marry (apart from the logistics of the wedding ceremony) is a completely private one. Nor, on this record, is the marriage restriction reasonably related to the articulated rehabilitation goal. First, in requiring refusal of permission absent a finding of a compelling reason to allow the marriage, the rule sweeps much more broadly than can be explained by petitioners’ penological objectives. Missouri prison officials testified that generally they had experienced no problem with the marriage of male inmates, see, e. g., 2 Tr. 21-22, and the District Court found that such marriages had routinely been allowed as a matter of practice at Missouri correctional institutions prior to adoption of the rule, 586 F. Supp., at 592. The proffered justification thus does not explain the adoption of a rule banning TURNER v. SAFLEY 99 78 Opinion of the Court marriages by these inmates. Nor does it account for the prohibition on inmate marriages to civilians. Missouri prison officials testified that generally they had no objection to inmate-civilian marriages, see, e. g., 4 Tr. 240-241, and Superintendent Turner testified that he usually did not object to the marriage of either male or female prisoners to civilians, 2 id., at 141-142. The rehabilitation concern appears from the record to have been centered almost exclusively on female inmates marrying other inmates or exfelons; it does not account for the ban on inmate-civilian marriages. Moreover, although not necessary to the disposition of this case, we note that on this record the rehabilitative objective asserted to support the regulation itself is suspect. Of the several female inmates whose marriage requests were discussed by prison officials at trial, only one was refused on the basis of fostering excessive dependency. The District Court found that the Missouri prison system operated on the basis of excessive paternalism in that the proposed marriages of all female inmates were scrutinized carefully even before adoption of the current regulation—only one was approved at Renz in the period from 1979-1983—whereas the marriages of male inmates during the same period were routinely approved. That kind of lopsided rehabilitation concern cannot provide a justification for the broad Missouri marriage rule. It is undisputed that Missouri prison officials may regulate the time and circumstances under which the marriage ceremony itself takes place. See Brief for Respondents 5. On this record, however, the almost complete ban on the decision to marry is not reasonably related to legitimate penological objectives. We conclude, therefore, that the Missouri marriage regulation is facially invalid. IV We uphold the facial validity of the correspondence regulation, but we conclude that the marriage rule is constitution 100 OCTOBER TERM, 1986 Opinion of Stevens, J. 482 U. S. ally infirm. We read petitioners’ additional challenge to the District Court’s findings of fact to be a claim that the District Court erred in holding that the correspondence regulation had been applied by prison officials in an arbitrary and capricious manner. Because the Court of Appeals did not address this question, we remand the issue to the Court of Appeals for its consideration. Accordingly, the judgment of the Court of Appeals striking down the Missouri marriage regulation is affirmed; its judgment invalidating the correspondence rule is reversed; and the case is remanded to the Court of Appeals for further proceedings consistent with this opinion. It is so ordered. Justice Stevens, with whom Justice Brennan, Justice Marshall, and Justice Blackmun join, concurring in part and dissenting in part. How a court describes its standard of review when a prison regulation infringes fundamental constitutional rights often has far less consequence for the inmates than the actual showing that the court demands of the State in order to uphold the regulation. This case provides a prime example. There would not appear to be much difference between the question whether a prison regulation that burdens fundamental rights in the quest for security is “needlessly broad”— the standard applied by the District Court and the Court of Appeals—and this Court’s requirement that the regulation must be “reasonably related to legitimate penological interests,” ante, at 89, and may not represent “an ‘exaggerated response’ to those concerns.” Ante, at 87. But if the standard can be satisfied by nothing more than a “logical connection” between the regulation and any legitimate penological concern perceived by a cautious warden, see ante, at 94, n. (emphasis in original), it is virtually meaningless. Application of the standard would seem to permit disregard for inmates’ constitutional rights whenever the imagination of the TURNER v. SAFLEY 101 78 Opinion of Stevens, J. warden produces a plausible security concern and a deferential trial court is able to discern a logical connection between that concern and the challenged regulation. Indeed, there is a logical connection between prison discipline and the use of bullwhips on prisoners; and security is logically furthered by a total ban on inmate communication, not only with other inmates but also with outsiders who conceivably might be interested in arranging an attack within the prison or an escape from it. Thus, I dissent from Part II of the Court’s opinion.1 I am able to join Part III-B because the Court’s invalidation of the marriage regulation does not rely on a rejection of a standard of review more stringent than the one announced in Part II. See ante, at 97. The Court in Part III-B concludes after careful examination that, even applying a “reasonableness” standard, the marriage regulation must fail because the justifications asserted on its behalf lack record support. Part III-A, however, is not only based on an application of the Court’s newly minted standard, see ante, at 89, but also represents the product of a plainly improper appellate encroachment into the factfinding domain of the District Court. See Icicle Seafoods, Inc. v. Worthington, 475 U. S. 709, 714 (1986). Indeed, a fundamental difference between the Court of Appeals and this Court in this case—and the principal point of this dissent—rests in the respective ways the two courts have examined and made use of the trial record. In my opinion the Court of Appeals correctly held that the trial court’s findings of fact adequately supported its judgment sustaining the inmates’ challenge to the mail 'The Court’s rather open-ended “reasonableness” standard makes it much too easy to uphold restrictions on prisoners’ First Amendment rights on the basis of administrative concerns and speculation about possible security risks rather than on the basis of evidence that the restrictions are needed to further an important governmental interest. Judge Kaufman’s opinion in Abdul Wali v. Coughlin, 754 F. 2d 1015, 1033 (CA2 1985), makes a more careful attempt to strike a fair balance between legitimate penological concerns and the well-settled proposition that inmates do not give up all constitutional rights by virtue of incarceration. 102 OCTOBER TERM, 1986 Opinion of Stevens, J. 482 U. S. regulation as it has been administered at the Renz Correctional Center in Cedar City, Missouri. In contrast, this Court sifts the trial testimony on its own2 in order to uphold a general prohibition against correspondence between unrelated inmates. I This is not a case in which it is particularly helpful to begin by determining the “proper” standard of review, as if the result of that preliminary activity would somehow lighten the Court’s duty to decide this case. The precise issue before us is evident from respondents’ complaint, which makes clear that they were not launching an exclusively facial attack against the correspondence regulation. Respondents instead leveled their primary challenge against the application of this regulation to mail addressed to or sent by inmates at Renz: “20. On information and belief, correspondence between non-family members at different institutions within the Missouri Division of Correction system is permitted at all institutions with the exception of Renz. On information and belief, defendant Turner and other employees of the Missouri Division of Corrections have a pattern and practice of refusing to permit inmates of Renz to correspond with or receive letters from inmates at other correctional institutions, a situation which appears to be unique within the Missouri Division of Corrections. “21. On information and belief, the reason given for refusing such correspondence is that Superintendent Turner feels that correspondence between inmates is not 2 The Court cites portions of the trial transcript and the amicus curiae brief filed by the State of Texas, ante, at 91, 93, but completely ignores the findings of fact that were made by the District Court and that bind appellate courts unless clearly erroneous. Fed. Rule Civ. Proc. 52(a). The Court does not and could not deem these particular findings clearly erroneous. TURNER v. SAFLEY 103 78 Opinion of Stevens, J. in the best interest of any inmate. In this manner defendant Turner has violated the constitutional right of every inmate residing at Renz and any inmate who desires to correspond with an inmate residing at Renz.” Amended Complaint, App. 11-12. On their face, the regulations generally applicable to the Missouri Correctional System permit correspondence between unrelated inmates “if the classification/treatment team of each inmate deems it in the best interests of the parties involved.”3 After a bench trial, however, the District Court found that there was a total ban on such correspondence at Renz: “6. The provisions of the divisional correspondence regulation allowing the classification/treatment team of each inmate to prohibit inmate-to-inmate correspondence have not been followed at Renz. Theoretically the classification/treatment team uses psychological reports, conduct violations, and progress reports in deciding whether to permit correspondence. At Renz, however, the rule as practiced is that inmates may not write non-family inmates or receive mail from non-family inmates. The more restrictive practice is set forth in the Renz Inmate Orientation Booklet presented to each inmate upon arrival at Renz. The restrictive rule at Renz is commonly known throughout the Missouri Correctional System. “7. The Renz rule against inmate-to-inmate correspondence is enforced without a determination that the security or order of Renz or the rehabilitation of the inmate would be harmed by allowing the particular correspondence to proceed and without a determination that there is no less restrictive alternative to resolve any legitimate concerns of the Department of Corrections short of prohibiting all correspondence. 3 586 F. Supp. 589, 591 (WD Mo. 1984). 104 OCTOBER TERM, 1986 Opinion of Stevens, J. 482 U. S. “8. Inmates at most institutions in the Missouri Correctional System are permitted to correspond with inmates in most other institutions. The greatest restriction on inmate correspondence is practiced at Renz.” 586 F. Supp. 589, 591 (WD Mo. 1984). “13. Correspondence between inmates has been denied despite evidence that the correspondence was desired simply to maintain wholesome friendships.” Id., at 591-592. These factual findings, which bear out respondents’ complaint, served as the basis for the District Court’s injunction: “Even if some restriction on inmate-to-inmate correspondence can be justified, the regulations and practices at bar must fall. The prohibitions are unnecessarily sweeping. Correspondence is a sufficiently protected right that it cannot be cut off simply because the recipient is in another prison, and the inmates cannot demonstrate special cause for the correspondence. . . . “Defendants have failed to demonstrate that the needs of Renz are sufficiently different to justify greater censorship than is applied by other well-run institutions.” Id., at 596. After reviewing the District Court’s findings and conclusions, the Court of Appeals held: “[W]ithout strong evidence that the relationship in question is or will be abusive, the connection between permitting the desired correspondence or marriage and the subsequent commission of a crime caused thereby is simply too tenuous to justify denial of those constitutionally protected rights. As to the security concerns, we think the prison officials’ authority to open and read all prisoner mail is sufficient to meet the problem of illegal conspiracies.” 777 F. 2d 1307, 1315-1316 (CA8 1985) (emphasis added). TURNER v. SAFLEY 105 78 Opinion of Stevens, J. The Court of Appeals’ affirmance of the District Court thus ultimately rests upon a conclusion with which I fully agree: absent a showing that prison officials would be unable to anticipate and avoid any security problems associated with the inmate-to-inmate mail that would result from application of the correspondence rule as it is written and as enforced at other Missouri prisons, the total ban at Renz found by the District Court offends the First Amendment. The ostensible breadth of the Court of Appeals’ opinion4 furnishes no license for this Court to reverse with another unnecessarily broad holding. Moreover, even under the Court’s newly minted standard, the findings of the District Court that were upheld by the Court of Appeals clearly dictate affirmance of the judgment below. II Without explicitly disagreeing with any of the District Court’s findings of fact, this Court rejects the trial judge’s conclusion that the total ban on correspondence between inmates at Renz and unrelated inmates in other correctional facilities was “unnecessarily sweeping” or, to use the language the Court seems to prefer, was an “exaggerated response” to the security problems predicted by petitioner’s expert witnesses. Instead, the Court bases its holding upon its own highly selective use of factual evidence. The reasons the Court advances in support of its conclusion include: (1) speculation about possible “gang problems,” escapes, and secret codes, ante, at 91-93; (2) the fact that the correspondence regulation “does not deprive prisoners of all means of expression,” ante, at 92; and (3) testimony indicat 4 The Court of Appeals may have used unnecessarily sweeping language in its opinion: “We conclude that the exchange of inmate-to-inmate mail is not presumptively dangerous nor inherently inconsistent with legitimate penological objectives. We therefore affirm the district court’s application of the Martinez strict scrutiny standard and its decision finding the Renz correspondence rule unconstitutional.” 777 F. 2d, at 1313. 106 OCTOBER TERM, 1986 Opinion of Stevens, J. 482 U. S. ing “that it would be impossible to read every piece of inmate-to-inmate correspondence,” ante, at 93. None of these reasons has a sufficient basis in the record to support the Court’s holding on the mail regulation. Speculation about the possible adverse consequences of allowing inmates in different institutions to correspond with one another is found in the testimony of three witnesses: William Turner, the Superintendent of Renz Correctional Center; Sally Halford, the Director of the Kansas Correctional Institution at Lansing; and David Blackwell, the former Director of the Division of Adult Institutions of the Missouri Department of Corrections. Superintendent Turner was unable to offer proof that prohibiting inmate-to-inmate correspondence prevented the formation or dissemination of escape plots. He merely asserted that the mail regulation assisted him in his duties to maintain security at Renz “[f]rom the standpoint that we don’t have escapes, we don’t have the problems that are experienced in other institutions.” 2 Tr. 75. Nor did the Superintendent’s testimony establish that permitting such correspondence would create a security risk; he could only surmise that the mail policy would inhibit communications between institutions in the early stages of an uprising. Id., at 76. The Superintendent’s testimony is entirely consistent with the District Court’s conclusion that the correspondence regulation was an exaggerated response to the potential gang problem at Renz.5 6 Superintendent Turner had not experienced any problem with gang warfare at Renz. 2 Tr. 117. He had not found any correspondence between gang members coming into Renz. Id., at 118. He also conceded that it would be possible to screen out correspondence that posed the danger of leading to gang warfare: “Q: Is there any reason that you could not read correspondence from other institutions to determine if these people were writing about gang warfare or something like that? “A: I think from the standpoint of the dictates of the department and, of course, the dictates of the court, I could if there was a problem. From the TURNER v. SAFLEY 107 78 Opinion of Stevens, J. Neither of the outside witnesses had any special knowledge of conditions at Renz. Ms. Halford had reviewed the prison’s rules and regulations relevant to this case, had discussed the case with Superintendent Turner, and had visited Renz for “a couple of hours.” 3 id., at 146. Mr. Blackwell was charged with the overall management of Missouri’s adult correctional facilities and did not make daily decisions concerning the inmate correspondence permitted at Renz. Id., at 259-260. He was “not sure” if he was specifically familiar with the policy at Renz that an inmate is allowed to correspond with inmates of other institutions only if they are members of the inmate’s immediate family. 4 id., at 44. Neither of them, and indeed, no other witness, even mentioned the possibility of the use of secret codes by inmates. The Kansas witness testified that Kansas followed a policy of “open correspondence.. . . An inmate can write to whomever they please.” 3 id., at 158. She identified two problems that might result from that policy. First, in the preceding year a male inmate had escaped from a minimum security area and helped a female inmate to escape and remain at large for over a week. The witness speculated that they must have used the mails to plan their escape. The trial judge discounted this testimony because there was no proof that this or any other escape had been discussed in correspondence. Id., at 158-159. Second, the Kansas witness suggested that a ban on inmate correspondence would frustrate the development of a “gang problem.” Id., at 160. In view of her acknowledgment that no gang problem had developed in Kansas despite its open correspondence rule, id., at standpoint of dealing with these people individually or personally, no. It would be a problem.” Ibid. “Q: Now, let’s limit it to people who you suspect might be involved in gang warfare, for example. Do you have any reason to say it would be impossible to read all the mail of those particular people? “A: Those that we know of that have been identified, no, it wouldn’t be impossible.” Id., at 119. 108 OCTOBER TERM, 1986 Opinion of Stevens, J. 482 U. S. 158, the trial judge presumably also attached little weight to this prediction. Indeed, there is a certain irony in the fact that the Kansas expert witness was unable to persuade her superiors in Kansas to prohibit inmate-to-inmate correspondence, id., at 168, yet this Court apparently finds no reason to discount her speculative testimony.6 The Missouri witness, Mr. Blackwell, also testified that one method of trying to discourage the organization of “gangs” of prisoners with ethnic or religious similarities is “by restricting correspondence.” Id., at 267. He did not testify, however, that a total ban on inmate-to-inmate correspondence was an appropriate response to the potential gang problem. Indeed, he stated that the State’s policy did not include a “carte blanche” denial of such correspondence,7 and he did not even know that Renz was enforcing such a total ban.8 His assertion that an open correspon 6 There is a further irony. While Missouri ostensibly does not have sufficient resources to permit and screen inmate-to-inmate mail, Kansas apparently lacks sufficient resources to ban it. Ms. Halford testified that open correspondence was not abrogated in the Kansas correctional system despite security concerns because her superiors felt that it was “too much of an effort to restrict it, that it tied up staff to send out all forms to the various and sundry institutions. So I think we’re all basically in agreement that even though it is a problem to have open correspondence, the reason that we don’t do it is simply staff time.” 3 id., at 168. 7 “Q. Those inmates who are allowed to write, you do not find it necessary to stop their correspondence as a matter of course; isn’t that true? “A. No, we don’t stop it as a matter of course and we don’t authorize it as a matter of course. There is no carte blanche approval or denial at any facility. It is done on a case by case individual basis and would have to be. “Q. Let me refer specifically to inmate-to-inmate. Are you saying there is no carte blanche denial of inmate-to-inmate or the inmates aren’t told that at Renz Correctional Center? “A. The Division policy is not carte blanche [to] deny inmate-to-inmate, or to approve it.” 4 id., at 43. 8“Q. You do know that is the rule at Renz that they cannot write to other institutions unless the inmate is a relative? “A. I am not certain that that is the rule, no. TURNER v. SAFLEY 109 78 Opinion of Stevens, J. dence policy would pose security problems was backed only by speculation: “[A]: ... I am sure that there are some inmates at Renz who would write other inmates at other facilities in an illegitimate fashion. I also feel certain that there is more of a probability that they would be writing about things other than just sound positive letter writing, given the nature of the offenders at Renz. “Q: What percentage of the [mail] inmate-to-inmate from Renz Correctional Center have you personally read? “A: Very, very little. “Q: So you are basically speculating about what inmates might write about? “A: Yes.” 4 id., at 82-83. Quite clearly, Mr. Blackwell’s estimate of the problems justifying some restrictions on inmate-to-inmate correspondence provides no support for the Renz policy that he did not even know about and that did not conform to the more liberal policy applicable to other institutions in which more serious offenders are incarcerated.9 As the District Court concluded, petitioners “failed to demonstrate that the needs of Renz are sufficiently different to justify greater censorship than is applied by other well-run institutions.” 586 F. Supp., at 596. “Q. Let me hand you Plaintiffs’ Exhibit B, excuse me, Defendants’ Exhibit B. I don’t have the plaintiffs’ number. This is in evidence. It is the inmate orientation manual, February 1983. I direct your attention to the paragraph that says correspondence with inmates of other institutions is permitted with immediate family members only. “Now, were you familiar with that being the policy at Renz Correctional Center? “A. I am not sure if I was specifically or not.” Id., at 44. 9 At the time of trial, the Renz Correctional Center contained both male and female prisoners of varying security level classifications. Most of the female inmates were medium and maximum security offenders, while most of the male inmates were minimum security offenders. 777 F. 2d 1307, 1308 (CA8 1985). 110 OCTOBER TERM, 1986 Opinion of Stevens, J. 482 U. S. The Court also relies on the fact that the inmates at Renz were not totally deprived of the opportunity to communicate with the outside world. This observation is simply irrelevant to the question whether the restrictions that were enforced were unnecessarily broad. Moreover, an evenhanded acceptance of this sort of argument would require upholding the Renz marriage regulation—which the Court quite properly invalidates—because that regulation also could have been even more restrictive. The Court’s final reason for concluding that the Renz prohibition on inmate-to-inmate correspondence is reasonable is its belief that it would be “impossible” to read all such correspondence sent or received by the inmates at Renz. No such finding of impossibility was made by the District Court, nor would it be supported by any of the findings that it did make. The record tells us nothing about the total volume of inmate mail sent or received at Renz; much less does it indicate how many letters are sent to, or received from, inmates at other institutions. As the State itself observed at oral argument about the volume of correspondence: “The difficulty with our position in the case is, since we had never permitted [mail between inmates], we didn’t have an idea except to say that—you know, except that we had 8,000 inmates, and we figured that they would write.” Tr. of Oral Arg. 14. The testimony the Court does cite to support its conclusion that reviewing inmate-to-inmate mail would be an insurmountable task was provided by Mr. Blackwell and Ms. Halford. Mr. Blackwell testified that “[t]here is no way we can read all the mail nor would we want to ... it is impossible.” 4 Tr. 41-43.10 Ms. Halford gave similar testi 10 “Q. The question was do you realize the plaintiffs in this case accept the rights of the Division of Corrections to read all their mail if the Division wants to? “A. There is no way we can read all the mail nor would we want to.” 4 Tr. 41. TURNER v. SAFLEY 111 78 Opinion of Stevens, J. mony,11 but again she was referring to “all incoming mail,” not to inmate-to-inmate correspondence and, of course, her testimony related to Kansas, not to the relatively small facility at Renz.12 In short, the evidence in the record is plainly “Q. Let me hand you Exhibit No. 3, sir, the mail and visiting rule for the Department of Corrections, specifically concerning inmate mail signed by you. “I direct your attention to paragraph 1(C), outgoing letters will not be sealed by the inmate. And further down in the paragraph, all letters may be inspected in the mail room and examined for contraband, escape plots, forgery, fraud, and other schemes. “Now, tell me, sir, how do you examine a letter for an escape plot without reading it? “A. We do not read mail. This does not say mail will be read. The only time we read a letter is when we have reason to believe, for example, that an escape is being planned. [W]hen a letter is being planned, there is no way we want to or know to read all inmate mail. It is impossible.” Id., at 42-43. There was no record indication of the amount of correspondence between inmates that would occur if it were permitted. Mr. Blackwell stated only that in his opinion, “if we do allow inmates to write other inmates pretty much at will, the vast majority will be writing one another, at least one other offender in another institution. I think it is obvious what it will do to mail room load.” Id., at 108. 11 “[I]n Kansas we have, our rules and regulations allow us to read all incoming mail. Due to the volume of mail that is absolutely impossible to do.” 3 id., at 159. 12 The average population at Renz in the 1983 fiscal year was 270. See American Correctional Assn., Juvenile and Adult Correctional Departments, Institutions, Agencies, and Paroling Authorities 214 (1984). When Ms. Halford was asked why the prison officials did not read all of the inmate mail, she gave this response: “A. To begin with it’s very boring reading. Another thing, I think it’s a poor use of staff time. If I get more staff in, I would like to have them doing something more important than reading inmate mail. That seems to me to be kind of a waste of time.” Tr. 176. Earl Englebrecht testified that at Renz he scanned the contents of all approved incoming mail from other institutions, and that this task and scanning some outgoing mail together took approximately one hour a day. 5 id., at 97, 99. He could not indicate with any certainty the additional screening burden that more frequent inmate-to-inmate correspondence 112 OCTOBER TERM, 1986 Opinion of Stevens, J. 482 U. S. insufficient to support the Court’s de novo finding of impossibility.13 It does, however, adequately support this finding by the District Court that the Court ignores: “14. The staff at Renz has been able to scan and control outgoing and incoming mail, including inmate-to-inmate correspondence.” 586 F. Supp., at 592. Because the record contradicts the conclusion that the administrative burden of screening all inmate-to-inmate mail would be unbearable, an outright ban is intolerable. The blanket prohibition enforced at Renz is not only an “excessive response” to any legitimate security concern; it is inconsistent with a consensus of expert opinion—including Kansas correctional authorities—that is far more reliable than the speculation to which this Court accords deference.14 Ill The contrasts between the Court’s acceptance of the challenge to the marriage regulation as overbroad and its rejection of the challenge to the correspondence rule are striking would impose on him and on the mail room. Id., at 102. The testimony of these two witnesses is hardly consistent with the C.ourt’s assumption that it would be “impossible” to read the portion of the correspondence that is addressed to, or received from, inmates in other institutions. 13 The Court’s speculation, ante, at 88, 93, about the ability of prisoners to use codes is based on a suggestion in an amicus curiae brief, see Brief for State of Texas as Amicus Curiae 7-9, and is totally unsupported by record evidence. 14 See ABA Standards for Criminal Justice 23-6.1, Commentary, p. 23*76 (2d ed. 1980) (“[P]risoners can write at any length they choose, using any language they desire, to correspondents of their selection, including present or former prisoners, with no more controls than those which govern the public at large”). The American Correctional Association has set forth the “current standards deemed appropriate by detention facility managers and recognized organizations representing corrections.” ACA, Standards for Adult Local Detention Facilities xiii (2d ed. 1981). Standard 2-5328 requires clear and convincing evidence to justify “limitations for reasons of public safety or facility order and security” on the volume, “length, language, content or source” of mail which an inmate may send or receive. Id., at 88. TURNER v. SAFLEY 113 78 Opinion of Stevens, J. and puzzling.15 The Court inexplicably expresses different views about the security concerns common to prison marriages and prison mail. In the marriage context expert speculation about the security problems associated with “love triangles” is summarily rejected, while in the mail context speculation about the potential “gang problem” and the possible use of codes by prisoners receives virtually total deference. Moreover, while the Court correctly dismisses as a defense to the marriage rule the speculation that the inmate’s spouse, once released from incarceration, would attempt to aid the inmate in escaping,16 the Court grants virtually total credence to similar speculation about escape plans concealed in letters. In addition, the Court disregards the same considerations it relies on to invalidate the marriage regulation when it turns to the mail regulation. The marriage rule is said to sweep too broadly because it is more restrictive than the routine practices at other Missouri correctional institutions, but the mail rule at Renz is not an “exaggerated response” even though it is more restrictive than practices in the remainder of the State. The Court finds the rehabilitative value of marriage apparent, but dismisses the value of corresponding with a friend who is also an inmate for the reason that communication with the outside world is not totally prohibited. The Court relies on the District Court’s finding that the marriage regulation operated on the basis of “excessive paternal 15 The Court’s bifurcated treatment of the mail and marriage regulations leads to the absurd result that an inmate at Renz may marry another inmate, but may not carry on the courtship leading to the marriage by corresponding with him or her beforehand because he or she would not then be an “immediate family member.” 16 Explaining why the request of inmate Diana Finley to be married to inmate William Quillam was denied, Superintendent Turner stated: “If he gets out, then we have got some security problems. . . . The threat, if a man gets out of the penitentiary and he is married to her and he wants his wife with him, there is very little that we can do to stop an escape from that institution because we don’t have the security, sophisticated security, like a maximum security institution.” 1 Tr. 185-186. See also id., at 187. 114 OCTOBER TERM, 1986 Opinion of Stevens, J. 482 U. S. ism” toward female inmates, ante, at 99, but rejects the same court’s factual findings on the correspondence regulation. Unfathomably, while rejecting the Superintendent’s concerns about love triangles as an insufficient and invalid basis for the marriage regulation, the Court apparently accepts the same concerns as a valid basis for the mail regulation.17 17 One of Superintendent Turner’s articulated reasons for preventing one female inmate from corresponding with a male inmate closely tracks the “love triangle” rationale advanced for the marriage regulation: “Q: Let’s take Ms. Flowers. Do you know of any reason why she should not be allowed to write to Mr. Barks? “A: Yes. “Q: Why? “A: She has two other men. One she wants to get married to, another man that she was involved with at Renz resides with Mr. Barks. “Q: Let me ask you this. You have mentioned on two or three occasions that people want to get married to one man or the other. Is it your understanding that the only possible relationship between a woman and a man is one of intending to get married? “A: Well, when they speak of love and want to marry two people, I think that one of them is going to be cut short.” Id., at 237-238. The Superintendent later elaborated on redirect examination: “Q: Now you have given an example of a problem that in your opinion justifies restrictions on correspondence as being, say, two men who were corresponding with a particular woman. Would it also be possible to call the two men in and have a chat with them in your office and try to resolve that between them? “A: I don’t see where that is necessary in my position.” 2 id., at 116-117. The paternalistic enforcement of the correspondence rule to “protect” female inmates prevents them from exchanging letters with more than one male inmate. Assuming a woman has received permission to correspond with a man: “Q: Now, what if the female inmate finds somebody new in the institution, and that person gets [pa]roled, can she then write to the new fellow? “A: Then we have two situations then. “Q: And, therefore, she cannot? “A: I would say that would be a positive [triggering security concerns] situation. It wouldn’t be a wholesome situation, no.” Id., at 134-135. “Q: And suppose she comes to you and says, I don’t want to write this old fellow anymore, I want to write to the new fellow. Is she then allowed to write to the new fellow? TURNER v. SAFLEY 115 78 Opinion of Stevens, J. In pointing out these inconsistencies, I do not suggest that the Court’s treatment of the marriage regulation is flawed; as I stated, I concur fully in that part of its opinion. I do suggest that consistent application of the Court’s reasoning necessarily leads to a finding that the mail regulation applied at Renz is unconstitutional.18 IV To the extent that this Court affirms the judgment of the Court of Appeals, I concur in its opinion. I respectfully dissent from the Court’s partial reversal of that judgment on the basis of its own selective forays into the record. When all “A: Then we still have a problem. “Q: Once an inmate makes a decision to write to—once a female inmate makes a decision to write to another male inmate, then she can’t write to anybody? “A: You keep saying females. We have the same situation with the male, too, that could exist.” Id., at 135. David Blackwell testified along the same lines: “If, for example, a male offender was believed to be in love with a female offender and another male offender wants to cause him some difficulty, he can start a rumor or confront the man with her seeing someone else or corresponding with someone else; and it’s caused a variety of security problems by way of love triangles and situations such as that.” 3 id., at 271 (emphasis added). Donald Wyrick, Director of Adult Institutions, Missouri Department of Corrections, similarly testified on the security considerations raised by women writing men at other prisons: ‘Well, many times love affairs develop, then the inmate inside . . . becomes extremely worried about the female inmate, he thinks she is messing around with somebody else, all those kind of things. He becomes agitated, worried, and frustrated, this type thing. In my professional opinion, that could cause him to do bad things. It might even cause him to explode and hurt someone or attempt to escape.” 4 id., at 231-232. 18 Having found a constitutional violation, the District Court has broad discretion in fashioning an appropriate remedy. Cf. United States v. Paradise, 480 U. S. 149,155-156, n. 4 (1987) (Stevens, J., concurring in judgment). The difficulties that a correspondence policy is likely to impose on prison officials screening inmate-to-inmate mail bear on the shaping of an appropriate remedy. It is improper, however, to rely on speculation about these difficulties to obliterate effective judicial review of state actions that abridge a prisoner’s constitutional right to send and receive mail. 116 OCTOBER TERM, 1986 Opinion of Stevens, J. 482 U. S. the language about deference and security is set to one side, the Court’s erratic use of the record to affirm the Court of Appeals only partially may rest on an unarticulated assumption that the marital state is fundamentally different from the exchange of mail in the satisfaction, solace, and support it affords to a confined inmate. Even if such a difference is recognized in literature, history, or anthropology, the text of the Constitution more clearly protects the right to communicate than the right to marry. In this case, both of these rights should receive constitutional recognition and protection. COMMISSIONER v. ASPHALT PRODUCTS CO., INC. 117 Syllabus COMMISSIONER OF INTERNAL REVENUE v. ASPHALT PRODUCTS CO., INC. ON PETITION FOR WRIT OF CERTIORARI TO THE UNITED STATES COURT OF APPEALS FOR THE SIXTH CIRCUIT No. 86-1053. Decided June 1, 1987* Because Asphalt Products Co. (APC) kept its books, and prepared its 1974 federal income tax return, on a cash receipts and disbursement basis, its reported 1974 taxable income did not fully reflect that its 1974 year-end inventories and accounts receivable were substantially higher than in prior years. APC’s 1974 return also claimed a deduction for the expense of driving two trucks to APC from their place of purchase, even though they detoured to pick up equipment bought by APC’s shareholders in their individual capacities. After determining that APC was required to compute its 1974 income on an accrual basis and disallowing the truck transportation deduction as a personal expense of the shareholders, the Commissioner of Internal Revenue, pursuant to 26 U. S. C. § 6653(a)(1), added to the resulting deficiency a penalty in the amount of 5% of the full alleged underpayment, contending that the use of the wrong accounting method and the deduction of the truck transportation expense constituted negligence. Although concluding that APC’s use of cash-basis accounting was nonnegligent, the Tax Court agreed that APC had negligently deducted the truck transportation expense, and therefore added to APC’s deficiency—almost all of which was due to the change in accounting methods—a negligence penalty computed by reference to the full amount of the deficiency. Affirming the finding that the truck transportation deduction was negligent, the Court of Appeals nevertheless reversed the imposition of the negligence penalty on the full amount of the deficiency, concluding that the penalty “should be applied only to that portion of the deficiency attributable to the disallowed deduction.” Held: Section 6653(a)(l)’s plain language—whereby, if “any part of any underpayment” is due to negligence, the Commissioner shall add to the tax a penalty of “5 percent of the underpayment”—clearly establishes that the penalty is imposed on the entire amount of “the underpayment,” not just on the “part of [the] underpayment” attributable to negligence. This conclusion is supported by the Government’s plausible interest in *Together with No. 86-1054, Asphalt Products Co., Inc. v. Commissioner of Internal Revenue, also on petition for writ of certiorari to the same court. 118 OCTOBER TERM, 1986 Per Curiam 482 U. S. deterring negligent tax preparation and by the statute’s explicit limitation of other penalties to the amount of a negligent or fraudulent underpayment. Certiorari denied in No. 86-1054. Certiorari granted in No. 86-1053; 796 F. 2d 843, reversed. Per Curiam. Asphalt Products Co. (APC) manufactures emulsified asphalt, a paving material containing oil refining residues, principally for sale to Tennessee county governments for use in highway construction. For reasons related to the rise in oil prices attending the 1973 Arab oil embargo, APC’s 1974 year-end inventories and accounts receivable were substantially higher than in prior years. Because APC kept its books, and prepared its 1974 federal tax return, on a cash receipts and disbursements basis, its reported 1974 taxable income did not fully reflect these changes. APC’s 1974 return also claimed a deduction of $1,103.04 for the expense of transporting two trucks from their place of purchase in Seattle to Tennessee. The trucks were driven to Tennessee by way of California, where they picked up two trailer-mounted waste water treatment plants bought by APC’s shareholders in their individual capacities. The Commissioner of Internal Revenue determined that, because of the increases in APC’s inventories and accounts receivable, the company’s traditional cash-basis bookkeeping did not “clearly reflect income,” 26 U. S. C. § 446(b), for the 1974 tax year, and APC was therefore required to compute its 1974 income on an accrual basis. The Commissioner also disallowed the deduction for the expense of transporting the trucks and trailers, on the ground that it was a personal expense of the shareholders. After several other adjustments, the Commissioner recomputed APC’s taxes to show a deficiency of $154,332.16. The Commissioner further contended that APC’s use of the wrong accounting method and its deduction of the truck transportation expenses constituted negligence, and it added COMMISSIONER v. ASPHALT PRODUCTS CO., INC. 119 117 Per Curiam to the deficiency a penalty under 26 U. S. C. § 6653(a)(1), which then provided: “If any part of any underpayment. . . is due to negligence or intentional disregard of rules or regulations (but without intent to defraud), there shall be added to the tax an amount equal to 5 percent of the underpayment.” (Section 6653 was amended in minor respects by the Tax Reform Act of 1986, Pub. L. 99-514, § 1503, 100 Stat. 2742; unless otherwise indicated, all further references are to the pre-1986 statute.) The penalty totaled $7,716.61—5% of the full alleged underpayment of $154,332.16. In the Tax Court, APC stipulated that the truck transportation expenses were not properly deductible, unsuccessfully contested the requirement that it use accrual accounting, and successfully contested certain other determinations, resulting in a recalculated deficiency of $133,248.69—almost all of which was due to the change in accounting methods. The Tax Court concluded that APC’s use of cash-basis accounting was nonnegligent, but affirmed the Commissioner’s finding that APC had negligently deducted the truck transportation expenses. It thus added to APC’s tax a negligence penalty of $6,943.37, computed as before by reference to the full amount of the deficiency (adjusted for carryback credits, see 26 U. S. C. §§6211, 6653(c)(1)). The Court of Appeals for the Sixth Circuit affirmed, over a dissent, the Tax Court’s determination that APC was required to use accrual accounting, and unanimously (albeit with little enthusiasm) affirmed the finding that the deduction for truck transportation expenses was negligent. 796 F. 2d 843 (1986). APC has petitioned for certiorari on those two issues in No. 86-1054, and we deny that petition. Accordingly, for purposes of this opinion we accept, without approving, the Commissioner’s finding of negligence. The Court of Appeals reversed the Tax Court’s imposition of the negligence penalty on the full amount of the deficiency, concluding that the penalty “should be applied only to that portion of the deficiency attributable to the disallowed deduc 120 OCTOBER TERM, 1986 Per Curiam 482 U. S. tion.” Id., at 850. The Commissioner has petitioned for certiorari on that issue in No. 86-1053. Because this holding is in apparent conflict with Abrams n. United States, 449 F. 2d 662 (CA2 1971), and is in obvious conflict with the plain language of the statute, we grant certiorari in No. 86-1053 and reverse. Section 6653(a)(1) could not be clearer. If “any part of any underpayment” is due to negligence, the Commissioner shall add to the tax a penalty of “5 percent of the underpayment.” It is impossible further to explain the statute without merely repeating its language—the penalty is imposed on “the underpayment,” not on the “part of [the] underpayment” attributable to negligence. By contrast (if contrast is thought necessary), the very next paragraph of the statute, § 6653(a)(2) (added in 1981, see Pub. L. 97-34, § 722(b)(1), 95 Stat. 342), limits the 50% penalty on interest due on negligent underpayments to “the portion of the underpayment . . . which is attributable to the [taxpayer’s] negligence.” The section imposing interest penalties on fraudulent underpayments contains the same proviso, § 6653(b)(2)(A), as does (after the 1986 Tax Reform Act) the provision for direct penalties on fraudulent underpayments. See § 1503(b)(1)(A), 100 Stat. 2742 (“If any part of any underpayment... is due to fraud, there shall be added to the tax ... 75 percent of the portion of the underpayment which is attributable to fraud”). As the Court of Appeals for the Second Circuit held in Abrams v. United States: “It is evident that it was intended that the five percent was to be assessed not just against that segment of the deficiency due to negligence but against the entire amount. The language is clear and leads to no other interpretation.” 449 F. 2d, at 664. The taxpayers in Abrams argued “that a literal application of the statute could lead to absurd results where a comparatively insignificant item of income is negligently omitted,” ibid., and the court in Abrams expressly reserved judgment on that situation. Ibid. (“That case is not before us on this COMMISSIONER v. ASPHALT PRODUCTS CO., INC. 121 117 Opinion of Marshall, J. appeal and we therefore express no opinion whatever as to its proper disposition if it should ever arise”). The Court of Appeals in this litigation relied on that reservation, and on the absence of any “egregious attempts [by APC] to avoid the payment of taxes,” 796 F. 2d, at 849, to distinguish Abrams, concluding that the Commissioner’s construction of the statute lets “the tail wag the dog.” 796 F. 2d, at 850; ibid. (Nelson, J., concurring in part and dissenting in part) (“Where the taxpayer is subject to a penalty only because of the negligent omission of a comparatively insignificant item of income, and the Commissioner also asserts that there is a large underpayment not claimed to be due to negligence, I agree with the court that it would be absurd to let the Commissioner calculate the negligence penalty by applying the statutory percentage to the sum of the negligent and non-negligent underpayments”). This was error. Judicial perception that a particular result would be unreasonable may enter into the construction of ambiguous provisions, but cannot justify disregard of what Congress has plainly and intentionally provided. Given the Government’s plausible interest in deterring negligent tax preparation and given the statute’s explicit limitation of other penalties to the amount of the negligent or fraudulent underpayment, no conclusion can be drawn from the provision here at issue except that Congress desired to impose a modest penalty (5%) upon underpayments any part of which was attributable to negligence of the taxpayer. It is not our assigned role to alter that disposition. The decision of the Court of Appeals limiting the amount of the negligence penalty is Reversed. Justice Marshall, concurring in part and dissenting in part. Once again the Court decides a case summarily without benefit of full briefing on the merits of the question decided. As I noted recently, Montana v. Hall, 481 U. S. 400, 405-406 (1987) (dissenting from summary disposition), this Court’s 122 OCTOBER TERM, 1986 Opinion of Marshall, J. 482 U. S. Rules governing the filing of petitions for certiorari instruct the parties to address whether plenary consideration of the case would be appropriate, and do not encourage detailed discussions of the merits. In this case, adhering to the admonition in this Court’s Rule 22.2 that a response be “as short as possible,” respondent filed a nine-page brief in opposition to the petition for certiorari, of which only four pages dealt with the issue of the proper construction of 26 U. S. C. § 6653(a)(1). It is, in my view, unfair to decide a case such as this without first permitting the litigants to brief in full the merits of the issues decided. The wisdom of summary disposition of this case is particularly doubtful. The legislative history of the Tax Reform Act of 1986, not mentioned by the Court, indicates that Congress considered carefully the scheme for imposing negligence penalties, see H. R. Conf. Rep. No. 99-841, pt. 2, pp. 779-782 (1986), and expressly disapproved the decision of the Court of Appeals in this case. Id., at 782, n. 3. Because Congress has definitively stated that the decision below is not to be followed in cases arising under the 1986 Act, this case, if left undisturbed, would have negligible precedential value. Moreover, courts considering the issue even with regard to tax years before 1987, while not bound by the current Congress’ view of the intent of a previous Congress, would probably pay some heed to the congressional view of the proper reading of § 6653(a)(1). * Under the circumstances it appears the reason for summarily reversing the judgment of the Court of Appeals in this case is simply that the majority perceives it to be wrong. But this Court routinely denies petitions for certiorari seeking review of decisions that, on the face of the petitions or the *Indeed, petitioner relies on this legislative history to support his contention that the decision of the Court of Appeals in this case was erroneous, Pet. for Cert. 6-8, and suggests that the Court vacate and remand the decision below for further consideration in light of the Conference Report. Id., at 8, n. 4. COMMISSIONER v. ASPHALT PRODUCTS CO., INC. 123 117 Opinion of Blackmun, J. petitions and responses, appear to be wrong. I can discern no intelligible principle distinguishing from that large number of cases those the Court chooses to decide without first giving the parties the opportunity to brief the merits of the case in full, and giving itself the opportunity to ensure that its initial impression is borne out by more thoughtful consideration. That our jurisdiction is discretionary should not lead us to be arbitrary in its exercise. I would not decide this case without first giving the parties the opportunity to file briefs on the merits. Accordingly, I dissent from the Court’s summary disposition in No. 86-1053. Because I too would deny the petition for certiorari in No. 86-1054, I concur in that part of the Court’s per curiam opinion. Justice Blackmun, concurring in part and dissenting in part. I agree with the Court in its denial of the petition for certiorari in No. 86-1054. I dissent from its summary reversal of the judgment of the Sixth Circuit on the negligence penalty issue. I do not agree that the correct result is so obvious and the Court of Appeals so clearly in error that summary reversal is warranted. I hope the Court’s action is not due to an innate reluctance to review a federal income tax case. After all, United States Courts of Appeals have reached conflicting conclusions on the issue, and income tax law often has its special vagaries. I would grant certiorari in No. 86-1053 and give that case plenary consideration. 124 OCTOBER TERM, 1986 Syllabus 482 U. S. TEXAS v. NEW MEXICO ON EXCEPTIONS TO REPORT OF SPECIAL MASTER No. 65, Orig. Argued April 29, 1987—Decided June 8, 1987 The 1949 Pecos River Compact between New Mexico and Texas divides the water of the Pecos River between the States, but, because of the river’s irregular flow, does not specify a particular amount of water to be delivered by New Mexico to Texas each year. Instead, Article 111(a) of the Compact provides that “New Mexico shall not deplete by man’s activities the flow of the Pecos River at the New Mexico-Texas state line below an amount which will give to Texas a quantity of water equivalent to that available to Texas under the 1947 condition.” In 1974, Texas filed this original action to resolve a dispute between the States with respect to the “1947 condition” and other matters. A Special Master was appointed, and this Court previously adopted his report specifying the methodology to be used in calculating Texas’ entitlement to water. The case is now before the Court on both parties’ exceptions to the Master’s recent report calculating the acre-feet shortfall of water that should have been delivered to Texas for the years 1950-1983, and recommending that, in addition to performing its ongoing obligation under the Compact, New Mexico be ordered to make up the accumulated shortfall by delivering a specified amount of water each year for 10 years, with a penalty in kind, i. e., “water interest,” for any bad-faith failure to deliver the additional amounts. Held: 1. Both parties’ exceptions with respect to the Master’s calculation of the shortfall that is chargeable to New Mexico are rejected. P. 128. 2. There is no merit to New Mexico’s contention that this Court may order only prospective relief and may not provide a remedy for past breaches of the Compact. Although a compact, when approved by Congress, becomes a law of the United States, it is still a contract, subject to construction and application in accordance with its terms. There is nothing in the nature of compacts generally or of the Pecos River Compact in particular that counsels against rectifying a failure to perform in the past as well as ordering future performance. Moreover, goodfaith differences (as here) about the scope of contractual undertakings do not relieve either party from performance. A court should provide a remedy if the contract’s terms provide a sufficiently certain basis for determining both that a breach has occurred and the nature of the remedy. Pp. 128-129. TEXAS v. NEW MEXICO 125 124 Syllabus 3. New Mexico contends that, in any event, it should be afforded the option of paying money damages for past shortages. Although the Master’s report noted that both sides would possibly be better off with monetary repayment, he concluded that the Compact, which does not specify a remedy in case of a breach, contemplates delivery of water and that this Court may not order relief inconsistent with the Compact’s terms. However, the Compact itself does not prevent the ordering of a suitable remedy, whether in water or money, and the Eleventh Amendment is no barrier to a monetary judgment, since that Amendment applies only to suits by citizens against a State. Any concern as to difficulties in enforcing judgments against States is insubstantial here, since if money damages were to be awarded, it would only be on the basis that if the sum awarded is not forthcoming in a timely manner, a judgment for repayment in water would be entered. This matter is returned to the Master for such further proceedings as he deems necessary and for his recommendations as to whether New Mexico should be allowed to elect a monetary remedy and, if so, the size of the payment and other terms that New Mexico must satisfy. Pp. 129-132. 4. A decree is entered with respect to New Mexico’s current and future obligation to deliver water pursuant to Article 111(a) of the Compact. Moreover, both the Master’s recommendation that, because applying the approved apportionment formula is not entirely mechanical and involves a degree of judgment, an additional enforcement mechanism be supplied, and his preferred solution—the appointment of a River Master to make the required periodic calculations—are accepted. This Court’s jurisdiction over original actions like this one provides it with ample authority to appoint such a master. On remand, the Special Master is requested to recommend an amendment to the decree, specifying as he deems necessary the River Master’s duties and the consequences of his determinations. Pp. 133-135. Exceptions to Special Master’s report sustained in part and overruled in part; decree entered. White, J., delivered the opinion of the Court, in which all other Members joined, except Stevens, J., who took no part in the consideration or decision of the case. Charlotte Uram argued the cause for defendant. With her on the briefs were Paul G. Bardacke, former Attorney General of New Mexico, Hal Stratton, Attorney General, and Peter Thomas White and Vickie L. Gabin, Special Assistant Attorneys General. 126 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. Renea Hicks, Assistant Attorney General of Texas, argued the cause for plaintiff. With him on the briefs were Jim Mattox, Attorney General, Mary F. Keller, Executive Assistant Attorney General, Nancy N. Lynch, and Paul Elliott, Assistant Attorney General.* Justice White delivered the opinion of the Court. This original case, which is here for the fourth time, involves the construction and enforcement of the 1949 Compact1 between New Mexico and Texas dividing the water of the Pecos River between the two States. Because of the irregular flow of the Pecos River, the Compact did not specify a particular amount of water to be delivered by New Mexico to Texas each year. Instead, Article 111(a) of the Compact provides that “New Mexico shall not deplete by man’s activities the flow of the Pecos River at the New Mexico-Texas state line below an amount which will give to Texas a quantity of water equivalent to that available to Texas under the 1947 condition.” Pecos River Compact, S. Doc. No. 109, 81st Congress, 1st Sess., Art. 111(a) (1949). The parties have had different views with respect to the “1947 condition” as well as other matters that could not be resolved through the Pecos River Commission, which Article V of the Compact established to carry out its provisions and which can effectively act only by mutual agreement of the two States.2 After years of relatively fruitless negotiation, Texas filed this original action in June 1974. We granted leave to file the *Briefs of amici curiae were filed for the incorporated municipality of Alamogordo, New Mexico, et al. by John B. Draper, Peter B. Shoenfeld, Neil C. Stillinger, and Michael G. Rosenberg; and for the Red Bluff Water Power Control District by Frank R. Booth. ^he Compact was signed by the States in 1948 and was approved by Congress in 1949. Article I, § 10, cl. 3, of the Constitution provides that “No State shall, without the Consent of Congress, . . . Compact with another State, or with a foreign Power . . . .” 2 The Commission is composed of a representative of each of the States and a third, but nonvoting, representative of the United States. TEXAS v. NEW MEXICO 127 124 Opinion of the Court complaint, 421 U. S. 927 (1975), and appointed a Special Master, 423 U. S. 942 (1975), the Honorable Jean Breitenstein, now deceased, who was then a judge of the Court of Appeals for the Tenth Circuit and a recognized expert in western water law. In 1979, the Special Master filed a report defining “the 1947 condition” and proposed a river routing study and adoption of a new inflow-outflow manual to be used in determining how much water Texas should be expected to receive over any particular period for any particular level of precipitation under the consumption conditions prevailing in New Mexico in 1947. We adopted that report in its entirety. 446 U. S. 540 (1980). When the case was next here, we decided against attempting to restructure the Commission to enable it to determine the method for allocating river water, preferring that the case continue in the litigation mode.3 462 U. S. 554 (1983). On June 11, 1984, we summarily approved the Special Master’s report specifying the inflow-outflow methodology to be used in calculating Texas’ entitlement.4 467 U. S. 1238. Special Master Charles Meyers, Judge Breitenstein’s successor, 468 U. S. 1202 (1984), then held hearings on the question whether New Mexico had fulfilled its obligation under Article 111(a) of the Compact. He issued a report containing his findings and conclusion that for the years 1950-1983, New Mexico should have delivered 340,100 acre-feet more water at the state line than Texas had received over those years. The Master recommended that in addition to performing its ongoing obligation under the Compact, New Mexico be ordered to make up the accumulated shortfall by delivering 3 We also rejected the submission of Texas that “Double Mass Analysis” rather than “Inflow-Outflow” be adopted as the method for determining New Mexico’s delivery obligation under the Compact. 4 The Special Master recommended and we agreed that Figure 1 and Table 1 of Texas Exhibit 68, pp. 3, 4, properly described the method to be used. 128 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. 34,010 acre-feet of water each year for 10 years, with a penalty in kind, i. e., “water interest,” for any bad-faith failure to deliver these additional amounts. Both sides excepted to the Master’s report, and we have heard oral argument. We find no merit in and reject the exceptions filed by Texas and New Mexico with respect to the Master’s calculation of the shortfall that is chargeable to New Mexico.5 New Mexico also excepts to the proposed remedy for the short deliveries in past years. We find no merit in its submission that we may order only prospective relief, that is, requiring future performance of compact obligations without a remedy for past breaches. If that were the case, New Mexico’s defaults could never be remedied. This was not our approach when the case was here in 1983. We then affirmed our authority to hear and decide Texas’ claim and remanded the case to the Master for a determination of the shortfall. As we said then, a compact when approved by Congress becomes a law of the United States, 462 U. S., at 564, but “[a] Compact is, after all, a contract.” Petty n. Tennessee-Missouri Bridge Common, 359 U. S. 275, 285 (1959) (Frankfurter, J., dissenting). It remains a legal document that must be construed and applied in accordance with its terms. West Virginia ex rel. Dyer v. Sims, 341 U. S. 22, 28 (1951); 462 U. S., at 564. There is nothing in the nature of compacts generally or of this Compact in particular that counsels against rectifying a failure to perform in the past as well as ordering future performance called for by the Compact. By ratifying the Constitution, the States gave this Court complete judicial power to adjudicate disputes among them, Rhode Island n. Massachusetts, 12 Pet. 657, 720 (1838), and this power includes the capacity to provide one State a remedy for the breach of another. 6 New Mexico’s ongoing obligation under Article 111(a) of the Compact, as now construed and applied by this Court, will be on the average 10,000 acre-feet higher than New Mexico’s deliveries have been in the past. TEXAS v. NEW MEXICO 129 124 Opinion of the Court New Mexico, however, argues that it has no obligation to deliver water that it, in good faith, believed it had no obligation to refrain from using. It is true that Texas and New Mexico have been at odds on the interpretation of the Compact and that their respective views have not been without substantial foundation. Both Special Masters recognized that New Mexico acted in good faith, and as Judge Breitenstein said in his 1982 report, New Mexico’s “obligation is still uncertain because the definition of the 1947 condition must be translated into water quantities to provide a numerical standard.” Report of Special Master 18. The basic meaning of the 1947 condition was not defined until 1979 in the course of this litigation; and a workable methodology for translating New Mexico’s obligation into quantities of water was not achieved until 1984, also in this litigation.6 But good-faith differences about the scope of contractual undertakings do not relieve either party from performance. A court should provide a remedy if the parties intended to make a contract and the contract’s terms provide a sufficiently certain basis for determining both that a breach has in fact occurred and the nature of the remedy called for. Restatement (Second) of Contracts § 33(2), and Comment b (1981). There is often a retroactive impact when courts resolve contract disputes about the scope of a promisor’s undertaking; parties must perform today or pay damages for what a court decides they promised to do yesterday and did not. In our view, New Mexico cannot escape liability for what has been adjudicated to be past failures to perform its duties under the Compact. New Mexico submits that in the event Texas is found to be entitled to a remedy for the past shortages now ascertained, it should be afforded the option of paying money damages rather than paying in kind. New Mexico’s Exceptions to the Report of the Special Master 40-41. This possibility was discussed to some extent in hearings before the Master, who 6 The Inflow-Outflow Manual incorporated in the Compact proved to be so faulty as to be unusable. 130 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. more than once stated that damages might be best for both parties. New Mexico “stipulated” that if relief was to be awarded, damages were preferable. Tr. of Hearing Before the Special Master 94 (Apr. 16, 1985). The Special Master’s report also states that both sides would possibly be better off with monetary repayment but refers to difficulties suggested by counsel and observes that the Compact contains no explicit provision for monetary relief. The Master concluded that the Compact contemplated delivery of water and that the Court could not order relief inconsistent with the Compact terms. The State of Texas supports the Master’s view. The Special Master was rightfully cautious, but the lack of specific provision for a remedy in case of breach does not, in our view, mandate repayment in water and preclude damages. Nor does our opinion in 462 U. S. 554 (1983), necessarily foreclose such relief. There, we asserted our authority in this original action to resolve the case judicially, rather than by restructuring the administrative mechanism established by the Compact. That authority extended to devising a method by which New Mexico’s obligation could be ascertained and then quantifying New Mexico’s past obligation, as the Master has now done. We have now agreed with him that New Mexico has not fully performed, and we are quite sure that the Compact itself does not prevent our ordering a suitable remedy, whether in water or money. The Court has recognized the propriety of money judgments against a State in an original action, South Dakota v. North Carolina, 192 U. S. 286 (1904); United States v. Michigan, 190 U. S. 379 (1903); and specifically in a case involving a compact, Virginia n. West Virginia, 246 U. S. 565 (1918). In proper original actions, the Eleventh Amendment is no barrier, for by its terms, it applies only to suits by citizens against a State. Maryland n. Louisiana, 451 U. S. 725, 745, n. 21 (1981); United States v. Mississippi, 380 U. S. 128, 140 (1965); South Dakota v. North Carolina, supra. That there may be difficulties in enforcing judgments against TEXAS v. NEW MEXICO 131 124 Opinion of the Court States counsels caution but does not undermine our authority to enter judgments against defendant States in cases over which the Court has undoubted jurisdiction, authority that is attested to by the fact that almost invariably the “States against which judgments were rendered, conformably to their duty under the Constitution, voluntarily respected and gave effect to the same.” Virginia v. West Virginia, supra, at 592. In any event, that concern is insubstantial here, for if money damages were to be awarded, it would only be on the basis that if the sum awarded is not forthcoming in a timely manner, a judgment for repayment in water would be entered. As we understand the Master, he did not pursue the matter of monetary relief because he thought it foreclosed by the Compact, not because he thought it inadequate, unfair, or impractical. As we have said, the issue was raised in the hearings, but the record does not permit a confident judgment as to whether a remedy in money, rather than water, would be equitable or feasible. To order making up the shortfalls by delivering more water has all the earmarks of specific performance, an equitable remedy that requires some attention to the relative benefits and burdens that the parties may enjoy or suffer as compared with a legal remedy in damages. “[S]pecific performance is never demandable as a matter of absolute right, but as one which rests entirely in judicial discretion, to be exercised, it is true, according to the settled principles of equity, but not arbitrarily and capriciously, and always with reference to the facts of the particular case.” Haffner n. Dobrinski, 215 U. S. 446, 450 (1910). Specific performance will not be compelled “if under all the circumstances it would be inequitable to do so.” Wesley v. Eells, 177 U. S. 370, 376 (1900). It might be said that those users who have suffered the water shortages caused by New Mexico’s underdeliveries over the years, rather than the State, should be the recipients of damages, and that they would be difficult if not impos 132 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. sible to identify. But repayment in water would also likely fail to benefit all those who were deprived in the past.7 It might also be said that awarding only a sum of money would permit New Mexico to ignore its obligation to deliver water as long as it is willing to suffer the financial penalty. But in light of the authority to order remedying shortfalls to be made up in kind, with whatever additional sanction might be thought necessary for deliberate failure to perform, that concern is not substantial in our view. We conclude that the matter of remedying past shortages should be returned to the Special Master for such further proceedings as he deems necessary and for his ensuing recommendation as to whether New Mexico should be allowed to elect a monetary remedy and, if so, to suggest the size of the payment and other terms that New Mexico must satisfy.8 ’Texas counsel suggested that a money judgment might find its way into the general coffers of the State, rather than benefit those who were hurt. But the basis on which Texas was permitted to bring this original action is that enforcement of the Compact was of such general public interest that the sovereign State was a proper plaintiff. See Maryland v. Louisiana, 451 U. S. 725, 735-739 (1981). It is wholly consistent with that view that the State should recover any damages that may be awarded, money it would be free to spend in the way it determines is in the public interest. 8 If the Special Master recommends and we approve a judgment for money damages, Texas will be entitled to postjudgment interest until the judgment is paid. If damages are not awarded or a damages judgment is not paid, it would appear it would be necessary to make up the shortfall by delivering more water over a period of years as the Master has recommended in his report. In that event, Texas would have a judgment against New Mexico for 340,100 acre-feet of water, plus any additional net shortfalls accruing to the date hereof, which, if not delivered as ordered by the Court, would entitle Texas to apply to this Court for enforcement, cf. Wyoming v. Colorado, 309 U. S. 572, 573 (1940), and to some form of postjudgment interest for the period during which that judgment is not satisfied. We are unpersuaded, however, that “water interest,” rather than money, should be awarded unless and until it proves to be necessary. New Mexico submits that there is no statutory authority for this Court to allow postjudgment interest in any form and that we are therefore with TEXAS v. NEW MEXICO 133 124 Opinion of the Court Meanwhile, a decree in the form discussed below will issue with respect to New Mexico’s current and future obligation to deliver water pursuant to Article 111(a) of the Pecos River Compact as interpreted and applied by the judgments of this Court. The attached decree enjoins New Mexico to comply with its Article 111(a) obligation under the Pecos River Compact and to determine the extent of its obligation in accordance with the formula approved by the decisions of this Court. That formula was fashioned in the course of this litigation, which was occasioned by the inability of the Pecos River Commission, on which Texas and New Mexico have the only votes, to agree on how river water should be divided. Neither this opinion nor the decree, however, displaces the authority of the Commission to perform what it has not been able to perform before, namely, an agreed upon and mutually satisfactory formula for division and utilization of Pecos River water. If history repeats itself, the Commission will not come forth with an apportionment different from that which the Court has now approved. If it does, the parties should petition the Court to terminate or appropriately modify its decree as the case may be. Even if the Commission takes no action, it may be that because of the unpredictability and peculiarities of the Pecos, the inflow-outflow methodology we have ordered implemented will not reflect the realities of the river. In that event, it would be appropriate to seek an amendment of the decree, as has been done in other original actions. out power to do so in this original action. It relies on the statement in Pierce v. United States, 255 U. S. 398, 406 (1921), that postjudgment interest may not be awarded absent statutory authority. But we are not bound by this rule in exercising our original jurisdiction. In Virginia v. West Virginia, 238 U. S. 202, 242 (1915), the Court awarded interest on its judgment, an action consistent with express statutory authority for other federal courts to award postjudgment interest. 134 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. The decree now issued goes no further, but the Master has recommended that because applying the approved apportionment formula is not entirely mechanical and involves a degree of judgment, an additional enforcement mechanism be supplied. We accept his recommendation and also his preferred solution: the appointment of a River Master to make the required periodic calculations. In 1983, because we thought the Compact foreclosed it, we declined to order a tie breaker in order that the Commission itself could arrive at a method to allocate water. We accordingly proceeded in the litigative mode to construe and enforce the Compact, asserting our authority to do so in unequivocal terms. We have arrived at what we deem to be a fair and equitable solution that is consistent with the Compact terms, and we are quite sure that our jurisdiction over original actions like this provides us with ample authority to appoint a master and to enforce our judgment. Virginia v. West Virginia, 246 U. S., at 591. In exercising this power, we have taken a distinctly jaundiced view of appointing an agent or functionary to implement our decrees. Vermont v. New York, 417 U. S. 270 (1974), emphatically expressed this reluctance. But as we recognized, id., at 275-276, that solution, or a like one, has been employed when the occasion demands. New Jersey n. New York, 283 U. S. 805 (1931); Wisconsin v. Illinois, 281 U. S. 179 (1930). This is one of those occasions when such a mechanism should be employed. The natural propensity of these two States to disagree if an allocation formula leaves room to do so cannot be ignored. Absent some disinterested authority to make determinations binding on the parties, we could anticipate a series of original actions to determine the periodic division of the water flowing in the Pecos. A River Master should therefore be appointed to make the calculations provided for in this decree, annually and as promptly as possible as data are available, and to report the calculations TEXAS v. NEW MEXICO 135 124 Opinion of the Court to appropriate representatives of New Mexico and of Texas. His calculations will include determinations of negative or positive departures from New Mexico’s delivery obligation and such shortfalls or credits will be reflected in that State’s later delivery obligations. Provision for a River Master will occasion an amendment to the decree. On remand, the Special Master is requested to recommend an amendment to the decree, specifying as he deems necessary the duties of the River Master and the consequences of his determinations. Any other suggestions for amendments should also be called to our attention. The River Master’s compensation shall be borne equally by the parties. The parties, as well as the Special Master, are welcome to suggest candidates for appointment as River Master.9 DECREE It is Ordered, Adjudged, and Decreed that the State of New Mexico, its officers, attorneys, agents, and employees are hereby enjoined: (A) To comply with the Article 111(a) obligation of the Pecos River Compact by delivering to Texas at state line each year an amount of water calculated in accordance with the inflow-outflow equation contained in Texas Exhibit 68, at page 2. (B) To calculate the Index Inflow component of the inflow-outflow and channel-loss equations contained in Texas Exhibit 79, modified to reflect the Court’s decision of June 8, 1987, as to manmade depletions chargeable to New Mexico. “Index Inflow” shall mean the 3-year progressive average of “annual flood inflows” as those terms are defined in Texas Exhibit 79, Table 2, p. 5. 9 Justice Stevens took no part in the consideration or decision of this case. 136 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. The Court retains jurisdiction of this suit for the purpose of any order, direction, or modification of the decree, or any supplementary decree, that may at any time be deemed proper in relation to the subject matter in controversy. BOWEN v. YUCKERT 137 Syllabus BOWEN, SECRETARY OF HEALTH AND HUMAN SERVICES v. YUCKERT ON WRIT OF CERTIORARI TO THE UNITED STATES COURT OF APPEALS FOR THE NINTH CIRCUIT No. 85-1409. Argued January 13, 1987—Decided June 8, 1987 The Social Security Act (Act) defines “disability” as the “inability to engage in any substantial gainful activity by reason of any medically determinable physical or mental impairment . . . .” 42 U. S. C. § 423(d)(1)(A). The Act also provides that an individual “shall be determined to be under a disability only if his . . . impairment [is] of such severity that he is not only unable to do his previous work but cannot, considering his age, education, and work experience, engage in any other . . . substantial gainful work . . . .” § 423(d)(2)(A). The Secretary of Health and Human Services (Secretary) has established a five-step sequential evaluation process for determining whether a person is disabled. In step two of that process, the “severity regulation” provides: “If you do not have any impairment . . . which significantly limits your . . . ability to do basic work activities, we will find that you do not have a severe impairment and are, therefore, not disabled. We will not consider your age, education, and work experience.” Respondent applied for disability benefits, but the appropriate state agency determined that she was not disabled. In light of the medical evidence and evidence of her activities, the Social Security Administration (SSA) Administrative Law Judge concluded that her medically determinable impairments were not severe under the severity regulation, and thé SSA’s Appeals Council denied her request for review. The Federal District Court affirmed, but the Court of Appeals reversed and remanded, holding that the Act does not authorize benefits denials based solely on a determination that the claimant is not severely impaired, and that § 423(d)(2)(A) requires that both medical and vocational factors such as age, education, and work experience be considered in determining disability. The court rejected the Secretary’s contention that the 1984 amendments to the Act endorsed step two of the disability evaluation process, and invalidated the severity regulation. Held: 1. The severity regulation is valid on its face under the language of the Act and the legislative history. Pp. 142-152. (a) The severity regulation is not inconsistent with § 423(d)(1)(A), which defines “disability” in terms of the effect an impairment has on a 138 OCTOBER TERM, 1986 Syllabus 482 U. S. person’s ability to function in the workplace. The regulation adopts precisely this functional approach to determining the effects of medical impairments, when it requires the claimant to show that he has an “impairment . . . which significantly limits” “the abilities and aptitudes necessary to do most jobs.” If the impairment is not severe enough to so limit the claimant, by definition it does not prevent the claimant from engaging in any substantial gainful activity. Moreover, § 423(d)(5)(A) expressly gives the Secretary the authority to place the burden of showing a medically determinable impairment on the claimant. The requirement of a threshold showing of severity also is consistent with the legislative history of § 423(d)(1)(A). Pp. 146-147. (b) The severity regulation is not inconsistent with § 423(d)(2)(A), which restricts disability benefit eligibility to claimants whose medically severe impairments prevent them from doing their previous work and any other substantial gainful work in the national economy. If a claimant is unable to show that he has a medically severe impairment, he is not eligible for benefits, and there is no reason for the Secretary to consider his age, education, and work experience. The legislative history reinforces this understanding of the statutory language. Pp. 147-149. (c) In enacting § 4(a)(1) of the Social Security Disability Benefits Reform Act of 1984, 42 U. S. C. § 423(d)(2)(C), Congress expressed its approval of the severity regulation both in the statute and in the accompanying Reports, recognizing that the Secretary may make an initial determination of medical severity, and that he need not consider the claimant’s age, education, and experience unless he finds “a medically severe combination of impairments.” Pp. 149-152. 2. The severity regulation increases the efficiency and reliability of the disability evaluation process by identifying at an early stage those claimants whose medical impairments are so slight that it is unlikely they would be found to be disabled even if their age, education, and experience were taken into account. Pp. 153-154. 774 F. 2d 1365, reversed and remanded. Powell, J., delivered the opinion of the Court, in which Rehnquist, C. J., and White, Stevens, O’Connor, and Scalia, JJ., joined. O’Connor, J., filed a concurring opinion, in which Stevens, J., joined, post, p. 155. Blackmun, J., filed a dissenting opinion, in which Brennan and Marshall, JJ., joined, post, p. 159. Edwin S. Kneedler argued the cause for petitioner. With him on the briefs were Solicitor General Fried, Assistant Attorney General Willard, and Deputy Solicitor General Wallace. BOWEN v. YUCKERT 139 137 Opinion of the Court Carole F. Grossman argued the cause for respondent. With her on the brief were James A. Douglas and Peter Komlos-Hrobsky. * Justice Powell delivered the opinion of the Court. The question in this case is whether the Secretary of Health and Human Services may deny a claim for Social Security disability benefits on the basis of a determination that the claimant does not suffer from a medically severe impairment that significantly limits the claimant’s ability to perform basic work activities. *Briefs of amici curiae urging affirmance were filed for the State of Alabama et al. by Robert Abrams, Attorney General of New York, 0. Peter Sherwood, Solicitor General, Charles A. Graddick, Attorney General of Alabama, Harold M. Brown, Attorney General of Alaska, John Steven Clark, Attorney General of Arkansas, Duane Woodard, Attorney General of Colorado, Jim Smith, Attorney General of Florida, Corinne K. A. Watanabe, Attorney General of Hawaii, Neil F. Hartigan, Attorney General of Illinois, and Roma J. Stewart, Solicitor General, Linley E. Pearson, Attorney General of Indiana, Thomas J. Miller, Attorney General of Iowa, Robert T. Stephan, Attorney General of Kansas, David L. Armstrong, Attorney General of Kentucky, William J. Guste, Jr., Attorney General of Louisiana, Stephen H. Sachs, Attorney General of Maryland, Frank J. Kelley, Attorney General of Michigan, Edwin Lloyd Pittman, Attorney General of Mississippi, Robert M. Spire, Attorney General of Nebraska, W Cary Edwards, Attorney General of New Jersey, Paul Bardacke, Attorney General of New Mexico, Lacy H. Thornburg, Attorney General of North Carolina, Nicholas J. Spaeth, Attorney General of North Dakota, Anthony J. Celebrezze, Jr., Attorney General of Ohio, Michael C. Turpen, Attorney General of Oklahoma, LeRoy S. Zimmerman, Attorney General of Pennsylvania, Mark V. Meierhenry, Attorney General of South Dakota, Jim Mattox, Attorney General of Texas, Jeffrey L. Amestoy, Attorney General of Vermont, Bronson C. La Follette, Attorney General of Wisconsin, and A. G. McClintock, Attorney General of Wyoming; for the city of New York et al. by Frederick A. 0. Schwarz, Jr., Leonard Koerner, Michael D. Young, Julie Downey, Jessica Heinz, and Judson H. Miner; for the American Association of Retired Persons by Alfred Miller; and for the American Diabetes Association et al. by Frederick M. Stanczak, Richard E. Yaskin, Kalman Finkel, John E. Kirklin, Nancy Morawetz, Robert E. Lehrer, Joseph A. Antolin, and Shelley Davis. 140 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. I Title II of the Social Security Act (Act), 49 Stat. 620, as amended, provides for the payment of insurance benefits to persons who have contributed to the program and who suffer from a physical or mental disability. 42 U. S. C. § 423(a)(1)(D) (1982 ed., Supp. III). Title XVI of the Act provides for the payment of disability benefits to indigent persons under the Supplemental Security Income (SSI) program. § 1382(a). Both titles of the Act define “disability” as the “inability to engage in any substantial gainful activity by reason of any medically determinable physical or mental impairment which can be expected to result in death or which has lasted or can be expected to last for a continuous period of not less than 12 months . . . .” §423(d)(1)(A). See § 1382c(a)(3)(A). The Act further provides that an individual “shall be determined to be under a disability only if his physical or mental impairment or impairments are of such severity that he is not only unable to do his previous work but cannot, considering his age, education, and work experience, engage in any other kind of substantial gainful work which exists in the national economy, regardless of whether such work exists in the immediate area in which he lives, or whether a specific job vacancy exists for him, or whether he would be hired if he applied for work.” §§ 423(d)(2)(A), 1382c(a)(3)(B) (1982 ed. and Supp. III). The Secretary has established a five-step sequential evaluation process for determining whether a person is disabled. 20 CFR §§404.1520, 416.920 (1986). Step one determines whether the claimant is engaged in “substantial gainful activity.” If he is, disability benefits are denied. §§ 404.1520(b), 416.920(b). If he is not, the decisionmaker proceeds to step two, which determines whether the claimant has a medically BOWEN v. YUCKERT 141 137 Opinion of the Court severe impairment or combination of impairments. That determination is governed by the “severity regulation” at issue in this case. The severity regulation provides: “If you do not have any impairment or combination of impairments which significantly limits your physical or mental ability to do basic work activities, we will find that you do not have a severe impairment and are, therefore, not disabled. We will not consider your age, education, and work experience.” §§404.1520(c), 416.920(c). The ability to do basic work activities is defined as “the abilities and aptitudes necessary to do most jobs.” §§404. 1521(b), 416.921(b). Such abilities and aptitudes include “[p]hysical functions such as walking, standing, sitting, lifting, pushing, pulling, reaching, carrying, or handling”; “[c]a-pacities for seeing, hearing, and speaking”; “[u]nderstanding, carrying out, and remembering simple instructions”; “[u]se of judgment”; “[r]esponding appropriately to supervision, coworkers, and usual work situations”; and “[d]ealing with changes in a routine work setting.” Ibid. If the claimant does not have a severe impairment or combination of impairments, the disability claim is denied. If the impairment is severe, the evaluation proceeds to the third step, which determines whether the impairment is equivalent to one of a number of listed impairments that the Secretary acknowledges are so severe as to preclude substantial gainful activity. §§404.1520(d), 416.920(d); 20 CFR pt. 404, subpt. P, App. 1 (1986). If the impairment meets or equals one of the listed impairments, the claimant is conclusively presumed to be disabled. If the impairment is not one that is conclusively presumed to be disabling, the evaluation proceeds to the fourth step, which determines whether the impairment prevents the claimant from performing work he has performed in the past. If the claimant is able to perform his previous work, he is not disabled. §§404.1520(e), 142 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. 416.920(e). If the claimant cannot perform this work, the fifth and final step of the process determines whether he is able to perform other work in the national economy in view of his age, education, and work experience. The claimant is entitled to disability benefits only if he is not able to perform other work. §§ 404.1520(f), 416.920(f). The initial disability determination is made by a state agency acting under the authority and supervision of the Secretary. 42 U. S. C. §§ 421(a), 1383b(a); 20 CFR §§ 404.1503, 416.903 (1986). If the state agency denies the disability claim, the claimant may pursue a three-stage administrative review process. First, the determination is reconsidered de novo by the state agency. §§ 404.909(a), 416.1409(a). Second, the claimant is entitled to a hearing before an administrative law judge (ALJ) within the Bureau of Hearings and Appeals of the Social Security Administration. 42 U. S. C. §§ 405(b)(1), 1383(c)(1) (1982 ed. and Supp. Ill); 20 CFR §§404.929, 416.1429, 422.201 et seq. (1986). Third, the claimant may seek review by the Appeals Council. 20 CFR §§404.967 et seq., 416.1467 et seq. (1986). Once the claimant has exhausted these administrative remedies, he may seek review in federal district court. 42 U. S. C. § 405(g). See generally Bowen v. City of New York, 476 U. S. 467, 472 (1986). II Respondent Janet Yuckert applied for both Social Security disability insurance benefits and SSI benefits in October 1980. She alleged that she was disabled by an inner ear dysfunction, dizzy spells, headaches, an inability to focus her eyes, and flatfeet. Yuckert had been employed as a travel agent from 1963 to 1977. In 1978 and 1979, she had worked intermittently as a real estate salesperson. Yuckert was 45 years old at the time of her application. She has a high school education, two years of business college, and real estate training. BOWEN v. YUCKERT 143 137 Opinion of the Court The Washington Department of Social and Health Services determined that Yuckert was not disabled. The agency reconsidered Yuckert’s application at her request, and again determined that she was not disabled. At the next stage of the administrative review process, the ALJ found that, although Yuckert suffered from “episodes of dizziness, or vision problems,” App. to Pet. for Cert. 28a, “[m]ultiple tests . . . failed to divulge objective clinical findings of abnormalities that support the claimant’s severity of the stated impairments.” Id., at 27a.1 The ALJ also found that Yuckert was pursuing a “relatively difficult” 2-year course in computer programming at a community college and was able to drive her car 80 to 90 miles each week. Id., at 27a-28a. In light of the medical evidence and the evidence of her activities, the ALJ concluded that her medically determinable impairments were not severe under 20 CFR §§404.1520(c) and 416.920(c) (1986). The Appeals Council denied Yuckert’s request for review on the ground that the results of additional psychological tests supported the ALJ’s finding that she had not suffered a significant impairment of any work-related abilities. App. to Pet. for Cert. 22a. Yuckert then sought review in the United States District Court for the Western District of Washington. The case was referred to a Magistrate, who concluded that the Secretary’s determination was supported by substantial evidence. The District Court adopted the Magistrate’s report and affirmed the denial of Yuckert’s claim. Id., at 14a. The United States Court of Appeals for the Ninth Circuit reversed and remanded without considering the substantiality of the evidence. Yuckert v. Heckler, 774 F. 2d 1365, 1370 (1985). The court held that the Act does not authorize 'Yuckert’s physician diagnosed her condition as bilateral labyrinthine dysfunction. App. to Pet. for Cert. 26a. Another physician found only ‘non-specific congestion of the nasal and middle ear mucous membranes.” Ibid. X rays, an electrocardiogram, and a spinal puncture revealed no abnormalities, id., at 27a. 144 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. the Secretary to deny benefits on the basis of a determination that the claimant is not severely impaired. The court focused on the statutory provision that a person is disabled “only if his physical or mental impairment or impairments are of such severity that he is not only unable to do his previous work but cannot, considering his age, education, and work experience, engage in any other kind of substantial gainful work . . . .” 42 U. S. C. §423(d)(2)(A) (1982 ed. and Supp. III). In the court’s view, this provision requires that “both medical and vocational factors [i. e., age, education, and work experience] be considered in determining disability.” Yuckert v. Heckler, 774 F. 2d, at 1370. The court rejected the Secretary’s contention that the 1984 amendments to the Act endorsed step two of the disability evaluation process. The court concluded that “[t]he legislative history does not suggest that Congress intended to permit findings of nondisability based on medical factors alone.” Ibid, (citation omitted). Finally, the court relied upon Court of Appeals holdings that the burden of proof shifts to the Secretary once the claimant shows an inability to perform his previous work.2 In the court’s view, step two of the Secretary’s evaluation process is inconsistent with this assignment of burdens of proof, because it allows the Secretary to deny benefits to a claimant who is unable to perform past work without requiring the Secretary to show that the claimant can perform other work. Accordingly, the court invalidated the severity regulation, 20 CFR §404.1520(c) (1986).3 Because of the importance of the issue, and because the court’s decision con 2E.g., Valencia v. Heckler, 751 F. 2d 1082, 1086-1087 (CA9 1985); Francis v. Heckler, 749 F. 2d 1562, 1564 (CA11 1985). 3 Although Yuckert had applied for SSI benefits as well as disability insurance benefits, the complaint she filed in District Court referred only to the disability insurance program of Title II. Accordingly, the Court of Appeals did not invalidate 20 CFR § 416.920(c) (1986), the severity regulation applicable to the SSI program. BOWEN v. YUCKERT 145 137 Opinion of the Court flicts with the holdings of other Courts of Appeals,4 we granted certiorari. 476 U. S. 1114 (1986). We now reverse. Ill Our prior decisions recognize that “Congress has ‘conferred on the Secretary exceptionally broad authority to prescribe standards for applying certain sections of the Act.’” Heckler v. Campbell, 461 U. S. 458, 466 (1983) (quoting Schweiker v. Gray Panthers, 453 U. S. 34, 43 (1981)). The Act authorizes the Secretary to “adopt reasonable and proper rules and regulations to regulate and provide for the nature and extent of the proofs and evidence and the method of taking and furnishing the same” in disability cases. 42 U. S. C. § 405(a). We have held that “[w]here, as here, the statute expressly entrusts the Secretary with the responsibility for implementing a provision by regulation, our review is limited to determining whether the regulations promulgated exceeded the Secretary’s statutory authority and whether they are arbitrary and capricious.” Heckler n. Campbell, supra, at 466 (footnote and citations omitted). In our view, both the language of the Act and its legislative history support the Secretary’s decision to require disability claimants to make a threshold showing that their “medically determinable” impairments are severe enough to satisfy the regulatory standards. 4 Some Courts of Appeals have upheld the facial validity of the severity regulation. McDonald v. Secretary of Health and Human Services, 795 F. 2d 1118, 1121-1126 (CAI 1986); Hampton v. Bowen, 785 F. 2d 1308, 1311 (CA5 1986); Farris v. Secretary of Health and Human Services, 773 F. 2d 85, 89-90 (CA6 1985); Flynn v. Heckler, 768 F. 2d 1273, 1274-1275 (CA11 1985) (per curiam). Others have joined the Court of Appeals for the Ninth Circuit in holding the severity regulation invalid on its face. Wilson v. Secretary of Health and Human Services, 796 F. 2d 36, 40-42 (CA3 1986); Johnson v. Heckler, 769 F. 2d 1202, 1209-1213 (CA7 1985); Brown v. Heckler, 786 F. 2d 870, 871-872 (CA8 1986); Hansen n. Heckler, 783 F. 2d 170, 174-176 (CAIO 1986). 146 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. A As noted above, the Social Security Amendments Act of 1954 defined “disability” as “inability to engage in any substantial gainful activity by reason of any medically determinable physical or mental impairment . . . .” 68 Stat. 1080, 42 U. S. C. § 423(d)(1)(A). The severity regulation requires the claimant to show that he has an “impairment or combination of impairments which significantly limits” “the abilities and aptitudes necessary to do most jobs.” 20 CFR §§404.1520(c), 404.1521(b) (1986). On its face, the regulation is not inconsistent with the statutory definition of disability. The Act “defines ‘disability’ in terms of the effect a physical or mental impairment has on a person’s ability to function in the workplace.” See Heckler n. Campbell, supra, at 459-460. The regulation adopts precisely this functional approach to determining the effects of medical impairments. If the impairments are not severe enough to limit significantly the claimant’s ability to perform most jobs, by definition the impairment does not prevent the claimant from engaging in any substantial gainful activity. The Secretary, moreover, has express statutory authority to place the burden of showing a medically determinable impairment on the claimant. The Act provides that “[a]n individual shall not be considered to be under a disability unless he furnishes such medical and other evidence of the existence thereof as the Secretary may require.” § 423(d)(5)(A) (1982 ed. and Supp. III). See Mathews n. Eldridge, 424 U. S. 319, 336 (1976).5 6 The severity regulation does not change the settled allocation of burdens of proof in disability proceedings. It is true, as Yuckert notes, that the Secretary bears the burden of proof at step five, which determines whether the claimant is able to perform work available in the national economy. But the Secretary is required to bear this burden only if the sequential evaluation process proceeds to the fifth step. The claimant first must bear the burden at step one of showing that he is not working, at step two that he has a medically severe impairment or combination of impairments, and at step four that the impairment prevents him from performing his past work. If the process ends at step two, the burden of proof never BOWEN v. YUCKERT 147 137 Opinion of the Court The requirement of a threshold showing of severity also is consistent with the legislative history of § 423(d)(1)(A). The Senate Report accompanying the 1954 Amendments states: “The physical or mental impairment must be of a nature and degree of severity sufficient to justify its consideration as the cause of failure to obtain any substantial gainful work. Standards for evaluating the severity of disabling conditions will be worked out in consultation with the State agencies.” S. Rep. No. 1987, 83d Cong., 2d Sess., 21 (1954). House Rep. No. 1698, 83d Cong., 2d Sess., 23 (1954), contains virtually identical language. Shortly after the 1954 Amendments were enacted, the Secretary promulgated a regulation stating that “medical considerations alone may justify a finding that the individual is not under a disability where the only impairment is a slight neurosis, slight impairment of sight or hearing, or other similar abnormality or combination of slight abnormalities.” 20 CFR §404.1502(a) (1961). This regulation, with minor revisions, remained in effect until the sequential evaluation regulations were promulgated in 1978. B The Court of Appeals placed little weight on § 423(d)(1)(A) or its legislative history, but concluded that the severity regulation is inconsistent with § 423(d)(2)(A). We find no basis for this holding. Section 423(d)(2)(A), set forth supra, at 140, was enacted as part of the Social Security Amend- shifts to the Secretary. Similarly, if the impairment is one that is conclusively presumed to be disabling, the claimant is not required to bear the burden of showing that he is unable to perform his prior work. See Bluv-band v. Heckler, 730 F. 2d 886, 891 (CA2 1984). This allocation of burdens of proof is well within the Secretary’s “exceptionally broad authority” under the statute. Schweiker v. Gray Panthers, 453 U. S. 34, 43 (1981). It is not unreasonable to require the claimant, who is in a better position to provide information about his own medical condition, to do so. 148 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. ments of 1967, 81 Stat. 868. It states that “an individual . . . shall be determined to be under a disability only if his physical or mental impairment or impairments are of such severity that he is not only unable to do his previous work but cannot, considering his age, education, and work experience, engage in any other kind of substantial gainful work . . . .” Ibid. The words of this provision limit the Secretary’s authority to grant disability benefits, not to deny them.6 Section 423(d)(2)(A) restricts eligibility for disability benefits to claimants whose medically severe impairments prevent them from doing their previous work and also prevent them from doing any other substantial gainful work in the national economy. If a claimant is unable to show that he has a medically severe impairment, he is not eligible for disability benefits. In such a case, there is no reason for the Secretary to consider the claimant’s age, education, and work experience. The legislative history reinforces this understanding of the statutory language. Section 423(d)(2)(A) was intended to “reemphasize the predominant importance of medical factors in the disability determination.” S. Rep. No. 744, 90th Cong., 1st Sess., 48 (1967). The 1967 Amendments left undisturbed the longstanding regulatory provision that “medical considerations alone may justify a finding that the individual is not under a disability.” 20 CFR §404.1502(a) (1966). Indeed, it is clear that Congress contemplated a sequential evaluation process: “The bill would provide that such an individual would be disabled [i] only if it is shown that he has a severe medically determinable physical or mental impairment or impairments; [ii] that if, despite his impairment or impairments, an individual still can do his previous work, he is not under a disability; and [iii] that if, considering the severity of his impairment together with his age, educa 6 According to the dissent our opinion implies that the Secretary has unlimited authority to deny meritorious claims. Post, at 160, n. 1. It hardly needs saying that our opinion carries no such implication. BOWEN v. YUCKERT 149 137 Opinion of the Court tion, and experience, he has the ability to engage in some other type of substantial gainful work that exists in the national economy even though he can no longer do his previous work, he also is not under a disability . . . .” S. Rep. No. 744, supra, at 48-49. See H. R. Rep. No. 544, 90th Cong., 1st Sess., 30 (1967).7 C If there was any lingering doubt as to the Secretary’s authority to require disability claimants to make a threshold ’Justice Blackmun’s dissent argues that a “straightforward reading” of the statute requires the Secretary expressly to consider the age, education, and work experience of any claimant who is unable to perform his past work, and who is able to show a medically determinable impairment, however trivial. Post, at 163. The dissent’s reading would make the severity of the claimant’s medical impairment turn on nonmedical factors such as education and experience. For example, the dissent asserts that the Court’s “reasoning begs the very question presented for resolution today— whether the severity of a claimant’s medical impairment can be discerned without reference to the individual’s age, education, and work experience.” Post, at 168, n. 7 (emphasis added). Moreover, the dissent ignores the fact that, below a threshold level of medical severity, an individual is not prevented from engaging in gainful activity “by reason of” the physical or mental impairment. 68 Stat. 1080, 42 U. S. C. § 423(d)(1)(A). Curiously, the dissent bases its position largely on § 423(d)(2)(A), a provision added to “reemphasize the predominant importance of medical factors,” S. Rep. No. 744, 90th Cong., 1st Sess., 48 (1967). The dissent’s reading of § 423(d)(2)(C) also is novel. That provision applies to the Secretary’s determination “whether an individual’s physical or mental impairment or impairments are of a sufficient medical severity that such impairment or impairments could be the basis of eligibility . . . .” As the dissenters read this provision, it does not support the severity regulation because it “says nothing of the severity level necessary to meet the eligibility requirements.” Post, at 174 (emphasis added). Of course, any threshold, however low, is still a threshold. Finally, the fact that the disability claims of widows and widowers are decided solely on the basis of medical factors, see 42 U. S. C. § 423(d)(2)(B); post, at 163-164, does not imply that Congress intended the Secretary expressly to consider nonmedical factors in other cases, no matter how trivial the medical impairment. In sum, the dissent’s reading of the statute is less than “straightforward.” 150 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. showing of medical severity, we think it was removed by § 4 of the Social Security Disability Benefits Reform Act of 1984, 98 Stat. 1800. It is true that “ ‘[t]he Reform Act is remedial legislation, enacted principally to be of assistance to large numbers of persons whose disability benefits have been terminated.’” Bowen n. City of New York, 476 U. S., at 486, n. 14 (quoting City of New York n. Heckler, 755 F. 2d 31, 33 (CA2 1985)). But Congress nevertheless expressed its approval of the severity regulation both in the statute and in the accompanying Reports.8 Sections 4(a)(1) and (b) of the 1984 Act provide: “In determining whether an individual’s physical or mental impairment or impairments are of a sufficient medical severity that such impairment or impairments could be the basis of eligibility under this section, the Secretary shall consider the combined effect of all of the individual’s impairments without regard to whether any such impairment, if considered separately, would be of such severity. If the Secretary does find a medically severe 8Justice Blackmun’s dissent recognizes, as it must, that the Secretary’s “severity regulation” requires a claimant to show a medically severe impairment or combination of impairments. Absent such a showing, it is unnecessary to consider the claimant’s “age, education, and work experience.” The dissent concludes, however, that the regulation “contradicts the statutory language” and therefore is invalid. Post, at 159. It is explicitly clear from the legislative history of the 1984 amendments that Congress perceived no such inconsistency. Indeed, both the Senate and House Reports endorse the severity regulation. The Senate Report, for example, states that a “claim must be disallowed” unless the Secretary determines “first, on a strictly medical basis and without regard to vocational factors, whether the individual’s impairments, considered in combination, are medically severe. If they are not, the claim must be disallowed.” S. Rep. No. 98-466, p. 22 (1984). The House Report is not inconsistent, and the Conference Report is in full accord. See infra, at 151-152. The dissent nevertheless views much of the legislative history as “ambiguous, post, at 175; see post, at 177. Even if we agreed that there was some ambiguity, we would defer to the Secretary’s interpretation of the statute. See Heckler Campbell, 461 U. S. 458, 466 (1983); supra, at 145. BOWEN v. YUCKERT 151 137 Opinion of the Court combination of impairments, the combined effect of the impairments shall be considered throughout the disability determination process.” 42 U. S. C. §§ 423(d)(2)(C), 1382c(a)(3)(F) (1982 ed. and Supp. III). Congress thus recognized once again that the Secretary may make an initial determination of medical severity, and that he need not consider the claimant’s age, education, and experience unless he finds “a medically severe combination of impairments.” The Senate Report accompanying the 1984 amendments expressly endorses the severity regulation. “[T]he new rule [requiring consideration of the combined effects of multiple impairments] is to be applied in accordance with the existing sequential evaluation process and is not to be interpreted as authorizing a departure from that process. . . . The amendment requires the Secretary to determine first, on a strictly medical basis and without regard to vocational factors, whether the individual’s impairments, considered in combination, are medically severe. If they are not, the claim must be disallowed. Of course, if the Secretary does find a medically severe combination of impairments, the combined impact of the impairments would also be considered during the remaining stages of the sequential evaluation process.” S. Rep. No. 98-466, p. 22 (1984). The House Report agrees: “[I]n the interests of reasonable administrative flexibility and efficiency, a determination that a person is not disabled may be based on a judgment that the person has no impairment, or that the impairment or combination of impairments [is] slight enough to warrant a presumption that the person’s work ability is not seriously affected. The current ‘sequential evaluation process’ allows such a determination, and the committee does not wish to elimi 152 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. nate or seriously impair use of that process.” H. R. Rep. No. 98-618, p. 8 (1984).9 Finally, the Conference Report stated: “[I]n the interests of reasonable administrative flexibility and efficiency, a determination that an individual is not disabled may be based on a judgment that an individual has no impairment, or that the medical severity of his impairment or combination of impairments is slight enough to warrant a presumption, even without a full evaluation of vocational factors, that the individual’s ability to perform [substantial gainful activity] is not seriously affected. The current ‘sequential evaluation process’ allows such a determination and the conferees do not intend to either eliminate or impair the use of that process.” H. R. Conf. Rep. No. 98-1039, p. 30 (1984).10 9 The House Report observed that the Secretary had “been criticized for basing terminations of benefits solely and erroneously on the judgment that the person’s medical evaluation is ‘slight,’ according to very strict criteria, and is therefore not disabling, without making any further evaluation of the person’s ability to work.” H. R. Rep. No. 98-618, p. 7 (1984). The Report “notes that the Secretary has already planned to re-evaluate the current criteria for non-severe impairments, and urges that all due consideration be given to revising those criteria to reflect the real impact of impairments upon the ability to work.” Id., at 8. These comments about the Secretary’s application of the severity regulation hardly suggest that the regulation is invalid on its face. 10 Senator Long, a ranking Member of the Conference Committee, observed that “[s]ome courts . . . have ruled that the Secretary cannot deny claims solely on the basis that the individual has no severe medical condition but must always make an evaluation of vocational capacities.” 130 Cong. Rec. 25981 (1984). Senator Long went on to state that the Senate bill, that was followed by the conference bill with only “minor language changes of a technical nature,” ibid., was “carefully drawn to reaffirm the authority of the Secretary to limit benefits to only those individuals with conditions which can be shown to be severe from a strictly medical standpoint —that is, without vocational evaluation,” ibid. Senator Long was one of the sponsors of the disability program when it was enacted in 1956, see S. Rep. No. 2133, 84th Cong., 2d Sess., 140 (1956), and also was Chairman of the Senate Finance Committee when the 1967 BOWEN v. YUCKERT 153 137 Opinion of the Court IV We have recognized that other aspects of the Secretary’s sequential evaluation process contribute to the uniformity and efficiency of disability determinations. Heckler n. Campbell, 461 U. S., at 461. The need for such an evaluation process is particularly acute because the Secretary decides more than 2 million claims for disability benefits each year, of which more than 200,000 are reviewed by administrative law judges. Department of Health and Human Services, Social Security Administration 1986 Annual Report to Congress, pp. 40, 42, 46. The severity regulation increases the efficiency and reliability of the evaluation process by identifying at an early stage those claimants whose medical impairments are so slight that it is unlikely they would be found to be disabled even if their age, education, and experience were taken into account. Similarly, step three streamlines the decision process by identifying those claimants whose medical impairments are so severe that it is likely they would be found disabled regardless of their vocational background. Respondent Yuckert has conceded that the Secretary may require claimants to make a “de minimis” showing that their impairment is severe enough to interfere with their ability to work.11 Brief for Respondent 22-23; Tr. of Oral Arg. 30. Yuckert apparently means that the Secretary may require a showing that the “impairment is so slight that it could not interfere with [the claimant’s] ability to work, irrespective of Amendments to the Act were enacted, see S. Rep. No. 744, 90th Cong., 1st Sess., 1 (1967). 11 Although the issue was not briefed or argued by the parties, the dissent nevertheless concludes that the severity regulation should be invalidated because it is excessively vague. Post, at 168. The severity regulation plainly adopts a standard for determining the threshold level of severity: the impairment must be one that “significantly limits your physical or mental ability to do basic work activities.” 20 CFR §404.1520(c) (1986). Moreover, as discussed supra, at 141, the Secretary’s regulations define “basic work activities” in detail. 154 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. age, education, and work experience.” Brief for Respondent 22. She contends that the Secretary imposed only a “de minimis” requirement prior to 1978, but has required a greater showing of severity since then. As we have noted, however, Congress expressly approved the facial validity of the 1978 severity regulation in the 1984 amendments to the Act. Particularly in light of those amendments and the legislative history, we conclude that the regulation is valid on its face.12 12 As the Court of Appeals for the Ninth Circuit invalidated the regulation on its face, we have no occasion to consider whether it is valid as applied. A number of Courts of Appeals have held that the Secretary has exceeded his authority by denying large numbers of meritorious disability claims at step two. See cases cited in n. 4, supra. We have noted that the House Report accompanying the 1984 amendments urged the Secretary to reevaluate the severity criteria to determine whether they were too strict. See n. 9, supra. Subsequent to the adjudication of Yuckert’s disability claim, the Secretary issued a ruling “[t]o clarify the policy for determining when a person’s impairment(s) may be found ‘not severe’. . . .” Social Security Ruling 85-28, App. to Pet. for Cert. 37a. The ruling states: “An impairment or combination of impairments is found ‘not severe’ and a finding of ‘not disabled’ is made at [step two] when medical evidence establishes only a slight abnormality or a combination of slight abnormalities which would have no more than a minimal effect on an individual’s ability to work even if the individual’s age, education, or work experience were specifically considered (i. e., the person’s impairment(s) has no more than a minimal effect on his or her physical or mental ability(ies) to perform basic work activities).” Id., at 41a. If the “evidence shows that the person cannot perform his or her past relevant work because of the unique features of that work,” the decisionmaker will conduct a “further evaluation of the individual’s ability to do other work considering age, education and work experience.” Id., at 43a. We do not undertake to construe this ruling today. We do, however, reject Yuckert’s contention that invalidation of the regulation is an appropriate remedy for the Secretary’s allegedly unlawful application of the regulation. See Brief for Respondent 44-47. The Court of Appeals did not invalidate the regulation on this ground. Moreover, there is no indication in the record that less drastic remedies would not have been effective. BOWEN v. YUCKERT 155 137 O’Connor, J., concurring V The judgment of the Court of Appeals for the Ninth Circuit is reversed. The case is remanded for the Court of Appeals to consider whether the agency’s decision is supported by substantial evidence. It is so ordered. Justice O’Connor, with whom Justice Stevens joins, concurring. The Court is, I believe, entirely correct to find that the “step two” regulation is not facially inconsistent with the Social Security Act’s definition of disability. Title 42 U. S. C. § 423(d)(2)(A) (1982 ed. and Supp. Ill) provides: “[A]n individual . . . shall be determined to be under a disability only if his physical or mental impairment or impairments are of such severity that he is not only unable to do his previous work but cannot, considering his age, education, and work experience, engage in any other kind of substantial gainful work which exists in the national economy.” Step two on its face requires only that the claimant show that he or she suffers from “an impairment or combination of impairments . . . [that] significantly limit[s] . . . physical or mental ability to do basic work activities.” 20 CFR §404.1521(a) (1986). “Basic work activities,” the regulation says, include “walking, standing, sitting, lifting, pulling, reaching, carrying, or handling[,] . . . seeing, hearing, and speaking, . . . [understanding, carrying out, and remembering simple instructions^] . . . [u]se of judgment[,] . . . [responding appropriately to supervision, co-workers and usual work situations[,] . . . [d]ealing with changes in a routine work setting.” §404.1521(b)(l)-(6). I do not see how a claimant unable to show a significant limitation in any of these areas can possibly meet the statutory definition of disability. For the reasons set out by the Court in Part III of 156 OCTOBER TERM, 1986 O’Connor, J., concurring 482 U. S. its opinion, I have no doubt that the Act authorizes the Secretary to weed out at an early stage of the administrative process those individuals who cannot possibly meet the statutory definition of disability. Accordingly, I concur in the Court’s opinion and judgment that the regulation is not facially invalid, and that the case must be remanded so that the lower courts may determine whether or not the Secretary’s conclusion that Janet Yuckert is not suffering from a sufficiently severe impairment is supported by substantial evidence. I write separately, however, to discuss the contention of respondent and various amici (including 29 States and 5 major cities) that this facially valid regulation has been applied systematically to deny benefits to claimants who do meet the statutory definition of disability. Respondent directs our attention to the chorus of judicial criticism concerning the step two regulation, as well as to substantially unrefuted statistical evidence. Despite the heavy deference ordinarily paid to the Secretary’s promulgation and application of his regulations, Schweiker v. Gray Panthers, 453 U. S. 34, 43 (1981), all 11 regional Federal Courts of Appeals have either enjoined the Secretary’s use of the step two regulation1 or imposed a narrowing construction upon it.2 The 1 Dixon v. Heckler, 785 F. 2d 1102 (CA2 1986) (preliminary injunction), cert, pending, No. 86-2; Wilson v. Secretary of Health and Human Services, 796 F. 2d 36 (CA3 1986); Boeder v. Heckler, 768 F. 2d 547 (CA3 1985); Johnson n. Heckler, 769 F. 2d 1202 (CA7 1985), cert, pending sub nom. Bowen n. Johnson, No. 85-1442; Brown v. Heckler, 786 F. 2d 870 (CA8 1986); Yuckert v. Heckler, 774 F. 2d 1365 (CA9 1985) (case below); Hansen v. Heckler, 783 F. 2d 170 (CAIO 1986). 2 McDonald n. Secretary of Health and Human Services, 795 F. 2d 1118 (CAI 1986) (relying upon Social Security Ruling 85-28); Evans v. Heckler, 734 F. 2d 1012 (CA4 1984); Stone v. Heckler, 752 F. 2d 1099 (CA5 1985); Estran v. Heckler, 745 F. 2d 340 (CA5 1984); Farris v. Secretary of Health and Human Services, 773 F. 2d 85 (CA6 1985); Salmi v. Secretary of Health and Human Services, 774 F. 2d 685 (CA6 1985); McCruter n. Bowen, 791 F. 2d 1544 (CA11 1986); Brady v. Heckler, 724 F. 2d 914 (CA11 1984). BOWEN v. YUCKERT 157 137 O’Connor, J., concurring frustration expressed by these courts in dealing with the Secretary’s application of step two in particular cases is substantial, and no doubt in part accounts for the Court of Appeals’ decision in this case to simply enjoin the regulation’s further use. Empirical evidence cited by respondent and the amici further supports the inference that the regulation has been used in a manner inconsistent with the statutory definition of disability. Before the step two regulations were promulgated approximately 8% of all claimants were denied benefits at the “not severe” stage of the administrative process; afterwards approximately 40% of all claims were denied at this stage. See Baeder n. Heckler, 768 F. 2d 547, 552 (CA3 1985). As the lower federal courts have enjoined use of step two and imposed narrowing constructions, the step two denial rate has fallen to about 25%. House Committee on Ways and Means, Background Material and Data on Programs Within the Jurisdiction of the Committee on Ways and Means, 99th Cong., 2d Sess., 114 (Comm. Print 1986). Allowance rates in Social Security disability cases have increased substantially when federal courts have demanded that the step two regulation not be used to disqualify those who are statutorily eligible. For example, in Illinois after entry of the injunction in Johnson v. Heckler, 769 F. 2d 1202 (CA7 1985), cert, pending sub nom. Bowen v. Johnson, No. 85-1442, the approval rate for claims climbed from 34.3% to 52% at the initial screening level and from 14.8% to 34.1% at the reconsideration level. See Brief for Alabama et al. as Amici Curiae 22. To be sure the Secretary faces an administrative task of staggering proportions in applying the disability benefits provisions of the Social Security Act. Perfection in processing millions of such claims annually is impossible. But respondent’s evidence suggests that step two has been applied systematically in a manner inconsistent with the statute. In 158 OCTOBER TERM, 1986 O’Connor, J., concurring 482 U. S. deed, the Secretary himself has recently acknowledged a need to “clarify” step two in light of this criticism and has attempted to do so by issuing new interpretative guidelines. See Social Security Ruling 85-28, App. to Pet. for Cert. 37a. In my view, step two may not be used to disqualify those who meet the statutory definition of disability. The statute does not permit the Secretary to deny benefits to a claimant who may fit within the statutory definition without determining whether the impairment prevents the claimant from engaging in either his prior work or substantial gainful employment that, in light of the claimant’s age, education, and experience, is available to him in the national economy. Only those claimants with slight abnormalities that do not significantly limit any “basic work activity” can be denied benefits without undertaking this vocational analysis. See Evans v. Heckler, 734 F. 2d 1012, 1014 (CA4 1984); Estran v. Heckler, 745 F. 2d 340, 341 (CA5 1984) (per curiam); Brady v. Heckler, 724 F. 2d 914, 920 (CA11 1984). As the Secretary has recently admonished in his new guideline: “Great care should be exercised in applying the not severe impairment concept. If an adjudicator is unable to determine clearly the effect of an impairment or combination of impairments on the individual’s ability to do basic work activities, the sequential evaluation process should not end with the not severe evaluation step. Rather, it should be continued. In such a circumstance, if the impairment does not meet or equal the severity level of the relevant medical listing, sequential evaluation requires that the adjudicator evaluate the individual’s ability to do past work, or to do other work based on the consideration of age, education, and prior work experience. ” Social Security Ruling 85-28, App. to Pet. for Cert. 44a. Applied in this manner, step two, I believe, can produce results consistent with the statute in the vast majority of cases BOWEN v. YUCKERT 159 137 Blackmun, J., dissenting and still facilitate the expeditious and just settlement of claims. Justice Blackmun, with whom Justice Brennan and Justice Marshall join, dissenting. The definition of “disability” for purposes of the disabilityinsurance benefits program is set forth in § 223(d) of the Social Security Act, codified, as amended, at 42 U. S. C. § 423(d) (1982 ed. and Supp. III). Paragraph (2)(A) of that section states: “An individual. . . shall be determined to be under a disability only if his physical or mental impairment or impairments are of such severity that he is not only unable to do his previous work but cannot, considering his age, education, and work experience, engage in any other kind of substantial gainful work” (emphasis added). The “severity regulation” promulgated by the Secretary of Health and Human Services for purposes of the program, however, explains to a claimant: “If you do not have any impairment or combination of impairments which significantly limits your physical or mental ability to do basic work activities, we will find that you do not have a severe impairment and are, therefore, not disabled. We will not consider your age, education, and work experience.” 20 CFR §404.1520(c) (1986) (emphasis added). This regulation, on its very face, directly contradicts the statutory language requiring that a claimant’s age, education, and work experience be considered in a case where the claimant cannot perform his past work. It is thus invalid. The legislative history of § 423(d) confirms that the severity regulation exceeds the Secretary’s statutory authority. Because the Court reverses the Court of Appeals’ judgment that correctly invalidated that regulation, I dissent. I A In its opinion today, the Court analyzes the facial validity of the Secretary’s severity regulation by interpreting § 423(d) 160 OCTOBER TERM, 1986 Blackmun, J., dissenting 482 U. S. in a manner that defeats the intent expressed through its language and structure. The Court isolates paragraph (1)(A) of § 423(d) and finds that the severity regulation does not conflict with the 1954 statutory definition of disability contained therein. Disregarding the fact that this definition was later amended to include paragraph (2) of § 423(d), the Court reaches a premature conclusion that the regulation “is not inconsistent with the statutory definition of disability.” Ante, at 146. After thus reasoning that the “statutory definition of disability” is not a bar to the Secretary’s severity regulation, the Court then characterizes paragraph (2)(A) as merely “limit[ing] the Secretary’s authority to grant disability benefits, not to deny them.”1 Ante, at 148. This allows the Court to conclude that there is no reason for the Secretary to consider the vocational factors of age, education, and work experience listed in paragraph (2)(A) in cases where he already has determined that the claimant does not have a severe impairment. The critical error in the Court’s analysis is readily apparent when one considers the language introducing paragraph (2) of § 423(d). Although the Court purports to set forth § 423(d) (2)(A) in its opinion, ante, at 140, it fails to quote the key language from the statute. The concurring opinion likewise ’The Court implies that the statute limits the Secretary’s ability to grant disability-insurance benefits but does not limit his ability to deny such benefits. This implication is inconsistent with the fact that the disability-insurance benefits program at issue here creates a statutory entitlement for those persons eligible under the statutory criteria. Section 423 begins by stating that every individual who is insured, is not of retirement age, has filed an application, and is disabled, “shall be entitled to a disability insurance benefit” during the time period deemed appropriate under the standards set forth in the statute (emphasis added). 42 U. S. C. § 423(a)(1) (1982 ed., Supp. III). The Court elsewhere takes note of the fact that claimants, such as respondent, who seek disabilityinsurance payments due to inability to continue working have contributed to the insurance program. Ante, at 140. As under any insurance program, a contributor to the plan is entitled to payment if he or she meets the agreed-upon terms for coverage. BOWEN v. YUCKERT 161 137 Blackmun, J., dissenting presents an abridged version of the statute. See ante, at 155. Neither places the language that it does quote within its proper context. Section 423(d) provides in relevant part: “(1) The term ‘disability’ means — “(A) inability to engage in any substantial gainful activity by reason of any medically determinable physical or mental impairment which can be expected to result in death or which has lasted or can be expected to last for a continuous period of not less than 12 months; or “(B) in the case of an individual who has attained the age of 55 and is blind .... “(2) For purposes of paragraph (1)(A)— “(A) An individual (except a widow, surviving divorced wife, widower, or surviving divorced husband for purposes of section 402(e) or (f) of this title) shall be determined to be under a disability only if his physical or mental impairment or impairments are of such severity that he is not only unable to do his previous work but cannot, considering his age, education, and work experience, engage in any other kind of substantial gainful work which exists in the national economy, regardless of whether such work exists in the immediate area in which he lives, or whether a specific job vacancy exists for him, or whether he would be hired if he applied for work. For purposes of the preceding sentence (with respect to any individual), ‘work which exists in the national economy’ means work which exists in significant numbers either in the region where such individual lives or in several regions of the country. “(B) A widow, surviving divorced wife, widower, or surviving divorced husband shall not be determined to be under a disability (for purposes of section 402(e) or (f) of this title) unless his or her physical or mental impairment or impairments are of a level of severity which under regulations prescribed by the Secretary is deemed 162 OCTOBER TERM, 1986 Blackmun, J., dissenting 482 U. S. to be sufficient to. preclude an individual from engaging in any gainful activity.” 42 U. S. C. § 423(d) (1982 ed. and Supp. Ill) (emphasis added). By employing the phrase “for purposes of paragraph (1)(A)” to introduce paragraph (2), Congress made clear that paragraph (2) serves as an annotation to paragraph (1)(A), not as an independent requirement, as the Court implies. The language and structure of § 423(d) plainly indicate that paragraph (2) is relevant at the time the determination is made under paragraph (1)(A), not afterwards. Paragraph (2), in effect, explains how to determine whether a claimant is unable “to engage in any substantial gainful activity” within the meaning of paragraph (1)(A).2 How the determination is to be made in most cases, including those brought by insured workers such as respondent Janet Yuckert, is set forth in paragraph (2)(A), whereas paragraph (2)(B) relates to the category of claims by surviving spouses of insured workers which is specifically excepted from paragraph (2)(A). Whether a claimant under (2)(A) has proved an “inability” to work “by reason of” a medical impair 2 This interpretation is strongly reinforced by 42 U. S. C. § 416(i). That section provides a definition of “disability” and “period of disability” for various other sections of the statute. It states in relevant part: “(1) Except for purposes of sections 402(d), 402(e), 402(f), 423, and 425 of this title, the term ‘disability’ means (A) inability to engage in any substantial gainful activity by reason of any medically determinable physical or mental impairment which can be expected to result in death or has lasted or can be expected to last for a continuous period of not less than 12 months, or (B) blindness .... The provisions of paragraphs (2)(A), (2)(C), (3), (4), (5), and (6) of section 423(d) of this title shall be applied for purposes of determining whether an individual is under a disability within the meaning of the first sentence of this paragraph in the same manner as they are applied for purposes of paragraph (1) of such section.” (Emphasis added.) 42 U. S. C. §416(i)(l) (1982 ed., Supp. III). Clearly, Congress intended that paragraph (2)(A) of § 423(d) be applied for purposes of determining whether a claimant is under a disability within the meaning of paragraph (1)(A) of § 423(d). BOWEN v. YUCKERT 163 137 Blackmun, J., dissenting ment for purposes of (1)(A) depends upon whether the impairment limits the worker to such an extent that he is “not only unable to do his previous work but cannot, considering his age, education, and work experience, engage in any [other work].” A straightforward reading of §§ 423(d)(1)(A) and (2)(A) indicates that the claimant must establish that he has an impairment, that it is medically determinable, that it meets the duration requirement, and that it is severe enough to be disabling within the terms of the statute so as to render him eligible for benefits. Paragraph (1)(A) does not indicate how the Secretary is to assess whether any established medical impairments meet the statutory severity standard. Paragraph (2)(A), however, provides that guidance. Under paragraph (2)(A), if the claimant is able to do his previous work, the Secretary, of course, need not consider his age, education, and work experience. In such a case, the medically determinable impairment is automatically deemed nonsevere within the meaning of the Act. If, however, the claimant cannot perform his past work, the Secretary then must inquire into the severity of the impairment or combination of impairments. He is to determine whether, in light of the claimant’s age, education, and work experience, the impairment is so severe that the claimant cannot engage in substantial gainful work. A comparison of this process to that set forth in paragraph (2)(B) leaves no doubt whatsoever that consideration of the vocational factors is a key feature of the process in evaluating claims under paragraph (2)(A). In paragraph (2)(B), Congress authorized the Secretary to deny benefit claims by surviving spouses based on medical evidence alone. That paragraph specifies that the Secretary may promulgate listed severity levels of impairments at which an individual cannot engage in any gainful activity, and may deny benefits in such 164 OCTOBER TERM, 1986 Blackmun, J., dissenting 482 U. S. cases simply by comparison to this list.3 If Congress had intended to authorize the Secretary to deny benefits in that same manner in disability claims under paragraph (2)(A), without consideration of age, education, or work experience, it would have included the same language in paragraph (2)(A) that it used in paragraph (2)(B). B The § 423(d)(2)(A) inquiry furthers the purpose of the disability-benefits program by ensuring an individualized assessment of alleged disability in cases of insured workers. The inquiry takes into account the fact that the same medically determinable impairment affects persons with different vocational characteristics differently. A relatively young, well-educated, and experienced individual who can no longer perform his past work due to a medical impairment may be able to transfer his skills to another job and perform substantial gainful work. That same medical impairment may have a much greater effect on a person’s ability to perform substantial gainful work if the person is of advanced age and has minimal education and limited work experience. Thus, a particular medical impairment may not be disabling for the first individual while it could be for the second. Despite the clarity of the statutory language and the purpose of individualized disability determinations, the Secretary has promulgated as step two of his step-evaluation process the severity standard set forth in 20 CFR §404.1520(c) 3 In the regulations pertaining to surviving spouses, the Secretary explains to claimants: “To determine whether you are disabled, we consider only your physical or mental impairment. We do not consider your age, education and work experience.” 20 CFR §404.1577 (1986). He further explains that the claimant will be found to be disabled only if he is not doing any substantial gainful activity and his impairment meets the requirements of an impairment listed in an accompanying appendix. See § 404.1578; see also Hansen v. Heckler, 783 F. 2d 170, 172 (CAIO 1986) (statutory criteria for disability benefit claims by widows are more restrictive than standard applicable to claims by insured wage earners). BOWEN v. YUCKERT 165 137 Blackmun, J., dissenting (1986). Because that regulation prohibits agency adjudicators from considering a claimant’s age,4 education, and work experience in cases where the claimant cannot perform his past work, the regulation is invalid on its face.5 4 Perhaps the most disturbing result of the step two severity regulation is the disproportionate effect that its application has had on claimants in the older age categories. Some of the amici express concern that the Court realize that the instant case, which involves a claim by a relatively young and well-educated individual, is not typical of cases in which step two has operated to deny benefits to eligible claimants. See Brief for American Diabetes Association et al. as Amici Curiae 9, n. 5, 12-13; Brief for American Association of Retired Persons as Amicus Curiae 16, n. 13. A survey of a significant number of reported cases reveals that in those cases the claimants whose step two severity denials were reversed by the courts were individuals age 50 or older. See id., at 16, n. 12. In one of its earliest statements as to why the vocational factors must be considered in making disability determinations, the Secretary explained that “[t]he aging process makes itself felt with respect to healing, prognosis, physiological degeneration, psychological adaptability and, in consequence, on vocational capacity.” 1955 Disability Freeze State Manual §325B. Noting that chronological age, however, was only “some indication of the individual’s physiological age,” the Secretary specified that “the impact of the aging process upon the specific individual will have to be considered in connection with the particular impairment claimed to prevent substantial gainful activity.” Ibid. Elimination of the age factor from the disability calculus at step two inevitably diminishes the reliability of the determinations at that step. 5 Although the Court peremptorily finds “no basis” for holding that the severity regulation is inconsistent on its face with the statute, ante, at 147, no less than five Federal Courts of Appeals, including the court below, have found the same blatant contradiction in the plain language that I find. See Brown v. Heckler, 786 F. 2d 870, 871 (CA8 1986) (citing Courts of Appeals that have “point[ed] out that while the provision [in the severity regulation] explicitly requires the Secretary to disregard the claimant’s age, education, and work experience, the Act expressly requires those factors to be taken into account when determining disability”); Hansen v. Heckler, 783 F. 2d, at 174 (“regulation on its face . . . conflicts with the statutory directive” which is “to consider a claimant’s ability both to perform past work and, given individual vocational factors, to engage in other work”); Johnson v. Heckler, 769 F. 2d 1202, 1212 (CA7 1985) (“[O]n its face, the step two severity regulation conflicts with 166 OCTOBER TERM, 1986 Blackmun, J., dissenting 482 U. S. The reasoning upon which the Court relies to support its contrary conclusion is unconvincing. Rather than analyze the severity regulation’s validity in light of the actual language and purpose of the statute, the Court relies, ante, at 146, on a description of the Act’s definition of disability set forth in one of its own earlier opinions. See Heckler v. Campbell, 461 U. S. 458, 459-460 (1983) (“The Social Security Act defines ‘disability’ in terms of the effect a physical or mental impairment has on a person’s ability to function in the workplace”). It is important to note, however, that the Court quotes only part of that description. Based on this abbreviated description, the Court views the statute as requiring a “functional approach to determining the effects of medical impairments,” ante, at 146, and regards the regulation as adopting a similar approach. Merely because both the statute and the regulation require analysis of the effect of the medical impairments on the claimant’s ability to work does not mean, however, that the two are consistent in all respects. Moreover, examination of the description of the statutory scheme, as set forth in Heckler v. Campbell, reveals that the general declaration upon which the Court relies was supported with a discussion of the particulars of the statute that included both paragraphs (1)(A) and (2)(A) of § 423(d). By not including § 423(d)(2)(A) at this step of its analysis, however, the Court avoids the impossible task of explaining how the statutory scheme described in Campbell and the regulatory scheme set forth in the severity regulation can represent “precisely” the same approach when the Social Security Act’s purposes and the plain language of the statute’s definitions of disability”), cert, pending sub nom. Bowen v. Johnson, No. 85-1442; Baeder v. Heckler, 768 F. 2d 547, 553 (CA3 1985) (severity regulation cannot be analyzed “except according to its plain language and the manner in which the Secretary uses it”; “[a]s it stands,. . . [it] is inconsistent with the Social Security Act, and therefore, is invalid”). Obviously, these cases do not support the assertion in the concurring opinion that the courts rested their judgments on “frustration ... in dealing with the Secretary’s application of step two.” Ante, at 157. BOWEN v. YUCKERT 167 137 Blackmun, J., dissenting the statutory scheme includes consideration of vocational factors and the regulation does not. While still focusing on the comparison between the regulation and paragraph (1)(A) read in isolation, the Court states: “If the impairments are not severe enough to limit significantly the claimant’s ability to perform most jobs [apparently referring to 20 CFR §404.1521 (1986)],6 by definition the impairment does not prevent the claimant from engaging in any substantial gainful activity.” Ante, at 146. Although I agree that a claimant who can perform most jobs is not disabled under the Act, I do not agree with the Court’s implication that the statute authorizes the Secretary to review the medical evidence in a case and, solely on the basis of that information, to determine the claimant’s ability to “perform most jobs.” Under that interpretation of the statute, the agency adjudicators would decide whether a claimant covered by § 423(d)(2)(A) could perform the listed basic-work activities, including responsiveness to supervision and adaptability to change in the workplace, without taking into account the claimant’s age, education, and work experience. I simply cannot read the statutory language of §§ 423(d)(1)(A) 6 In § 404.1521, the Secretary explains what he means by “an impairment that is not severe”: “(a) Non-severe impairments). An impairment or combination of impairments is not severe if it does not significantly limit your physical or mental ability to do basic work activities. “(b) Basic work activities. When we talk about basic work activities, we mean the abilities and aptitudes necessary to do most jobs. Examples of these include— “(1) Physical functions such as walking, standing, sitting, lifting, pushing, pulling, reaching, carrying, or handling; “(2) Capacities for seeing, hearing, and speaking; “(3) Understanding, carrying out, and remembering simple instructions; “(4) Use of judgment; “(5) Responding appropriately to supervision, co-workers and usual work situations; and “(6) Dealing with changes in a routine work setting.” (Emphasis added.) 168 OCTOBER TERM, 1986 Blackmun, J., dissenting 482 U. S. and (2)(A) as authorizing the Secretary to permit that determination to be made in such a void. Even if a medical impairment affected different individuals’ abilities to perform such functions to the same extent, regardless of age, education, and work experience, there is no guidance in the severity regulation as to what constitutes a “significant” limitation on the ability, for example, to use judgment or to adapt to changes in work conditions, or as to how the degree of limitation caused by a medical impairment on such functions is to be determined based solely on medical evidence. Nor does the regulation explain whether the claimant must be able to perform a few, most, or all of the §404.1521 “[e]xamples” of “basic work activities” in order to be found capable of performing “most jobs.” The concurring opinion appears to assume that the Secretary can deny benefits at that stage only if a claimant can perform all the basic work activities listed without any significant limitations. Ante, at 155-156. Assuming this to be true, the regulation does not recognize that less than “significant” limitations on several of the activities in combination could equate with an overall significant limitation on the ability to perform most jobs. In sum, the regulation authorizes disability determinations to be made in a manner inconsistent with the statutory mandate. Congress clearly intended to prohibit these assessments from being made in a vacuum when it specified in § 423(d)(2)(A) that a claimant’s age, education, and work experience be taken into account in determining the effect of his medical impairment on his ability to work.7 7 The Court is correct in asserting that “[i]f a claimant is unable to show that he has a medically severe impairment, he is not eligible for disability benefits.” Ante, at 148. I disagree, however, with its conclusion, drawn from that assertion, that “[i]n such a case, there is no reason for the Secretary to consider the claimant’s age, education, and work experience.” Ibid. This reasoning begs the very question presented for resolution today—whether the severity of a claimant’s medical impairment can be discerned without reference to the individual’s age, education, and work experience. The statute expressly answers this question in the negative. BOWEN v. YUCKERT 169 137 Blackmun, J., dissenting There simply is no support in the language of the statute for the proposition that the Secretary can create his own definition of “severe impairment” for purposes of disability determinations in disabled-worker cases and exclude consideration of factors that Congress directed be considered. Whereas the Court perceives “no reason for the Secretary to consider the claimant’s age, education, and work experience” in cases where a premature showing of nonseverity has been made, ante, at 148, there is one compelling reason for the Secretary to consider those factors—the unambiguous language of the statute directs that he do so. II A An examination of the legislative history of § 423(d) provides strong additional support for respondent’s position. The disability definition in § 423(d) has its roots in another statutory provision that was first enacted in 1952. In that year, Congress amended the Social Security Act in part to guarantee that the insured status of workers would not be adversely affected if they were permanently and totally disabled for periods of time prior to retirement. As part of this amendment, Congress added to the Act its §216(i), which contains the definition of “disability” and “period of disability” for purposes of that program. 66 Stat. 771, 42 U. S. C. §416(i) (1952 ed.). In 1954, Congress replaced those definitions with slightly different ones contained in a new § 216(i). 68 Stat. 1080, 42 U. S. C. §416(i) (1952 ed., Supp. IV). When Congress amended the Act in 1956, in part to establish a program to provide benefits for certain insured disabled individuals prior to retirement, it adopted the § 216(i) definition for purposes of the new program. It added § 223 to the Act which set forth the terms of the new program and included a definition of “disability” nearly identical to that set forth in 170 OCTOBER TERM, 1986 Blackmun, J., dissenting 482 U. S. § 216(i).8 In 1965, Congress amended that definition to specify that the impairment must be expected to last for not less than 12 months. 79 Stat. 367. The statutory definition of disability again was a focus of congressional attention in 1967, when the current structure of the definition was adopted. One of the express aims of the Social Security Act Amendments of that year was to provide a more detailed definition of “disability” for purposes of the disability-insurance benefits program. The definition was set forth in a new § 223(d). 81 Stat. 868. The congressional Reports explain: “Paragraph (1) of the new section 223(d) states the basic definition of the term ‘disability’ exactly as it [was] stated in existing law .... “Paragraph (2)(A) of the new section 223(d) provides that in applying the basic definition (except the special definition for the blind, and except for purposes of widow’s or widower’s insurance benefits on the basis of disability), an individual shall be determined to be under a disability only if his impairment or impairments are so severe that he is not only unable to do his previous work but cannot, considering his age, education, and work experience, engage in any other kind of substantial gainful work which exists in the national economy, regardless of whether such work exists in the general area in which he lives, or whether a specific job vacancy exists, or whether he would be hired if he applied for work.” S. Rep. No. 8 That definition stated: “The term ‘disability’ means inability to engage in any substantial gainful activity by reason of any medically determinable physical or mental impairment which can be expected to result in death or to be of long-continued and indefinite duration. An individual shall not be considered to be under a disability unless he furnishes such proof of the existence thereof as may be required.” 70 Stat. 815-816, 42 U. S. C. § 423(c)(2) (1952 ed., Supp. IV). BOWEN v. YUCKERT 171 137 Blackmun, J., dissenting 744, 90th Cong., 1st Sess., 263-264 (1967); H. R. Rep. No. 544, 90th Cong., 1st Sess., 163 (1967).9 Congress intended that this provision “clarify and amplify the definition of‘disability’ for purposes of the social security program.” S. Rep. No. 744, at 263; H. R. Rep. No. 544, at 163. Congress felt the need to clarify the definition of disability because, in its view, the rising cost of the disability-insurance program was due in part to court decisions that had interpreted the definition too broadly. S. Rep. No. 744, at 46-47. In particular, Congress was concerned with decisions that had required agency adjudicators to focus on a narrow geographic area in determining whether a claimant could perform substantial gainful activity and to consider whether there existed specific job vacancies for which the claimant had a reasonable opportunity to be hired. Id., at 47-48. See, e. g., Tigner v. Gardner, 356 F. 2d 647 (CA5 1966); Wimmer v. Celebrezze, 355 F. 2d 289 (CA4 1966). Congress also noted that questions had arisen about what kind of medical evidence was necessary to “establish the existence and severity of an impairment,” and about what current work performance constituted “substantial gainful activity.” S. Rep. No. 744, at 48. The new language in § 423(d)(2)(A) was aimed at answering these questions. Congress made it clear that medical factors, and not local job conditions, are the primary focus in disability cases. It tempered the new restrictiveness of the statute, however, by specifying that consideration of the vocational factors is a necessary component of the disability determination in all cases where a claimant is not working and the medical impairment is not of a level presumed to be disabling, except those expressly exempted from § 423 9 Section 223(d)(2)(A) was in the original House bill. Although it was in the bill recommended out of Committee in the Senate, it was deleted by amendment on the floor. The Conference Committee restored the provision and the Senate accepted it. H. R. Conf. Rep. No. 1030, 90th Cong., 1st Sess., 51-52 (1967). 172 OCTOBER TERM, 1986 Blackmun, J., dissenting 482 U. S. (d)(2)(A). Congress summarized its view of the disability determination process as follows: “In most cases the decision that an individual is disabled can be made solely on the basis of an impairment, or impairments, which are of a level of severity presumed (under administrative rules) to be sufficient so that, in the absence of an actual demonstration of ability to engage in substantial gainful activity, it may be presumed that the person is unable to so engage because of the impairment or impairments. The language which would be added by the bill specifies the requirements which must be met in order to establish inability to engage in substantial gainful activity for those people with impairments to which the presumption mentioned above does not apply” (emphasis added). S. Rep. No. 744, at 49. Congress nowhere indicated an intention to authorize the Secretary to deny claims by insured workers not performing previous work based on medical factors alone. Congress’ intention that the vocational factors be considered in claims by insured workers such as respondent is further illustrated by comparing Congress’ own description of this process with its description of the simpler process it authorized in cases involving claims by disabled surviving spouses.10 It was explained: “The bill would also provide benefits ... for certain disabled widows . . . and disabled dependent widowers under a test of disability that is somewhat more restrictive than that for disabled workers and childhood disability beneficiaries. The determination of disability in the case of a widow or widower would be based solely on the level of severity of impairment. Determinations in disabled widow and widower cases would be made without 10 Paragraph (2)(B), which sets forth the method of determining disability eligibility for disabled surviving spouses, was also added by the 1967 Amendments. 81 Stat. 868. BOWEN v. YUCKERT 173 137 Blackmun, J., dissenting regard to nonmedical factors such as age, education, and work experience, which are considered in disabled worker cases” (emphasis added). Id., at 49-50. See also H. R. Conf. Rep. No. 1030, 90th Cong., 1st Sess., 52 (1967). Clearly, the nonmedical factors were considered by Congress to be a key ingredient in disability assessments under § 423(d)(2)(A). Out of this legislative history surrounding the enactment and amendment of the current disability definition, the Court grasps at three straws. First, it quotes the legislative Reports that accompanied the 1954 amendment to § 216(i) of the Act. Ante, at 147. The record is clear, however, that the 1967 Amendments to § 223 of the Act represent a decision by Congress to set forth new standards governing the severity assessment of medical impairments. Second, the Court relies upon language from the Senate Report that accompanied the 1967 Amendments. Once again, however, the context is incomplete, for the Court quotes only the remark concerning the “predominant importance of medical factors.” Ante, at 148. There is no question that Congress intended to emphasize that a claimant must produce adequate medical evidence to support his showing of a severe medically determinable impairment. Such an intent, however, is not at odds with Congress’ other clear aim of ensuring that an insured worker’s age, education, and work experience remain relevant factors in the disability determination. Finally, the Court quotes the 1967 Senate Report’s summary of the overall disability evaluation process which, as the Court points out, contemplated a sequential evaluation. Ante, at 148-149. Expressly included in that sequential evaluation, however, is the consideration of the vocational factors in cases where an insured worker cannot do his previous work.11 11 At various points in its opinion, the Court implies that the issue before the Court is the validity of the Secretary’s sequential evaluation process. Respondent, however, has not challenged the validity of that process. I agree that Congress foresaw that there would be various steps in the dis- 174 OCTOBER TERM, 1986 Blackmun, J., dissenting 482 U. S. B To avoid the force of the legislative history contemporaneous with the enactment of §§ 423(d)(1)(A) and (2)(A),12 the Court seeks refuge in § 4(b) of the Social Security Disability Benefits Reform Act of 1984, 98 Stat. 1800. It claims that by this provision, Congress approved the validity of the severity regulation. Yet § 4(b), on its face, says nothing of the severity level necessary to meet the eligibility requirements. See ante, at 149-151. According to that provision, in making a determination of the medical severity of a claimant’s impairment or impairments, the Secretary cannot simply consider each impairment in isolation but rather must consider the combined effect of the impairments. There thus is no “approval of the severity regulation,” as the Court would say, ante, at 150, in the language of that provision. The legislative history of the 1984 Act also does not stand as an endorsement of the severity regulation. Each of the three congressional Reports contains a brief description of the general disability-determination process. In each of these descriptions, the preliminary steps of the Secretary’s step-evaluation process were characterized somewhat differently. The Senate Report, see ante, at 151, explained that the new provision requiring consideration of combined impairments would not authorize a departure from the sequential evaluation process. Omitted from the heart of the Court’s quotation, however, is the Report’s express incorporation by reference of the 1967 interpretation. The Report ability determination process at which a claimant could be deemed ineligible. In Heckler n. Campbell, 461 U. S. 458 (1983), the Court upheld a significant part of the sequential evaluation process, but step two was not before it in that case and in fact was not even mentioned in the description of the current process. 12 The only substantive amendment to these sections since 1967 was in 1983 when § 423(d)(2) was amended to substitute “widower, or surviving divorced husband” for “or widower” throughout that paragraph. 97 Stat. 117. BOWEN v. YUCKERT 175 137 Blackmun, J., dissenting explained: “As the Committee stated in its report on the 1967 amendments, an individual is to be considered eligible ‘only if it is shown that he has a severe medically determinable physical or mental impairment or impairments.’” S. Rep. No. 98-466, p. 22 (1984) (emphasis added). Reference back to the congressional views supporting the 1967 Amendments evinces an intent to adhere to a consistent interpretation of that provision. For the reasons discussed above, the 1967 view necessarily considered the vocational factors to be a critical part of a disability determination in cases where the insured worker cannot do his previous work. This view stands in contradiction to the Senate Report’s apparent suggestion that the Secretary can deny benefits in such cases based on medical evidence alone. Hence, the Senate’s discussion of the disability determination process is ambiguous at best. The House Report accompanying the 1984 Act reflects dissatisfaction with the step two severity regulation. According to the House Report, under that process, “a determination that a person is not disabled may be based on a judgment that the person has no impairment, or that the impairment or combination of impairments are slight enough to warrant a presumption that the person’s work ability is not seriously affected.” H. R. Rep. No. 98-618, p. 8 (1984). While stating that it did not wish to undermine the Secretary’s entire step-evaluation process, the House Report nevertheless expressed reservations about the “slight impairment” approach as a threshold assessment at step two. It explained: “[T]he committee is concerned that the consideration of eligibility for disability benefits be conducted using criteria that clearly reflect the intent of Congress that all those who are unable to work receive benefits. It is of particular concern that the Social Security Administration has been criticized for basing terminations of benefits solely and erroneously on the judgment that the person’s medical impairment is ‘slight,’ according to very 176 OCTOBER TERM, 1986 Blackmun, J., dissenting 482 U. S. strict criteria, and is therefore not disabling, without making any further evaluation of the person’s ability to work.” Id., at 7. After stating that it did not wish to eliminate the sequential evaluation process, it continued: “However, the committee notes that the Secretary has already planned to re-evaluate the current criteria for non-severe impairments [i. e. step two], and urges that all due consideration be given to revising those criteria to reflect the real impact of impairments upon the ability to work” (emphasis added). Id., at 8. Hence, not only did the House Report read the current stepevaluation process as setting forth a “slight impairment” standard that was less onerous than the standard discussed in the Senate Report, but it also expressed concern that even that threshold step did not provide the necessary individualized consideration of a disability claim to determine the actual impact of the impairment on the individual’s ability to work. The House thus indicated a desire not to upset the Secretary’s step-evaluation process, but it did not approve the step two severity regulation. The Conference Report adopted the position set forth in the House Report. It referred to the Secretary’s “plan to reevaluate the current criteria for nonsevere impairments” and to the expectation that the Secretary would apprise Congress of the results of that evaluation. H. R. Conf. Rep. No. 98-1039, p. 30 (1984). Moreover, the description of the sequential evaluation process in the Conference Report is even more lenient than the House Report. The conferees approved of the flexibility and efficiency resulting from a threshold disability determination but indicated that “a determination that an individual is not disabled may be based on a judgment that an individual has no impairment, or that the medical severity of his impairment or combination of impairments is slight enough to warrant a BOWEN v. YUCKERT 177 137 Blackmun, J., dissenting presumption, even without a full evaluation of vocational factors, that the individual’s ability to perform [substantial gainful activity] is not seriously affected” (emphasis added). Ibid. The conferees stated that the current sequential evaluation process permitted that determination and they did not intend to eliminate the process. Ibid. This characterization of the process as permitting less than a full evaluation of the vocational factors indicates that the appropriate standard would include an implicit or limited analysis of vocational factors. Because the agency’s regulation states expressly that vocational factors will not be considered, however, the conferees’ statement can serve only as a description of what they believed a valid threshold standard would be, rather than as a description of the current severity regulation. The ambiguity in the congressional references to step two is understandable due to the fact that Congress did not have before it the question of that regulation’s validity. Examination of the totality of the legislative history of the 1984 Act reveals that Congress limited its focus to several major problems in the Social Security system. These problems included the standard of review for termination of disability benefits, for evaluating pain, for ensuring consideration of multiple impairments, and for evaluating the effect of mental impairments on ability to work. In sum, Congress acknowledged that the Secretary was in the midst of reevaluating the severity regulation and indicated its willingness to await the Secretary’s results rather than to address the matter in the midst of the overwhelming legislative task it already faced regarding the matters properly before it. The brief remarks about the step-evaluation process simply cannot be read as an endorsement of the facial validity of the severity regulation. These congressional comments in 1984 cannot outweigh the clear language of §§ 423(d)(1)(A) and (2)(A) and the legislative history of those provisions. 178 OCTOBER TERM, 1986 Blackmun, J., dissenting 482 U. S. Ill The Court makes much of the Secretary’s broad authority to prescribe standards for applying the Social Security Act and the limited nature of our review in light of that authority. Ante, at 145. This Court has recognized, however, that “[c]ourts need not defer to an administrative construction of a statute where there are ‘compelling indications that it is wrong.’” Espinoza v. Farah Mfg. Co., 414 U. S. 86, 94-95 (1973), quoting Red Lion Broadcasting Co. v. FCC, 395 U. S. 367, 381 (1969). When a regulation is facially inconsistent with the statute, the administrative construction of the statute is necessarily wrong and there is no need to consider further the position of the agency. The Secretary’s interpretation of the statute as reflected in his regulation “cannot supersede the language chosen by Congress.” Mohasco Corp. v. Silver, 447 U. S. 807, 825 (1980). Unlike the situation presented recently in Lukhard v. Reed, 481 U. S. 368 (1987), Congress unambiguously specified its intent when enacting § 423(d) that the vocational factors be considered in determining disability eligibility in cases such as respondent’s. The efficiency and reliability interests that the Court attributes to the Secretary,13 ante, at 151-152, cannot outweigh clear congressional intent. The Secretary attempts to avoid the facial contradiction between his severity regulation and the statute by interpreting the regulation as representing only a de minimis threshold standard. The Secretary apparently has recognized finally what every Federal Court of Appeals has concluded— application of a threshold severity regulation that is greater 13 The extensive litigation concerning the meaning of step two belies the assertion that it fosters efficiency. See ante at 145, n. 4. Justice O’Connor describes the evidence indicating that step two has proved to be a very unreliable indicator of disability eligibility. See ante, at 157-158. While a clearly drafted regulation encompassing a valid preliminary screening standard undoubtedly could increase efficiency and reliability, the current step two advances neither. BOWEN v. YUCKERT 179 137 Blackmun, J., dissenting than de minimis is invalid under the terms of the statute. See concurring opinion ante, at 156, and nn. 1 and 2. The Court explains that it has not considered the validity of the Secretary’s application of the regulation, ante, at 150, n. 8, although it appears to adopt the “slight” impairment interpretation. See ante, at 153. In her concurring opinion, Justice O’Connor expressly imposes on the severity regulation a narrowing interpretation that permits only a de minimis threshold standard. I cannot, however, join that approach in this case. I agree with respondent’s position that, although a de minimis standard that implicitly draws the vocational factors into the disability determination may be permitted under the statute, this Court cannot resolve that question on the record in this case. Such a standard was not applied by the agency adjudicators who reviewed respondent’s claim, and there is no record evidence as to the Secretary’s application of a de minimis standard subsequent to the 1978 adoption of the sequential evaluation. Indeed, Justice O’Connor aptly demonstrates that even if the Secretary is currently attempting to readopt the pre-1978 slight impairment standard, that standard is entirely inconsistent with the interpretation in effect at the time respondent’s claim was considered by the agency adjudicators. I agree with Justice O’Connor that the evidence suggests that step two has been “applied systematically in a manner inconsistent with the statute.” Ante, at 157; see also Stone v. Heckler, 752 F. 2d 1099 (CA5 1985). Little weight can be given to views of an agency when the views themselves are inconsistent. See United Housing Foundation, Inc. v. Forman, 421 U. S. 837, 858, n. 25 (1975). Whether a de minimis threshold standard is authorized under the statute is not before this Court. The regulation on its face simply does not describe a standard that incorporates into the threshold step an implicit consideration of the vocational factors. The language of step two does not represent 180 OCTOBER TERM, 1986 Blackmun, J., dissenting 482 U. S. a standard that denies disability claims only if the medical impairment is so minimal that no set of vocational factors, even if fully considered, could result in a finding of disability. Yet, in order to be valid under the terms of the statute, any de minimis threshold step would have to adopt such a standard. It would have to ensure that it did not preclude an individual evaluation of vocational factors at a later stage14 and a finding of disability if they affect the ultimate determination of that issue. I agree with the approach of the Court of Appeals in this case. Contrary to this Court’s implications, ante, at 147, that court did not address the question whether the statute authorizes a threshold showing of medical severity. The Court of Appeals addressed only the facial validity of the severity standard in step two of the sequential evaluation process. It expressly declined to consider whether other threshold severity standards, such as a de minimis standard, would be authorized under the statute. See Yuckert n. Heckler, 774 F. 2d 1365, 1369, n. 6 (CA9 1985). Invalidating step two does not prohibit the adoption of a threshold screening stand 14 Application of a greater than de minimis threshold severity standard can render the step-evaluation process internally inconsistent by denying benefits to claimants who would be found to be disabled under the criteria of a more advanced step in the sequential evaluation process. For example, under the medical-vocational guidelines that are applied at step five, the only impairments that will never be found to be disabling regardless of age, education, and work experience, are those that do not prevent the claimant from engaging in heavy work and do not impose nonexertional restrictions. Yet persons who are unable to perform heavy work have been found to be not disabled at step two at the administrative level, and the courts have had to reverse those initial findings. See Brief for American Diabetes Association et al. as Amici Curiae 8, 13-14 (profiling cases). The Secretary also has argued in the past that even if a claimant’s impairment meets the requirements in the Listing of Impairments used at step three of the process and would have been found to be disabling at that step, that fact is irrelevant if the claimant is found to be not disabled at the threshold step two standard. See Williamson v. Secretary of Health and Human Services, 796 F. 2d 146, 150 (CA6 1986). BOWEN v. YUCKERT 181 137 Blackmun, J., dissenting ard to eliminate frivolous claims at an early stage in the process if that standard takes into account vocational factors as required by the statute. Adoption of such a standard should take place through the administrative procedures required under the Act for the adoption of new regulations. See 42 U. S. C. § 421(k)(2) (1982 ed., Supp. III). Further agency interpretations of the invalid regulation are of no value. They cannot alter the fact that the regulation is facially invalid. IV Because the Secretary’s regulation directly conflicts with the statutory language set forth by Congress and because it plainly is inconsistent with the legislative history, it is highly inappropriate for this Court to permit the Secretary to continue to enforce that regulation. I dissent. 182 OCTOBER TERM, 1986 Syllabus 482 U. S. ROCKFORD LIFE INSURANCE CO. v. ILLINOIS DEPARTMENT OF REVENUE ET AL. APPEAL FROM THE SUPREME COURT OF ILLINOIS No. 86-251. Argued March 31, 1987—Decided June 8, 1987 Under financial instruments commonly known as “Ginnie Maes,” the issuing private financial institution has the primary obligation of making timely principal and interest payments. However, in order to attract investors into the private mortgage market, Ginnie Maes also contain a provision whereby the Government National Mortgage Association, a Government corporation, guarantees payment if the issuer defaults. After state taxing officials included the value of appellant’s Ginnie Mae portfolio in calculating net assets, appellant filed suit challenging its annual property tax assessment. The state courts rejected appellant’s contention that the Ginnie Maes could not be taxed under the constitutional principle of intergovernmental tax immunity and under Revised Statutes §3701, which exempts from state taxation “all stocks, bonds, Treasury notes, and other obligations of the United States.” Held: Ginnie Maes are not exempt from state taxation under § 3701. The statutory phrase “other obligations of the United States” refers only to obligations or securities of the same type as those specifically enumerated. Ginnie Maes are fundamentally different from the enumerated instruments in that the Government’s obligation as guarantor is secondary and contingent. Nor is the indirect, contingent, and unliquidated promise that the Government makes in Ginnie Maes the type of obligation that is protected by the constitutional principle of intergovernmental tax immunity. The purpose of that principle is to prevent States from taxing federal obligations in a manner which has an adverse effect on the United States’ borrowing ability. Ginnie Maes’ failure to include a binding governmental promise to pay specified sums at specified dates renders any threat to the federal borrowing power far too attenuated to support constitutional immunity. Pp. 187-192. 112 Ill. 2d 174, 492 N. E. 2d 1278, affirmed. Stevens, J., delivered the opinion for a unanimous Court. Erwin N. Griswold argued the cause for appellant. With him on the briefs were Karl L. Kellar, Ira L. Burleson, and John C. McCarthy. ROCKFORD LIFE INS. CO. v. ILL. DEPT. OF REV. 183 182 Opinion of the Court Patricia Rosen, Assistant Attorney General of Illinois, argued the cause for appellees. With her on the brief for appellees Illinois Department of Revenue et al. were Neil F. Hartigan, Attorney General, and Roma Jones Stewart, Solicitor General. Charles J. Prorok filed a brief for appellee Aurand.* Justice Stevens delivered the opinion of the Court. This case involves financial instruments commonly known as “Ginnie Maes.” These instruments are issued by private financial institutions, which are obliged to make timely payment of the principal and interest as set forth in the certificates. The Government National Mortgage Association (GNMA) guarantees that the payments will be made as scheduled. The question presented today is whether these instruments are exempt from state taxation under the constitutional principle of intergovernmental tax immunity, or under the relevant immunity statute.1 *Briefs of amici curiae urging affirmance were filed for the United States by Solicitor General Fried, Assistant Attorney General Olsen, Alan I. Horowitz, Michael L. Paup, and Ernest J. Brown; for the National Governor’s Association et al. by Benna Ruth Solomon, Beate Bloch, and Alan S. Madans; and for the California Franchise Tax Board by Benjamin F. Miller and Anna Jovanovich. 1 At the time relevant to this case, that statute was Rev. Stat. § 3701, as amended, and provided: “Except as otherwise provided by law, all stocks, bonds, Treasury notes, and other obligations of the United States, shall be exempt from taxation by or under State or municipal or local authority. This exemption extends to every form of taxation that would require that either the obligations or the interest thereon, or both, be considered, directly or indirectly, in the computation of the tax, except nondiscriminatory franchise or other nonproperty taxes in lieu thereof imposed on corporations and except estate taxes or inheritance taxes.” 31 U. S. C. § 742 (1976 ed.). The 1982 reformulation of the statute was “without substantive change” see Pub. L. 97-258, § 4(a), 96 Stat. 1067, and now appears at 31 U. S. C. § 3124(a) with some minor variations in its language, which are not relevant to this case. As in First National Bank of Atlanta v. Bartow County Tax Assessors, 470 U. S. 583 (1985), the tax at issue here was levied prior to 184 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. Prior to 1979 changes in Illinois’ tax law, Rockford Life Insurance Company (Rockford) paid an annual property tax on the assessed value of its capital stock. In 1978, the Illinois taxing authorities included the value of Rockford’s portfolio of Ginnie Maes in their calculation of the corporation’s net assets. Rockford challenged the assessment in the Illinois courts and the County Treasurer filed an action to collect the full amount of the assessment ($723,053.70). The Illinois courts uniformly rejected Rockford’s contention that the securities were exempt from state property taxes,2 reasoning that “the securities in question here were not ‘other obligations of the United States’ within the meaning of § 3701,” and that the constitutional and statutory inquiries were identical in this case. 112 Ill. 2d 174, 176-184, 492 N. E. 2d 1278, 1279-1283 (1986). We noted probable jurisdiction,3 479 U. S. 947 (1986), and now affirm. I The instruments involved here are standard securities bearing the title “Mortgage Backed Certificate Guaranteed by Government National Mortgage Association.” App. 56. the recodification, and “the pre-1982 form of the statute technically controls this case.” Id., at 585, n. 1. 2 Appellant’s state-court action also involved a variety of state-law claims, and claims that some other federally guaranteed securities were exempt from state taxation. See 112 Ill. 2d 174, 177, 185-187, 492 N. E. 2d 1278, 1279, 1283-1284 (1986). These claims are not at issue here. 3 The issue presented is not the type that would usually merit our attention if presented in a petition for certiorari. The issue has divided neither the federal courts of appeals nor the state courts. Indeed, aside from the Illinois courts, no court has ever considered whether Ginnie Maes are exempt from state taxes. Nor does it appear that this case presents an overly important question of federal law “which has not been, but should be, settled by this Court. ” This Court’s Rule 17.1(c). The fact is that the Illinois property tax imposed here was repealed in 1979. Nonetheless, this case arises under our mandatory jurisdiction, 28 U. S. C. § 1257(2), and Congress has not allowed us to consider these factors in deciding whether to rule on this case on its merits. ROCKFORD LIFE INS. CO. v. ILL. DEPT. OF REV. 185 182 Opinion of the Court True to that title, the instruments contain a provision in which GNMA pledges the “full faith and credit of the United States” to secure the timely payment of the interest and principal set forth in the instrument. The purpose of the guarantee, and the function of GNMA, which is a wholly owned government corporation within the Department of Housing and Urban Development, is to attract investors into the mortgage market by minimizing the risk of loss.4 See 12 U. S. C. § 1716(a). There is uncontradicted evidence in the record supporting the conclusion that GNMA’s guarantee is responsible for the ready marketability of these securities. That guarantee is not the primary obligation described in the instrument, however. The duty to make monthly payments of principal and interest to the investors falls squarely on the issuer of the certificate.5 4 “The Mortgage-Backed Securities Program provides a means for channeling funds from the Nation’s securities markets into the housing market. The U. S. Government full faith and credit guaranty of securities makes them widely accepted in those sectors of the capital markets that otherwise would not be likely to supply funds to the mortgage market. The funds raised through the securities issued are used to make residential and other mortgage loans. Through this process, the program serves to increase the overall supply of credit available for housing and helps to assure that this credit is available at reasonable interest rates.” Dept, of Housing and Urban Development, Handbook GNMA 5500.1 Rev. 6, GNMA I Mortgage Backed Securities Guide 1-1 (1984) (hereinafter GNMA Guide). 6 The promises set forth in the representative GNMA certificate in the record read as follow: “THE ISSUER, NAMED BELOW, PROMISES TO PAY TO THE ORDER OF: “ROCKFORD LIFE INSURANCE COMPANY “36 1695690 F “(HEREINAFTER CALLED THE HOLDER) The sum of $1,018,717 DOLS 20 CTS in principal amount, together with interest thereon and on portions thereof outstanding from time to time at the fixed rate set forth hereon, such payment to be in monthly installments, adjustable as set forth below. All monthly installments shall be for application first to interest at such fixed rate and then in reduction of principal balance then outstanding, 186 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. The issuer of the certificate is a private party, generally a financial institution, that posesses a pool of federally guaranteed mortgages.6 Those individual mortgages are the product of transactions between individual borrowers and private lending institutions. It is this pool of private obligations that provides the source of funds, as well as the primary security, for the principal and interest that the issuer promises to pay to the order of the holder of the instrument. After a pool of qualified mortgages is assembled by a qualified issuer, the issuer enters into an agreement with GNMA authorizing the issuer to sell one or more certificates, each of which is proportionately based on and backed by all the mortgages in the designated pool, and each of which is also guaranteed by GNMA. The issuer thereafter may sell the “mortgage-backed certificates” to holders such as Rockford. The issuer administers the pool by collecting principal and interest from and shall continue until payment in full of the principal amount, and of all interest accruing thereon. “[T]he issuer shall pay to the holder, whether or not collected by the issuer, and shall remit as set forth below, monthly payments of not less than the amounts of principal being due monthly on the mortgages and apportioned to the holder by reason of the aforesaid base and backing, together with any apportioned prepayments or other early recoveries of principal and interest at the fixed rate.” App. 56-57. Sample certificates are published in the GNMA Guide, at App. 39-43. The Ginnie Maes held by Rockford, are “modified pass-through securities” that provide for the payment of specific amounts whether or not timely collections are made from the individual mortgagors in the pool. See 128 Ill. App. 3d 302, 313, 470 N. E. 2d 596, 603 (2d Dist. 1984). GNMA also guarantees “straight pass-through securities” which provide that the issuer shall pay the holders of the securities the amounts collected from the pool, “as collected,” less specified administrative costs. See 24 CFR § 390.5(a) (1986); GNMA Guide, at 1-1. 6 The issuer must satisfy various financial requirements imposed by the Federal Housing Authority (FHA) and GNMA. See 24 CFR § 390.3 (1986). In addition each of the individual mortgages in the pool must be guaranteed by the FHA, the Veterans Administration, or another Government agency. Ibid. ROCKFORD LIFE INS. CO. v. ILL. DEPT. OF REV. 187 182 Opinion of the Court the individual mortgagors and remitting the amounts specified in the certificates to the holders. GNMA’s costs for the regulatory duties is covered by a fee charged to the issuer. Unless the issuer defaults in its payments to the holder of a certificate, no federal funds are used in connection with the issuance and sale of these securities, the administration of the pool of mortgages, or the payments of principal and interest set forth in the certificates. Under the type of Ginnie Maes involved in this case, see n. 5, supra, the issuer is required to continue to make payments to the holders even if an individual mortgage in the pool becomes delinquent. In such event, the issuer may pursue its remedies against the individual mortgagor, or the guarantor of the mortgage, but the issuer does not have any rights against GNMA. GNMA’s guarantee is implicated only if the issuer fails to meet its obligations to the holders under the certificates. In that event the holder proceeds directly against GNMA, and not against the issuer. But the risk of actual loss to GNMA is minimal because its guarantee is secured not only by the individual mortgages in the pool but also by the separate guarantee of each of those mortgages, and by a fidelity bond which the issuer is required to post. See 24 CFR § 390.1 (1986). II The GNMA guarantee of payment that is contained in the mortgage-backed certificates held by Rockford is a pledge of the “full faith and credit of the United States.”7 But that does not mean that it is the type of “obligation” of the United States which is subject to exemption under the Constitution or the immunity statute. Because the statutory immunity 7 GNMA is authorized to make this guarantee under 12 U. S. C. § 1721(g). The fact that the guarantee is executed by a federal agency, rather than by the United States itself, does not avoid the application of the immunity doctrine and statute. See Memphis Bank & Trust v. Garner, 459 U. S. 392, 396 (1983). 188 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. provision now codified at 31 U. S. C. § 3124(a) is “principally a restatement of the constitutional rule,” see Memphis Bank & Trust Co. v. Gamer, 459 U. S. 392, 397 (1983), we shall first decide whether the statute requires that Ginnie Maes be exempted from state property taxes, and then consider whether the constitutional doctrine of intergovernmental tax immunity requires any broader exemption. At the time relevant to this case,8 Rev. Stat. §3701, as amended, 31 U. S. C. §742 (1976 ed.), provided that “all stocks, bonds, Treasury notes, and other obligations of the United States, shall be exempt from taxation by or under State or municipal or local authority” (emphasis added). The full text of the sentence in which these words appear, rules of statutory construction, and the earlier legislation that was codified by the enactment of this statute, are all consistent with the conclusion that the phrase “other obligations” refers “only to obligations or securities of the same type as those specifically enumerated.” Smith v. Davis, 323 U. S. Ill, 117 (1944). This longstanding interpretation resolves the statutory question before us. GNMA certificates are fundamentally different from the securities specifically named in the statute. Most significantly, they are neither direct nor certain obligations of the United States. As the certificate provides, it is the issuer that bears the primary obligation to make timely payments—the United States’ obligation is secondary and contingent.9 In short, the United States is the 8 See n. 1, supra. 9 Appellant contends that the issuer is not an obligor at all because the certificate provides that the holder’s sole recourse is against the GNMA. We disagree. That GNMA is willing to pay the investor in case of default and then pursue its own remedies against the issuer does not detract from the reality that the primary obligor is in fact the issuer, and not the GNMA. While the holder of the certificate may not enforce the obligation through a direct action against the issuer, GNMA may, upon default, institute a claim against the issuer’s fidelity bond or extinguish the issuer’s interest in the underlying mortgages thereby making the mortgages the absolute property of GNMA “subject only to unsatisfied rights therein of ROCKFORD LIFE INS. CO. v. ILL. DEPT. OF REV. 189 182 Opinion of the Court guarantor—not the obligor. This distinction is more than adequate to support our conclusion that Ginnie Maes do not qualify as “other obligations of the United States” for the purposes of this statute. Nor does the constitutional doctrine of intergovernmental tax immunity exempt these instruments from state property taxes. In Smith v. Davis, supra, the United States owed money to a construction company for work that the company had performed on open account. In computing its assets for state tax purposes, the company sought to exclude the amount owed to it by the Federal Government, but a unanimous Court held that the debt was not exempt. The Court concluded that “a unilateral, unliquidated creditor’s claim, which by itself does not bind the United States and which in no way increases or affects the public debt, cannot be said to be a credit instrumentality of the United States for the purposes of tax immunity,” 323 U. S., at 114, and went on to explain that the claim differed “vitally from the type of credit instrumentalities which this Court in the past has recognized as constitutionally exempt from state and local taxation. Such instrumentalities in each instance have been characterized by (1) written documents, (2) the bearing of interest, (3) a binding promise by the United States to pay specified sums at specified dates and (4) specific Congressional authori- the holders of the securities.” 24 CFR §390.15(b) (1986); see also 12 U. S. C. § 1721(g); New York Guardian Mortgagee Corp. n. Cleland, 473 F. Supp. 409, 411 (SDNY 1979). As the GNMA Guide provides: An “issuer of GNMA-guaranteed mortgage-backed securities is responsible for • . . making the full and timely payment of all amounts due to securities holders.” GNMA Guide, at 2-1. This statement is supported by the regulations which prohibit GNMA from guaranteeing securities “if the pool arrangement proposed by the issuer does not satisfactorily provide for . . . [t]imely payment of principal and interest, in accordance with the terms of the guaranteed securities.” 24 CFR § 390.9(c) (1986). See also GNMA Guide, at App. 19, §4.01 (issuer’s contractual agreement with GNMA binds issuer to “remit to the holders all payments ... in a timely manner”). 190 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. zation, which also pledged the full faith and credit of the United States in support of the promise to pay.” Id., at 114-115. With respect to Ginnie Maes, the third element described in Smith v. Davis is clearly lacking, and its absence is critical in view of the purposes behind the intergovernmental tax immunity doctrine. That doctrine is based on the proposition that the borrowing power is an essential aspect of the Federal Government’s authority and, just as the Supremacy Clause bars the States from directly taxing federal property, it also bars the States from taxing federal obligations in a manner which has an adverse effect on the United States’ borrowing ability. See Weston v. City Council of Charleston, 2 Pet. 449 (1829); McCulloch v. Maryland, 4 Wheat. 316 (1819). The lack of a fixed and certain obligation by the United States in the Ginnie Mae context makes this concern far too attenuated to support constitutional immunity.10 Cf. Willcuts v. Bunn, 282 U. S. 216, 225 (1931); Hibernia Sav- 10 “But when effort is made, as is the case here, to establish the un-constititional character of a particular tax by claiming that its remote effect will be to impair the borrowing power of the government, courts in overturning statutes, long established and within the ordinary sphere of state legislation, ought to have something more substantial to act upon than mere conjecture. The injury ought to be obvious and appreciable.” Plummer v. Coler, 178 U. S. 115, 137-138 (1900). The proposition that a federal guarantee of a loan does not preclude state taxation is a “long established” one. See Board of Commas of Montgomery County v. Elston, 32 Ind. 27, 32 (1869); S.S. Silberblatt, Inc. n. Tax Comm’n of New York, 5 N. Y. 2d 635, 641, 159 N. E. 2d 195, 197-198, cert, denied, 361 U. S. 912 (1959); see also 47 Op. N. C. Atty. Gen. 19 (1977) (concluding that Ginnie Maes are not “obligation of the United States” for these purposes, and indicating that the Assistant Director of GNMA agreed with this position). In fact, during the debate on one of the predecessors to the current immunity statute, Senator Sherman assured the Senate that bonds of the Pacific Railroad, which had been guaranteed by the United States, were not subject to immunity. See Cong. Globe, 41st Cong., 2d Sess., 1591 (1870), discussing Act of July 14, 1870, 16 Stat. 272. ROCKFORD LIFE INS. CO. v. ILL. DEPT. OF REV. 191 182 Opinion of the Court ings Society v. San Francisco, 200 U. S. 310, 315 (1906); Plummer v. Coler, 178 U. S. 115, 136 (1900). Moreover, none of the proceeds of the issuance and sale of the GNMA certificates are received by the Federal Government or used to finance any governmental function. Indeed, given the fixed fees that GNMA charges issuers, and the lack of any GNMA profit sharing, it has not been suggested here that the federal fisc would at all benefit from a holding that Ginnie Maes are exempt from state taxation.11 Appellant asserts that Congress authorized the GNMA’s guarantee for the salutary purpose of facilitating the financing of private mortgages, and that an exemption from state taxation will further this purpose. But our job is neither to assess the underlying merits of the program, nor to opine on whether Congress would be wise to exempt Ginnie Maes from state taxation. Our task is simply to decide whether the indirect, contingent, and unliquidated promise that GNMA is authorized to make is the type of federal obligation for which the Constitution, in Congress’ silence, imposes an exemption from state taxation. We hold that it is not. Ill A court must proceed carefully when asked to recognize an exemption from state taxation that Congress has not clearly established. We do well to remember the concluding words in Smith, which although spoken in reference to the statute, are relevant to our role in applying the constitutional doctrine as well: “All of these related statutes are a clear indication of an intent to immunize from state taxation only the 11 Even if there were a somewhat more certain effect, state taxation of these privately issued instruments would not necessarily be invalid. See Alabama v. King & Boozer, 314 U. S. 1 (1941); Graves v. New York ex rel. O’Keefe, 306 U. S. 466 (1939); James v. Dravo Contracting Co., 302 U. S. 134 (1937). Immunity from taxation “may not be conferred simply because the tax has an effect on the United States.” United States v. New Mexico, 455 U. S. 720, 734 (1982). 192 OCTOBER TERM, 1986 482 U. S. Opinion of the Court interest-bearing obligations of the United States which are needed to secure credit to carry on the necessary functions of government. That intent, which is largely codified in §3701, should not be expanded or modified in any degree by the judiciary.” 323 U. S., at 119. The judgment is Affirmed. UTAH DIV. OF STATE LANDS v. UNITED STATES 193 Syllabus UTAH DIVISION OF STATE LANDS v. UNITED STATES ET al. CERTIORARI TO THE UNITED STATES COURT OF APPEALS FOR THE TENTH CIRCUIT No. 85-1772. Argued March 23, 1987—Decided June 8, 1987 After the Federal Government, in 1976, issued oil and gas leases for lands underlying Utah Lake, a navigable body of water located in Utah, the State brought suit in Federal District Court for injunctive relief and a declaratory judgment that it, rather than the United States, had title to the lakebed under the equal footing doctrine. Under that doctrine, the United States holds the lands under navigable waters in the Territories in trust for the future States, and, absent a prior conveyance by the Federal Government to third parties, a State acquires title to such lands upon entering the Union on an “equal footing” with the original 13 States. The Utah Enabling Act of 1894 provided that Utah was to be so admitted. The United States answered in the District Court that title to the lakebed remained in federal ownership by operation of a United States Geological Survey official’s selection of the lake as a reservoir site in 1889 pursuant to an 1888 Act that provided that all lands which might be so selected were reserved as the property of the United States and were not subject to entry, settlement, or occupation. Although the 1888 Act was repealed in 1890, the 1890 Act provided that “reservoir sites heretofore located or selected shall remain segregated and reserved from entry or settlement as provided by [the 1888 Act].” The District Court granted summary judgment for the United States, and the Court of Appeals affirmed. Held: Title to Utah Lake’s bed passed to Utah under the equal footing doctrine upon Utah’s admission to the Union. Pp. 200-209. (a) Even assuming, arguendo, that a federal reservation of the lakebed—as opposed to a conveyance by the Federal Government to a third party—could defeat Utah’s claim to title under the equal footing doctrine, such defeat was not accomplished on the facts here. There is a strong presumption against finding congressional intent to defeat a State’s title, and, in light of the longstanding policy of the Federal Government’s holding land under navigable waters for the ultimate benefit of future States absent exceptional circumstances, an intent to defeat a State’s equal footing entitlement could not be inferred from the mere act of reservation itself. The United States would not merely be required to establish that Congress clearly intended to include land under navi 194 OCTOBER TERM, 1986 Syllabus 482 U. S. gable waters within the federal reservation, but would additionally have to establish that Congress affirmatively intended to defeat the future State’s title to such land. Pp. 200-202. (b) The 1888 Act fails to make sufficiently plain a congressional intent to include the bed of Utah Lake within the Federal Government’s reservation. The Act’s language did not necessarily refer to lands under navigable waters, which lands were already the property of the United States, and were already exempt from sale, entry, settlement, or occupation under the general land laws. Moreover, the concerns that motivated Congress to enact the statute—concerns as to homesteaders’ possible monopolization of and speculation in arid lands suitable for reservoir sites or irrigation works—had nothing to do with the beds of navigable waters. There is no merit to the Federal Government’s contention that, in view of remarks made by the Geological Survey in reserving Utah Lake, Congress’ enactment of the 1890 Act ratified the Survey’s reservation of the lakebed. The Survey’s references to the “segregation” of the lakebed, placed in the proper context, could refer to the segregation of the lands adjacent to the lake. Moreover, neither the language nor the legislative history of the 1890 Act supports the conclusion that Congress intended to ratify a reservation of the lakebed. Pp. 202-207. (c) Even assuming that Congress did intend to reserve the lakebed in either the 1888 Act or the 1890 Act, Congress did not clearly express an intention to defeat Utah’s claim to the lakebed under the equal footing doctrine upon entry into statehood. The 1888 Act’s structure and history strongly suggest that Congress had no such intent. Moreover, the transfer of title of the lakebed to Utah would not necessarily prevent the Federal Government from subsequently developing a reservoir or water reclamation project at the lake in any event. The broad sweep of the 1888 Act, which had the practical effect of reserving all of the public lands in the West from settlement, cannot be reconciled with an intent to defeat the States’ title to the land under navigable waters under the equal footing doctrine. Pp. 208-209. 780 F. 2d 1515, reversed. O’Connor, J., delivered the opinion of the Court, in which Rehnquist, C. J., and Blackmun, Powell, and Scalia, JJ., joined. White, J., filed a dissenting opinion, in which Brennan, Marshall, and Stevens, JJ., joined, post, p. 209. Dallin W. Jensen, Solicitor General of Utah, argued the cause for petitioner. With him on the briefs were David L. Wilkinson, Attorney General, and Michael M. Quealy and R. Douglas Credille, Assistant Attorneys General. UTAH DIV. OF STATE LANDS v. UNITED STATES 195 193 Opinion of the Court Edwin S. Kneedler argued the cause for respondents. With him on the brief were Solicitor General Fried, Assistant Attorney General Habicht, Deputy Solicitor General Wallace, Jacques B. Gelin, and Dirk D. Snel* Justice O’Connor delivered the opinion of the Court. The issue in this case is whether title to the bed of Utah Lake passed to the State of Utah under the equal footing doctrine upon Utah’s admission to the Union in 1896. I A The equal footing doctrine is deeply rooted in history, and the proper application of the doctrine requires an understanding of its origins. Under English common law the English Crown held sovereign title to all lands underlying navigable waters. Because title to such land was important to the sovereign’s ability to control navigation, fishing, and other commercial activity on rivers and lakes, ownership of this land was considered an essential attribute of sovereignty. *A brief of amici curiae urging reversal was filed for the State of Alaska et al. by Ronald W. Lorensen, Acting Attorney General of Alaska, and G. Thomas Koester, Assistant Attorney General, and by the Attorneys General for their respective jurisdictions as follows: Charles A. Graddick of Alabama, Robert K. Corbin of Arizona, John Steven Clark of Arkansas, John Van de Kamp of California, Duane Woodard of Colorado, Jim Smith of Florida, Michael J. Bowers of Georgia, Corinne K A. Watanabe of Hawaii, Jim Jones of Idaho, Neil F. Hartigan of Illinois, Linley E. Pearson of Indiana, Thomas J. Miller of Iowa, Robert T. Stephan of Kansas, William J. Guste, Jr., of Louisiana, Francis X. Bellotti of Massachusetts, Frank J. Kelley of Michigan, Edwin Lloyd Pittman of Mississippi, William L. Webster of Missouri, Mike Greely of Montana, Robert M. Spire of Nebraska, Brian McKay of Nevada, Stephen E. Merrill of New Hampshire, Paul Bardacke of New Mexico, Lacy H. Thornburg of North Carolina, Nicholas J. Spaeth of North Dakota, Michael C. Turpen of Oklahoma, Dave Frohnmayer of Oregon, Jim Mattox of Texas, Ken Eikenberry of Washington, Charles G. Brown of West Virginia, Bronson C. La Follette of Wisconsin, and Archie G. McClintock of Wyoming. 196 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. Title to such land was therefore vested in the sovereign for the benefit of the whole people. See Shively v. Bowlby, 152 U. S. 1, 11-14 (1894). When the 13 Colonies became independent from Great Britain, they claimed title to the lands under navigable waters within their boundaries as the sovereign successors to the English Crown. Id., at 15. Because all subsequently admitted States enter the Union on an “equal footing” with the original 13 States, they too hold title to the land under navigable waters within their boundaries upon entry into the Union. Pollard’s Lessee v. Hagan, 3 How. 212 (1845). In Pollard’s Lessee this Court announced the principle that the United States held the lands under navigable waters in the Territories “in trust” for the future States that would be created, and in dicta even suggested that the equal footing doctrine absolutely prohibited the United States from taking any steps to defeat the passing of title to land underneath navigable waters to the States. Id., at 230. Half a century later, however, the Court disavowed the dicta in Pollard’s Lessee, and held that the Federal Government had the power, under the Property Clause, to convey such land to third parties: “By the Constitution, as is now well settled, the United States, having rightfully acquired the Territories, and being the only government which can impose laws upon them, have the entire dominion and sovereignty, national and municipal, Federal and state, over all the Territories, so long as they remain in territorial condition. . . . “We cannot doubt, therefore, that Congress has the power to make grants of lands below high water mark of navigable waters in any Territory of the United States, whenever it becomes necessary to do so in order to perform international obligations, or to effect the improvement of such lands for the promotion and convenience of commerce with foreign nations and among the several UTAH DIV. OF STATE LANDS v. UNITED STATES 197 193 Opinion of the Court States, or to carry out other public purposes appropriate to the objects for which the United States hold the Territory.” Shively v. Bowlby, 152 U. S., at 48. Thus, under the Constitution, the Federal Government could defeat a prospective State’s title to land under navigable waters by a prestatehood conveyance of the land to a private party for a public purpose appropriate to the Territory. The Court further noted, however, that Congress had never undertaken by general land laws to dispose of land under navigable waters. Ibid. From this, the Court inferred a congressional policy (although not a constitutional obligation) to grant away land under navigable waters only “in case of some international duty or public exigency.” Id., at 50. The principles articulated in Shively have been applied a number of times by this Court, and in each case we have consistently acknowledged congressional policy to dispose of sovereign lands only in the most unusual circumstances. In recognition of this policy, we do not lightly infer a congressional intent to defeat a State’s title to land under navigable waters: “[T]he United States early adopted and constantly has adhered to the policy of regarding lands under navigable waters in acquired territory, while under its sole dominion, as held for the ultimate benefit of future States, and so has refrained from making any disposal thereof, save in exceptional instances when impelled to particular disposals by some international duty or public exigency. It follows from this that disposals by the United States during the territorial period are not lightly to be inferred, and should not be regarded as intended unless the intention was definitely declared or otherwise made very plain.” United States v. Holt State Bank, 270 U. S. 49, 55 (1926). We have stated that “[a] court deciding a question of title to the bed of a navigable water must. . . begin with a strong presumption against conveyance by the United States, and 198 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. must not infer such a conveyance unless the intention was definitely declared or otherwise made very plain, or was rendered in clear and especial words, or unless the claim confirmed in terms embraces the land under the waters of the stream.” Montana n. United States, 450 U. S. 544, 552 (1981) (internal quotations omitted; citations omitted). Indeed, in only a single case—Choctaw Nation v. Oklahoma, 397 U. S. 620 (1970)—have we concluded that Congress intended to grant sovereign lands to a private party. The holding in Choctaw Nation, moreover, rested on the unusual history behind the Indian treaties at issue in that case, and indispensable to the holding was a promise to the Indian Tribe that no part of the reservation would become part of a State. Montana v. United States, supra, at 555, n. 5. Choctaw Nation was thus literally a “singular exception,” in which the result depended “on very peculiar circumstances.” 450 U. S., at 555, n. 5. B Utah Lake is a navigable body of freshwater covering 150 square miles. It is drained by the Jordan River, which flows northward and empties into the Great Salt Lake. Several years before the entry of Utah into the Union, “[t]he opening of the arid lands to homesteading raised the specter that settlers might claim lands more suitable for reservoir sites or other irrigation works, impeding future reclamation efforts.” California n. United States, 438 U. S. 645, 659 (1978). In response, Congress passed the Sundry Appropriations Act of 1888, 25 Stat. 505 (1888 Act), which authorized the United States Geological Survey to select “sites for reservoirs and other hydraulic works necessary for the storage and utilization of water for irrigation and the prevention of floods and overflows.” Id., at 526. The Act further provided that the United States would reserve the sites that might be so selected: “[A]ll the lands which may hereafter be designated or selected ... for sites for reservoirs, ditches or canals for UTAH DIV. OF STATE LANDS v. UNITED STATES 199 193 Opinion of the Court irrigation purposes and all the lands made susceptible of irrigation by such reservoirs, ditches or canals are from this time henceforth hereby reserved from sale as the property of the United States, and shall not be subject after the passage of this act, to entry, settlement or occupation until further provided by law.” Id., at 527. On April 6, 1889, Major John Wesley Powell, the Director of the United States Geological Survey, submitted a report to the Secretary of the Interior stating that the “site of Utah Lake in Utah County in the Territory of Utah is hereby selected as a reservoir site, together with all lands situate within two statute miles of the border of said lake at high water.” App. 19. The Commissioner of the General Land Office subsequently informed the Land Office at Salt Lake City of the selection of “the site of Utah Lake” as “a reservoir site” and instructed the Land Office “to refuse further entries or filing on the lands designated, in accordance with the [Sundry Appropriations] Act of October 2, 1888.” Letter of Apr. 11, 1889, App. 21. The selection of Utah Lake as a reservoir was confirmed in the official reports of the Geological Survey to Congress. Because the 1888 Act reserved all the land that “may” be designated, the 1888 Act had the practical effect of reserving all of the public lands in the West from public settlement. California v. United States, 438 U. S., at 659. Therefore, in 1890—in response to “a perfect storm of indignation from the people of the West,” ibid, (quoting 29 Cong. Rec. 1955 (1897) (statement of Cong. McRae))—Congress repealed the 1888 Act in the Sundry Appropriations Act of 1890, ch. 837, 26 Stat. 371 (1890 Act). In repealing the 1888 Act, however, Congress provided “that reservoir sites heretofore located or selected shall remain segregated and reserved from entry or settlement as provided by [the 1888 Act].” Id., at 391. Six years later, on January 4, 1896, Utah entered the Union. The Utah Enabling Act of July 16, 1894, provided that Utah 200 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. was “to be admitted into the Union on an equal footing with the original States.” 28 Stat. 107. In 1976, the Bureau of Land Management of the United States Department of the Interior issued oil and gas leases for lands underlying Utah Lake. Viewing this as a violation of its ownership and property rights to the bed of Utah Lake, the State of Utah brought suit in the District Court for the District of Utah seeking a declaratory judgment that it, rather than the United States, had title to the lakebed. Utah also sought an injunction against interference with its alleged ownership and management rights. In its complaint, Utah claimed that on January 4, 1896, by virtue of the State’s admission into the Union on an equal footing with all other States, the State of Utah became the owner of the bed of Utah Lake. The United States, in turn, answered that title to the lakebed remained in federal ownership by operation of Major Powell’s selection of the lake as a reservoir site in 1889. The District Court granted summary judgment for the United States, holding that the United States held title to the bed of Utah Lake. 624 F. Supp. 622 (1983). The District Court found that the withdrawal of the bed of Utah Lake in 1889 pursuant to the 1888 Act defeated Utah’s claim to title under the equal footing doctrine. The Court of Appeals for the Tenth Circuit affirmed. 780 F. 2d 1515 (1985). We granted certiorari, 479 U. S. 881 (1986), and now reverse. II The State of Utah contends that only a conveyance to a third party, and not merely a federal reservation of land, can defeat a State’s title to land under navigable waters upon entry into the Union. Although this Court has always spoken in terms of a “conveyance” by the United States before statehood, we have never decided whether Congress may defeat a State’s claim to title by a federal reservation or withdrawal of land under navigable waters. In Shively, this Court concluded that the only constitutional limitation on the UTAH DIV. OF STATE LANDS v. UNITED STATES 201 193 Opinion of the Court right to grant sovereign land is that such a grant must be for a “public purpos[e] appropriate to the objects for which the United States hold[s] the Territory.” 152 U. S., at 48. In the Court’s view, the power to make such a grant arose out of the Federal Government’s power over Territories under the Property Clause of the United States Constitution, which provides: “The Congress shall have Power to dispose of and make all needful Rules and Regulations respecting the Territory or other Property belonging to the United States . . . .” U. S. Const., Art. IV, §3, cl. 2. The Property Clause grants Congress plenary power to regulate and dispose of land within the Territories, and assuredly Congress also has the power to acquire land in aid of other powers conferred on it by the Constitution. Under Utah’s view, however, while the United States could create a reservoir site by granting title to Utah Lake to a private entity, the United States could not accomplish the same purpose by a means that would keep Utah Lake under federal control. We need not decide that question today, however, because even if a reservation of the bed of Utah Lake could defeat Utah’s claim, it was not accomplished on these facts. Although arguably there is nothing in the Constitution to prevent the Federal Government from defeating a State’s title to land under navigable waters by its own reservation for a particular use, the strong presumption is against finding an intent to defeat the State’s title. In Shively and Holt State Bank this Court observed that Congress “early adopted and constantly has adhered” to a policy of holding land under navigable waters “for the ultimate benefit of future States.” United States v. Holt State Bank, 270 U. S., at 55; Shively v. Bowlby, 152 U. S., at 49-50. Congress, therefore, will defeat a future State’s entitlement to land under navigable waters only “in exceptional instances,” and in light of this policy, whether faced with a reservation or a conveyance, we simply cannot infer that Congress intended to defeat a future State’s 202 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. title to land under navigable waters “unless the intention was definitely declared or otherwise made very plain.” United States v. Holt State Bank, supra, at 55. When Congress intends to convey land under navigable waters to a private party, of necessity it must also intend to defeat the future State’s claim to the land. When Congress reserves land for a particular purpose, however, it may not also intend to defeat a future State’s title to the land. The land remains in federal control, and therefore may still be held for the ultimate benefit of future States. Moreover, even if the land under navigable water passes to the State, the Federal Government may still control, develop, and use the waters for its own purposes. Arizona v. California, 373 U. S. 546, 597-598 (1963). Congress, for example, may intend to create a reservoir, but also intend to let the State obtain title to the land underneath this reservoir upon entry into statehood. Such an intent would not be unusual. In Montana v. United States, 450 U. S. 544 (1981), we found that Congress intended to permit the State to take title to the bed of a navigable river even though the river was in the midst of an Indian Reservation, and in United States v. Holt State Bank, supra, we held that Congress intended the State to hold title to the bed of a navigable lake wholly within the boundaries of an Indian Reservation. Given the longstanding policy of holding land under navigable waters for the ultimate benefit of the States, therefore, we would not infer an intent to defeat a State’s equal footing entitlement from the mere act of reservation itself. Assuming, arguendo, that a reservation of land could be effective to overcome the strong presumption against the defeat of state title, the United States would not merely be required to establish that Congress clearly intended to include land under navigable waters within the federal reservation; the United States would additionally have to establish that Congress affirmatively intended to defeat the future State’s title to such land. UTAH DIV. OF STATE LANDS v. UNITED STATES 203 193 Opinion of the Court III We conclude that the 1888 Act fails to make sufficiently plain either a congressional intent to include the bed of Utah Lake within the reservation or an intent to defeat Utah’s claim to title under the equal footing doctrine. The 1888 Act provided that the reserved lands were “reserved from sale as the property of the United States, and shall not be subject ... to entry, settlement or occupation until further provided by law.” 25 Stat. 527. The words of the 1888 Act did not necessarily refer to lands under navigable waters because lands under navigable lakes and rivers such as the bed of Utah Lake were already the property of the United States, and were already exempt from sale, entry, settlement, or occupation under the general land laws. As this Court recognized in Shively v. Bowlby, supra, at 48, “Congress has never undertaken by general laws to dispose of” land under navigable waters. See also Mann v. Tacoma Land Co., 153 U. S. 273, 284 (1894) (applying Shively v. Bowlby, supra, to hold that “the general legislation of Congress in respect to public lands does not extend to tide lands”); Illinois Central R. Co. v. Illinois, 146 U. S. 387, 437 (1892) (holding that “the same doctrine as to the dominion and sovereignty over and ownership of lands under the navigable waters . . . applies, which obtains at the common law as to the dominion and sovereignty over and ownership of lands under tide waters on the borders of the sea”). Therefore, little purpose would have been served by the reservation of the bed of Utah Lake. Moreover, the concerns with monopolization and speculation that motivated Congress to enact the 1888 Act, see P. Gates, History of Public Land Law Development 641 (1968), had nothing to do with the beds of navigable rivers and lakes. The intent to reach only land that would otherwise be available for sale and settlement is made manifest by the Act’s proviso: “Provided, That the President may at any time in his discretion by proclamation open any portion or all of the 204 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. lands reserved by this provision to settlement under the homestead laws.” 25 Stat. 527. This proviso would permit the President to open any land reserved under the 1888 Act to settlement under the homesteading laws. We find it inconceivable that Congress intended by this simple proviso to abandon its long-held and unyielding policy of never permitting the sale or settlement of land under navigable waters under the general land laws. Shively n. Bowlby, 152 U. S., at 48. The proviso can be interpreted consistently with that policy only if lands under navigable waters were not subject to reservation under the 1888 Act in the first instance. The United States, however, does not rely solely on the 1888 Act. It points to references to the bed of Utah Lake made by the Geological Survey in reserving Utah Lake, and contends that Congress ratified the Geological Survey’s reservation of the bed of Utah Lake in the 1890 Act. In the 1890 Act, Congress repealed the 1888 Act, but also specifically provided that “reservoir sites heretofore located or selected shall remain segregated and reserved from entry or settlement as provided by [the 1888] Act, until otherwise provided by law.” 26 Stat. 391. Thus, the United States argues, Congress ratified the reservation of the lakebed of Utah Lake. At first examination, statements made by the Geological Survey in reserving Utah Lake might seem to support this argument. The Tenth Annual Report of the Geological Survey (1890), which was transmitted to Congress, stated that an individual had been sent to examine Utah Lake “with reference to its capacity for a reservoir site,” in order that he might “furnish the specifications for its withdrawal as such under the law, so far as the lands covered or overflowed by it or the lands bordering upon it were still public lands.” App. 25. Furthermore, in the Eleventh Annual Report (1891), the Geological Survey reported that “the segregation” of UTAH DIV. OF STATE LANDS v. UNITED STATES 205 193 Opinion of the Court Utah Lake “was made to include not only the bed but the lowlands up to mean high water.” App. 29. The Geological Survey’s references to the “segregation” of the bed of Utah Lake, however, must be placed in the proper context. A “segregation” of land simply means that the land is no longer subject to disposal under the public land laws. See E. Baynard, Public Land Law and Procedure §5.32, p. 174 (1986). The bed of Utah Lake had already been “segregated” by the United States Geological Survey even before the adoption of the 1888 Act. The United States had surveyed Utah Lake between 1856 and 1878, and had established the “meander line”—the mean high-water elevation— segregating the land covered by navigable waters from land available for public sale and settlement.* 4 Record, Doc. F; U. S. Bureau of Land Management, Manual of Instructions for Survey of Public Lands of the United States § 3-115, p. 93 (1973) (“All navigable bodies of water and other important rivers and lakes are segregated from the public lands at *The dissent misconstrues our argument with regard to the segregation of Utah Lake between 1856 and 1878. Post, at 214, n. 5. Our point is not that the meander line was a “boundary” between the lands under the navigable waters and the adjacent lands granted by the Federal Government to private citizens, nor that this line settled the property rights of those who occupied exposed land within the meander line when Utah Lake receded. The resolution of these issues is complex, depending in large measure on the facts of the specific survey. See 4 Record, Doc. J, p. 27 (Department of Interior Memorandum discussing the effect of the exposure of land contained within the meander line to Utah Lake on land patents granted before 1888); Poynter v. Chipman, 8 Utah 442, 32 P. 690 (1893) (case involving title to land between meander line and shoreline of Utah Lake); Knudsen v. Omanson, 10 Utah 124, 37 P. 250 (1894) (same); Hinckley v. Peay, 22 Utah 21, 60 P. 1012 (1900) (same). We express no opinion on these matters. Instead, our point is a simpler one—that the meander line “segregated” the bed of Utah Lake from public sale even before the 1889 reservation, and, accordingly, that the references to the “segregation” of the lakebed by the United States Geological Survey cannot be taken as unambiguous statements of an intent to include the lakebed within the 1889 reservation. 206 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. mean high-water elevation”). Given that the bed of Utah Lake was already “segregated” from public sale, the United States Geological Survey Reports are best understood as reporting the further segregation of the lands adjacent to the lake which, until the reservation of Utah Lake in 1889, had not been segregated and thus had been available for public settlement. In the Eleventh Annual Report, for example, the Geological Survey’s announcement that “the segregation” of Utah Lake “includ[ed] not only the bed but the lowlands up to mean high water” in our view simply announced an increase in the segregated portion of Utah Lake. App. 29. Because the bed of Utah Lake had been segregated as early as 1878, the Geological Survey’s statement that the lakebed was segregated need not be taken as a statement that the bed was included within the reservation. Similarly, the Tenth Annual Report’s statement that a Geological Survey employee would furnish specifications for a withdrawal “so far as the lands covered or overflowed by [Utah Lake] or the lands bordering upon it were still public lands,” id., at 25 (emphasis supplied), is consistent with an intention that the Geological Survey would withdraw those lands still subject to public settlement, i. e., the lands that were “still public lands.” See Baynard, supra, §1.1, p. 2 (“Most enduringly, the public lands have been defined as those lands subject to sale or other disposal under the general land laws”) (emphasis in original). Because the bed of Utah Lake was not at that time “public land” subject to settlement, we think it doubtful that the Tenth Annual Report should be understood as informing Congress that the Geological Survey had reserved the bed of Utah Lake. The record reflects that the Geological Survey’s concern in 1889 was not with the bed of Utah Lake; rather its concern was that the land adjacent to the lake was then available for public sale and settlement under the general land laws. In Major Powell’s letter to the Department of the Interior announcing the selection of Utah Lake as a reservoir site he did UTAH DIV. OF STATE LANDS v. UNITED STATES 207 193 Opinion of the Court not discuss the bed of Utah Lake. Instead, he observed that “further entries of the lands adjoining Utah Lake will have a tendency to defeat the purposes of [the 1888 Act] and obstruct the use of the lake as a natural reservoir,” App. 20, and that “speedy action” was necessary to avoid settlement. Ibid. Thus, Major Powell recommended that “the Register of the Land Office at Salt Lake City be instructed to refuse entries of public land within” two miles of the lake. Ibid. The local land office was so instructed by the Department of the Interior. Id., at 21. We further find no clear demonstration that Congress intended to ratify any reservation of the bed of Utah Lake in the 1890 Act. At best, the United States points to only scattered references to the bed of Utah Lake in the material submitted to Congress, and presents no unambiguous evidence that Members of Congress actually understood these references as pointing to a reservation of the bed of Utah Lake. As with the 1888 Act, the language of the 1890 Act is consistent with the view that only land available for entry and sale was reserved: “[R]eservoir sites heretofore located or selected shall remain segregated and reserved from entry or settlement as provided by said act, until otherwise provided by law . . . .” 26 Stat. 391. In sum, the 1890 Act can be understood as ratifying a reservation of the bed of Utah Lake only by ignoring the language of the 1890 Act and by taking the Geological Survey’s references to the bed of Utah Lake out of context. Under our precedents, however, we cannot so lightly infer the reservation of land under navigable waters. We conclude, therefore, that the 1890 Act no more “‘definitely declared or otherwise made very plain’” Congress’ intention to reserve Utah Lake than had the 1888 Act. Montana v. United States, 450 U. S., at 552 (quoting United States v. Holt State Bank, 270 U. S., at 55). 208 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. IV Even if Congress did intend to reserve the bed of Utah Lake in either the 1888 Act or the 1890 Act, however, Congress did not clearly express an intention to defeat Utah’s claim to the lakebed under the equal footing doctrine upon entry into statehood. The United States points to no evidence of a congressional intent to defeat Utah’s entitlement to the bed of Utah Lake, and the structure and the history of the 1888 Act strongly suggest that Congress had no such intention. On its face, the 1888 Act does not purport to defeat the entitlement of future States to any land reserved. Instead, the Act merely provides that any reserved land is “reserved from sale” and “shall not be subject... to entry, settlement or occupation”; it makes no mention of the States’ entitlement to the beds of navigable rivers and lakes upon entry into statehood. The transfer of title of the bed of Utah Lake to Utah, moreover, would not necessarily prevent the Federal Government from subsequently developing a reservoir or water reclamation project at the lake in any event. See, e. g., Arizona v. California, 283 U. S. 423, 451-452, 457 (1931) (holding that the United States has power to construct a dam and reservoir on a navigable river and reserving question of such power for purpose of irrigating public lands). Finally, the broad sweep of the 1888 Act cannot be reconciled with an intent to defeat the States’ title to the land under navigable waters. As noted above, the 1888 Act “had the practical effect of reserving all of the public lands in the West from settlement.” California v. United States, 438 U. S., at 659. In light of the congressional policy of defeating the future States’ title to the lands under navigable waters only “in exceptional instances” in case of “international duty or public exigency,” United States v. Holt State Bank, supra, at 55, we find it inconceivable that Congress intended to defeat the future States’ title to all such land in the western United States. Such an action would be wholly at odds UTAH DIV. OF STATE LANDS v. UNITED STATES 209 193 White, J., dissenting with Congress’ policy of holding this land for the ultimate benefit of the future States. In sum, Congress did not definitely declare or otherwise make very plain either its intention to reserve the bed of Utah Lake or to defeat Utah’s title to the bed under the equal footing doctrine. Accordingly, we hold that the bed of Utah Lake passed to Utah upon that State’s entry into statehood on January 4, 1896. The judgment of the Court of Appeals is Reversed. Justice White, with whom Justice Brennan, Justice Marshall, and Justice Stevens join, dissenting. A State obtains title to the land underlying a navigable water upon its admission to the Union unless Congress’ intention to convey the land to a third party during the territorial period “was definitely declared or otherwise made very plain, or was rendered in clear and especial words, or unless the claim confirmed in terms embraces the land under the waters of the stream.” Montana n. United States, 450 U. S. 544, 552 (1981) (internal quotations omitted; citations omitted). In this case we are presented with the question whether a congressional reservation of land unto the United States during the territorial period has defeated a State’s claim to title under the equal footing doctrine. Contrary to the Court’s opinion and judgment today, I am confident that Congress has the power to prevent ownership of land underlying a navigable water from passing to a new State by reserving the land to itself for an appropriate public purpose and that Congress plainly and specifically expressed its intent to exercise that power with respect to Utah Lake in the Sundry Appropriations Act of Aug. 30, 1890, 26 Stat. 371, 390-392 (1890 Act). The Property Clause of the Constitution, Art. IV, § 3, cl. 2, is the source of the congressional power. See ante, at 200-201. In Shively v. Bowlby, 152 U. S. 1, 48 (1894), the Court stated: 210 OCTOBER TERM, 1986 White, J., dissenting 482 U. S. “We cannot doubt. . . that Congress has the power to make grants of lands below high water mark of navigable waters in any Territory of the United States, whenever it becomes necessary to do so in order to perform international obligations, or to effect the improvement of such lands for the promotion and convenience of commerce with foreign nations and among the several States, or to carry out other public purposes appropriate to the objects for which the United States hold the Territory.” (Emphasis added.) The development of reservoirs for irrigation in the arid West is surely an appropriate public purpose, and there is no reason to distinguish between a conveyance to a third party required for that purpose and a reservation unto the United States for the same purpose. Contrary to petitioner’s position, were I to make a distinction, I would more readily find a reservation constitutionally permissible than a conveyance. In the case of a reservation, the submerged lands retain their sovereign status. See ante, at 195-196. And if Congress later determines that the lands are no longer needed by the Federal Government for a public purpose, it can at that time transfer title to the State. Pursuant to the Sundry Appropriations Act of Oct. 2,1888, 25 Stat. 505, 526-527 (1888 Act), Major John Wesley Powell, famed western explorer, scientist, and Director of the United States Geological Survey (USGS), set out to identify reservoir sites.1 By letter of April 6, 1889, he reported to the ’Major Powell was quite familiar with the 1888 Act, having been for many years the leading proponent of a federal policy for reclamation of the arid West and essentially the only authority in the Federal Government on the science of irrigation. See W. Darrah, Powell of the Colorado 299-314 (1951). In 1878, he submitted to Congress his Report on the Lands of the Arid Region of the United States, with a More Detailed Account of the Lands of Utah, H. R. Exec. Doc. No. 73, 45th Cong., 2d Sess. (1878), a seminal work in the evolution of federal reclamation policy. See P. Gates, History of Public Land Law Development 645 (1968). In 1888, Major Powell reported to the Senate, at its request, 19 Cong. Rec. 2428-2429 UTAH DIV. OF STATE LANDS v. UNITED STATES 211 193 White, J., dissenting Secretary of the Interior that “the site of Utah Lake in Utah County in the Territory of Utah is hereby selected as a reservoir site, together with all lands situate within two statute miles of the border of said lake at high water.” Ante, at 199; App. 19.2 The selection of Utah Lake as a reservoir site was thereafter confirmed in the official reports of the USGS, which were formally transmitted to Congress as required by the 1888 Act.3 In the Tenth Annual Report of USGS to Secretary of the Interior 1888-1889, Part II—Irrigation, for the fiscal year ending June 30, 1889, Major Powell stated: “In April, Mr. Newell was sent to Utah to make certain examinations of Utah Lake with reference to its capacity for a reser- (1888), on the appropriation that would be required to “investigate the practicability of constructing reservoirs for the storage of water in the arid region of the United States,” the designation of sites for such reservoirs and related works, and the segregation of lands susceptible to irrigation. In the report, which was submitted to the Senate on May 11, 1888, Powell proposed language for an appropriations bill which was incorporated, with two changes not pertinent here, into the 1888 Act. See Tenth Annual Report of USGS to Secretary of the Interior 1888-1889, Part II—Irrigation, H. R. Exec. Doc. No. 1, 51st Cong., 1st Sess., pt. 5, pp. 8-14 (1890). 2 The majority makes much of the fact that Major Powell “did not dis- cuss the bed of Utah Lake” in his 1889 letter to the Secretary of the Inte- rior. Ante, at 206-207. It is true that the word “bed” is not found in the brief letter, but the land underlying the lake is clearly denoted by the words “the site of Utah Lake.” Major Powell selected as a reservoir site “the site of Utah Lake . . . together with all lands situate within two statute miles of the border of said lake at high water.” (Emphasis added.) Although it may have been the impending settlement of lands adjoining the lake which necessitated expeditious action, nothing in the letter suggested that the bed of the lake was forever unnecessary to the purpose of the reservation. 8 The 1888 Act provided that “the Director of the Geological Survey under the supervision of the Secretary of the Interior shall make a report to Congress on the first Monday in December of each year, showing in detail how the [money appropriated for the selection of sites for reservoirs] has been expended, the amount used for actual survey and engineer work in the field in locating sites for reservoires [sic] and an itemized account of the expenditures under this appropriation.” 25 Stat. 526-527. 212 OCTOBER TERM, 1986 White, J., dissenting 482 U. S. voir site and to furnish the specifications for its withdrawal as such under the law, so far as the lands covered or overflowed by it or the lands bordering upon it were still public lands.” Id., at 88; App. 25 (emphasis added). It is difficult to imagine a clearer statement to Congress of the reservation of the bed of Utah Lake.4 Major Powell, the director of the agency charged with implementing the 1888 Act, unquestionably understood the Act to authorize the reservation of lands underlying navigable waters. His contemporaneous construction of the Act is entitled to considerable deference. Udall n. Tailman, 380 U. S. 1, 16 (1965). The argument advanced by the majority in support of its position that the 1888 Act does not authorize the reservation of a lakebed, ante, at 203-204, is singularly unpersuasive as a basis for rejecting the USGS’s interpretation. Moreover, Congress clearly ratified the reservation of Utah Lake, including its bed, in the 1890 Act. Any concerns about the scope of the 1888 Act are put to rest by this ratification. Although the 1890 Act repealed the withdrawal provision of the 1888 Act, see ante, at 199, Congress provided “that reservoir sites heretofore located or selected shall remain segregated and reserved from entry or settlement as provided by [the 1888] act, until otherwise provided by law, and reservoir sites hereafter located or selected on public 4 The majority passes over the very clear, very specific reference to the bed of Utah Lake in the Tenth Annual Report and alights on the phrase “public lands.” That phrase, according to the majority, means “lands subject to sale or other disposal under the general land laws.” Ante, at 206. This interpretive approach is inconsistent with our recent opinion in Amoco Production Co. v. Gambell, 480 U. S. 531, 549, n. 15 (1987), where we “reject[ed] the assertion that the phrase ‘public lands,’ in and of itself, has a precise meaning, without reference to a definitional section or its context in a statute.” The most natural interpretation of “public lands” in this context is simply lands to which the Federal Government holds title. In Choctaw Nation v. Oklahoma, 397 U. S. 620, 633 (1970), for example, we stated that “the United States can dispose of lands underlying navigable waters just as it can dispose of other public lands.” (Emphasis added.) UTAH DIV. OF STATE LANDS v. UNITED STATES 213 193 White, J., dissenting lands shall in like manner be reserved from the date of the location or selection thereof.” 26 Stat. 391. The “broad sweep of the 1888 Act,” ante, at 208, is therefore irrelevant since that Act was repealed before Utah was admitted to the Union. The pertinent statute, the 1890 Act, is more limited in scope, reserving to the United States only reservoir sites actually selected by the USGS. Subsequent to the enactment of the 1890 Act, the Eleventh Annual Report of USGS to Secretary of the Interior 1889-1890, Part II—Irrigation, H. R. Exec. Doc. No. 1, 51st Cong., 2d Sess., pt. 5 (1890), for the fiscal year ending June 30, 1890, was transmitted to Congress. In that report, the USGS elaborated on its work at Utah Lake and described the reservation of the bed of the lake with unassailable clarity: “In Utah, in addition to the general reconnaissance of the storage facilities at the headwaters of the Sevier River and other streams, a careful survey was made of Utah Lake. This survey, run by level and transit around the lake, was for the purpose of determining the area which would be covered by damming or holding back the flood water. A description of the location and physical features of this body of water is to be found in this report under the head of Hydrography, and it will suffice to state here that after a careful study it was found that, on account of the excessive evaporation from such an enormous surface, the lake was too large to act in an economical manner as a storage reservoir. On the other hand, while it may not be advisable to hold back the water to a point above that of the average height, yet there is sufficient evidence to show that natural forces at times may raise the water level and increase the area to abnormal proportions by backing water over the great fringing marshes on the east and south. This land being, therefore, the natural flood ground of the lake, should be reserved up to the high-water line. Accordingly, the segregation, as shown on Pl. XCV and given in 214 OCTOBER TERM, 1986 White, J., dissenting 482 U. S. the following lists, was made to include not only the bed but the lowlands up to mean high water.” Id., at 183-184; App. 28-29 (emphasis added). There followed a designation of the land included in the reservation by enumeration of sections, half-sections, and quarter-sections, concluding: “Total area segregated, 125,440 acres.” Id., at 184-189; App. 29-38. This area indisputably included the bed of the lake and Congress must have so understood it.6 6 The majority’s efforts to interpret the report otherwise, ante, at 204-207, are unpersuasive. Its conclusion that the bed of the lake up to mean high water had been “segregated” as of 1878 is based on the affidavit of a Bureau of Land Management official which states that “the original surveyed meander line on Utah Lake was completed by 1878, except for three small segments approximating a total of ten miles of shoreland . . . which was completed in 1910,” 4 Record, Doc. F, and a 1973 Bureau of Land Management Manual which explains that that agency’s current survey practice is to run a meander line at the mean high-water elevation. From these documents the majority appears to deduce the location of the 1878 meander line, its relationship to the area segregated by the USGS under the 1888 Act, and its legal significance with respect to the general land laws. None of these matters would have been apparent to the 51st Congress. Among other possible complexities ignored in this analysis is the fluctuating surface area of Utah Lake. The Manual on which the majority relies explains that “mean” high water is the annual mean: “Practically all inland bodies of water pass through an annual cycle of changes, between the extremes of which will be found mean high water. . . . The most reliable indication of mean high-water elevation is the evidence made by the water’s action at its various stages, which are generally well marked in the soil. . . . “Mean high-water elevation is found at the margin of the area occupied by the water for the greater portion of each average year.” U. S. Bureau of Land Management, Manual of Instructions for Survey of Public Lands of the United States § 3-116, pp. 94-95 (1973). Mean high water, therefore, as defined in the Manual, does not account for variation from year to year. The Manual expressly states: “When by action of water the bed of the body of water changes, high-water mark changes, and the ownership of adjoining land progresses with it. Lane v. United States, 274 Fed. 290 (1921).” Id., §3-115, p. 94. The USGS reported in its Twelfth Annual Report, Part II-Irrigation, H. R. Exec. Doc. No. 1, 52d Cong., 1st Sess., pt. 5, p. 335 (1892), that the annual average UTAH DIV. OF STATE LANDS v. UNITED STATES 215 193 White, J., dissenting Several months after receiving the Eleventh Annual Report, Congress affirmed its intent to reserve the bed of Utah Lake for use as a reservoir. In the Act of Mar. 3, 1891, § 17, level of Utah Lake varied greatly through the years, “the extreme range of water level since the settlement of the country being about 12 feet.” Because the lake lies in a shallow basin, this fluctuation in water level results in substantial changes in surface area, “the shore advancing or retreating over a strip of land from 1 or 2 miles or even more in width.” Id., at 336. From 1884 to 1889, a drought period, the lake receded each year, exposing dry land to settlement. Id., at 336-337. Nothing before Congress, however, clearly documented the relationship between the surface area of the lake in 1878, when the meander line was run, and 1889, when Utah Lake was segregated pursuant to the 1888 Act. The majority’s assertion that the Eleventh Annual Report merely advised Congress of “the further segregation of the lands adjacent to the lake,” ante, at 206, is based on the assumption that the 1878 meander line lay within the area of the 1889 reservation, but even if that assumption is correct, it would not have been apparent to Congress from the information before it. The legal significance of the 1878 meander line was also less than obvious. When the lake receded between 1884 and 1889 the newly exposed lands were settled, being “of great value to the people dwelling around the shores of the lake,” since the arable and pasture lands of Utah County were fully utilized. Twelfth Annual Report, supra, at 336. This settlement was addressed at an August 19, 1889, hearing before the Senate Special Committee on Irrigation and Reclamation of Arid Lands. The Chairman of the Committee, Senator Stewart, engaged in the following exchange with the Water Master of Salt Lake City: “Mr. Wilcken. . . . [T]hey have a dam at [Utah] Lake to store water. There has been a little contention with the people in Utah County. The lake has been going down rapidly since 1884; people have crowded upon the land, and the moment we commenced to store water, thereby causing the lake to rise, there was a cry. “The Chairman. Within the last year there has been a reservation of any land needed for that purpose, and the Government will survey such land and set it apart; otherwise will there not be a disposition to crowd upon it and settle it up? “Mr. Wilcken. Of course, some of the land has been entered; but whether they have perfected their titles or not I do not know.” S. Rep. No. 928, 51st Cong., 1st Sess., pt. 3, p. 29 (1890). The Court unfortunately rejects the plain and obvious meaning of the Eleventh Annual Report for a meaning fraught with uncertainty, and I would not assume that Congress did so. The United States has had no opportu 216 OCTOBER TERM, 1986 White, J., dissenting 482 U. S. 26 Stat. 1101, Congress provided, inter alia, that reservoir sites selected or to be selected under the 1888 and 1890 Acts “shall be restricted to and shall contain only so much land as is actually necessary for the construction and maintenance of reservoirs.” Although the 1891 legislation reflected congressional concern about the extent of reservoir site reservations, Congress declined to disturb the reserved status of the bed of Utah Lake. Similarly, in the Act of Feb. 26, 1897, 29 Stat. 599, 43 U. S. C. § 664, Congress provided that all reservoir sites reserved or to be reserved by the United States were to be open for the construction of reservoirs, canals, and ditches for irrigation under rules prescribed by the Secretary of the Interior but once again declined to disturb the 1888 Act reservations themselves. The majority’s skewed interpretation of the pertinent statutes and administrative reports appears to result from the unsupportable assumption that Congress could have had no reason to reserve the bed of the lake. The USGS informed Congress as early as 1889, prior to Congress’ ratification of the reservation of Utah Lake in the 1890 Act, that when the lake was developed as a reservoir, the water level should be lowered beneath the natural shoreline in order to reduce its surface area and minimize the amount of water lost to evaporation. F. H. Newell of the USGS reported to the Senate Special Committee on the Irrigation and Reclamation of Arid Lands at an August 20, 1889, hearing on his examination of Utah Lake: “At first it was thought necessary to raise the lake in order to get more water, but on more careful study I think the lake can perform its full functions best by nity to brief the legal significance of the 1878 meander line, and, even though the majority disavows any intention of deciding property rights, ante, at 205, n., it would be most unfortunate if the majority’s unsolicited conclusion with respect to the issue is inconsistent with that of the General Land Office and spawns litigation concerning otherwise established title to the lands bordering Utah Lake. UTAH DIV. OF STATE LANDS v. UNITED STATES 217 193 White, J., dissenting drawing down below the natural shore lines, rather than by raising it above them. In other words, if raised above, the lake will be too large for the evaporation area. The evaporation is even now too great in proportion to the amount of water than can be taken out.” S. Rep. No. 928, 51st Cong., 1st Sess., pt. 3, p. 61 (1890).6 Congress could anticipate that if title to the bed of the lake passed to the State upon its admission to the Union and the United States thereafter developed a reservoir as proposed, state land would be exposed which the State presumably could develop or convey as it saw fit. This settlement would be incompatible with the Federal Government’s use of the lake as a reservoir, however, because in times of flooding, water would be impounded in the reservoir, inundating the 6 See also the Eleventh Annual Report. The Twelfth Annual Report for the fiscal year ending June 30, 1891, reiterated the USGS’s position that the water level of Utah Lake should be lowered below the natural shoreline: “[T]he lake is in effect too large to be most effective as a storage reservoir. . . . [T]he efficiency of the lake as a reservoir would be greatly increased if its area could be reduced even to less that [sic] half of its present extent; for by so doing in years of scarcity, as those of 1888 and 1889, a large proportion of the water which reaches the lake, instead of being lost by evaporation, would be retained and held for use in canals which cover the land of Salt Lake County. On the other hand, ... if the lake were only one-half its present area, the floods which come in years of exceptional precipitation would cause a far greater proportional increase of water surface than now takes place, for this water, being thrown into a smaller lake and being able to escape but slowly through the Jordan River, would of necessity encroach upon a far greater proportion of the surrounding lands. “Thus, while to obtain the maximum amount of water in years of scarcity it would be better if the lake were small, yet to take care of the floods, which will happen at intervals of from five to ten years, it is necessary that the lake have a flood area as large as it now has, or even what it would have at the highest water. From consideration of these points the segregation of the land around and under the lake was made to a contour line which should be 5 feet above the low-water mark of 1879.” Id., at 339. 218 OCTOBER TERM, 1986 White, J., dissenting 482 U. S. new settlements and potentially subjecting the Government to claims for compensation. Moreover, Congress could anticipate that if the Federal Government did not retain title to the lakebed, it might be required to pay compensation for the use of nonfederal lands on which it constructed dams, dikes, or other works. The majority relies on Arizona v. California, 373 U. S. 546, 597-598 (1963), for the proposition that “even if the land under navigable water passes to the State, the Federal Government may still control, develop, and use the waters for its own purposes.” Ante, at 202. But Arizona v. California concerned the issue of federal water rights in the Colorado River for use on Indian reservations, national forests, and recreational and wildlife areas, not the right to construct water control structures on state lands. Water rights are not at issue here. The majority also relies on an earlier opinion in Arizona n. California, 283 U. S. 423 (1931), for the proposition that “[t]he transfer of title of the bed of Utah Lake to Utah . . . would not necessarily prevent the federal government from subsequently developing a reservoir or water reclamation project at the lake in any event.” Ante, at 208. We held in that case only that Congress had the power to construct a dam and reservoir, one purpose of which was expressly declared to be “improving navigation and regulating the flow of the river” pursuant to the Federal Government’s navigational servitude. 283 U. S., at 455-456. We specifically reserved the question of the Federal Government’s power to use state land for the construction of a project with other purposes: “Since the grant of authority to build the dam and reservoir is valid as an exercise of the Constitutional power to improve navigation, we have no occasion to decide whether the authority to construct the dam and reservoir might not also have been constitutionally conferred for the specified purpose of irrigating public lands of the United States.” Id., at 457. Because the Federal Government’s right to construct irrigation works without the payment of UTAH DIV. OF STATE LANDS v. UNITED STATES 219 193 White, J., dissenting compensation is open to question, Congress may have intended to reserve the lakebed in order to avoid such claims. The majority’s refusal to acknowledge such intent because it is not absolutely certain that the reservation was necessary to effectuate Congress’ purpose is quite strange. In sum, the reservation by the USGS of Utah Lake by its plain “terms embraces the land under the waters of the [lake],” and Congress “definitely declared” its intent to ratify that reservation in the 1890 Act. See Montana v. United States, 450 U. S., at 552. As I see it, Utah did not obtain title to the bed of the lake upon its admission to the Union, and I therefore dissent. 220 OCTOBER TERM, 1986 Syllabus 482 U. S. SHEARSON/AMERICAN EXPRESS INC. ET AL. v. McMAHON et al. CERTIORARI TO THE UNITED STATES COURT OF APPEALS FOR THE SECOND CIRCUIT No. 86-44. Argued March 3, 1987—Decided June 8, 1987 Respondents were customers of petitioner Shearson/American Express Inc. (Shearson), a brokerage firm registered with the Securities and Exchange Commission (SEC), under customer agreements providing for arbitration of any controversy relating to their accounts. Respondents filed suit in Federal District Court against Shearson and its representative (also a petitioner here) who handled their accounts, alleging violations of the antifraud provisions in § 10(b) of the Securities Exchange Act of 1934 (Exchange Act) and SEC Rule 10b-5, and of the Racketeer Influenced and Corrupt Organizations Act (RICO). Petitioners moved to compel arbitration of the claims pursuant to § 3 of the Federal Arbitration Act, which requires a court to stay its proceedings if it is satisfied that an issue before it is arbitrable under an arbitration agreement. The District Court held that respondents’ Exchange Act claims were arbitrable, but that their RICO claim was not. The Court of Appeals affirmed as to the RICO claim, but reversed as to the Exchange Act claims. Held: 1. The Arbitration Act establishes a federal policy favoring arbitration, requiring that the courts rigorously enforce arbitration agreements. This duty is not diminished when a party bound by an agreement raises a claim founded on statutory rights. The Act’s mandate may be overridden by a contrary congressional command, but the burden is on the party opposing arbitration to show that Congress intended to preclude a waiver of judicial remedies for the statutory rights at issue. Such intent may be discernible from the statute’s text, history, or purposes. Pp. 225-227. 2. Respondents’ Exchange Act claims are arbitrable under the provisions of the Arbitration Act. Congressional intent to require a judicial forum for the resolution of § 10(b) claims cannot be deduced from § 29(a) of the Exchange Act, which declares void an agreement to waive “compliance with any provision of [the Act].” Section 29(a) only prohibits waiver of the Act’s substantive obligations and thus does not void waiver of § 27 of the Act, which confers exclusive district court jurisdiction of violations of the Act, but which does not impose any statutory duties. SHEARSON/AMERICAN EXPRESS INC. v. McMAHON 221 220 Syllabus Wilko v. Swan, 346 U. S. 427, which held that claims arising under the Securities Act of 1933, which has similar antiwaiver and jurisdictional provisions, were not subject to compulsory arbitration under an arbitration agreement, does not control here. That case must be read as barring waiver of a judicial forum only where arbitration is inadequate to protect the substantive rights at issue. Cf. Scherk n. Alberto-Culver Co., 417 U. S. 506. There is no merit to respondents’ contention, based on Wilko, that their arbitration agreements effected an impermissible waiver of the Exchange Act’s substantive protections. Even if Wilko’s assumptions regarding arbitration were valid at the time it was decided—when there was judicial mistrust of the arbitral process—such assumptions do not hold true today for arbitration procedures (such as those involved here) subject to the SEC’s oversight authority under the intervening changes in the regulatory structure of the securities laws. Nor does the legislative history support respondents’ argument that even if § 29(a) as enacted does not void predispute arbitration agreements, Congress subsequently has indicated that § 29(a) should be so interpreted. Pp. 227-238. 3. Respondents’ RICO claim is also arbitrable under the Arbitration Act. Nothing in RICO’s text or legislative history even arguably evinces congressional intent to exclude civil RICO claims for treble damages under 18 U. S. C. § 1964(c) from the Arbitration Act’s dictates. Nor is there any irreconcilable conflict between arbitration and RICO’s underlying purposes. Cf. Mitsubishi Motors Corp. v. Soler Chrysler-Plymouth, Inc., 473 U. S. 614. Neither the potential complexity of RICO claims, nor the “overlap” between RICO’s civil and criminal provisions, renders § 1964(c) claims nonarbitrable. Moreover, the public interest in the enforcement of RICO does not preclude submission of such claims to arbitration. The legislative history of § 1964(c) emphasized the remedial role of the treble-damages provision. Its policing function, although important, was a secondary concern. The private attorney general role for the typical RICO plaintiff does not support a finding that there is an irreconcilable conflict between arbitration and enforcement of RICO. Pp. 238-242. 788 F. 2d 94, reversed and remanded. O’Connor, J., delivered the opinion of the Court, in which Rehnquist, C. J., and White, Powell, and Scalia, JJ., joined, and in Parts I, II, and IV of which Brennan, Marshall, Blackmun, and Stevens, JJ., joined. Blackmun, J., filed an opinion concurring in part and dissenting in part, in which Brennan and Marshall, JJ., joined, post, p. 242. Stevens, J., filed an opinion concurring in part and dissenting in part, post, p. 268. 222 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. Theodore A. Krebsbach argued the cause and filed briefs for petitioners. Richard G. Taranto argued the cause for the United States as amicus curiae urging reversal. With him on the brief were Solicitor General Fried, Deputy Solicitor General Cohen, Daniel L. Goelzer, Paul Gonson, Jacob H. Stillman, and David A. Sirignano. Theodore G. Eppenstein argued the cause for respondents. With him on the brief was Madelaine Eppenstein. * Justice O’Connor delivered the opinion of the Court. This case presents two questions regarding the enforceability of predispute arbitration agreements between brokerage firms and their customers. The first is whether a claim brought under § 10(b) of the Securities Exchange Act of 1934 (Exchange Act), 48 Stat. 891, 15 U. S. C. § 78j(b), must be sent to arbitration in accordance with the terms of an arbitration agreement. The second is whether a claim brought under the Racketeer Influenced and Corrupt Organizations Act (RICO), 18 U. S. C. § 1961 et seq., must be arbitrated in accordance with the terms of such an agreement. I Between 1980 and 1982, respondents Eugene and Julia McMahon, individually and as trustees for various pension and profit-sharing plans, were customers of petitioner Shear- *Briefs of amici curiae urging reversal were filed for the American Arbitration Association by Michael F. Hoellering, Joseph T. McLaughlin, Rosemary S. Page, Thomas Thacher, Gerald Aksen, Sheldon L. Berens, Richard S. Lombard, Robert MacCrate, John R. Stevenson, and Robert B. von Mehren; and for the Attorneys for Securities Industry Association, Inc., et al. by Joseph G. Riemer III, Judith Welcom, Paul Windels III, William J. Fitzpatrick, Donald B. McNelley, Steven N. Machtinger, Paul J. Dubow, and Joseph McLaughlin. Briefs of amici curiae urging affirmance were filed for Willie D. Chandler et al. by Stirling Lathrop and Richard D. Greenfield; and for Bruce Cordray et al. by Denis A. Downey. SHEARSON/AMERICAN EXPRESS INC. v. McMAHON 223 220 Opinion of the Court son/American Express Inc. (Shearson), a brokerage firm registered with the Securities and Exchange Commission (SEC or Commission). Two customer agreements signed by Julia McMahon provided for arbitration of any controversy relating to the accounts the McMahons maintained with Shearson. The arbitration provision provided in relevant part as follows: “Unless unenforceable due to federal or state law, any controversy arising out of or relating to my accounts, to transactions with you for me or to this agreement or the breach thereof, shall be settled by arbitration in accordance with the rules, then in effect, of the National Association of Securities Dealers, Inc. or the Boards of Directors of the New York Stock Exchange, Inc. and/or the American Stock Exchange, Inc. as I may elect.” 618 F. Supp. 384, 385 (1985). In October 1984, the McMahons filed an amended complaint against Shearson and petitioner Mary Ann McNulty, the registered representative who handled their accounts, in the United States District Court for the Southern District of New York. The complaint alleged that McNulty, with Shearson’s knowledge, had violated § 10(b) of the Exchange Act and Rule 10b-5, 17 CFR § 240.10b-5 (1986), by engaging in fraudulent, excessive trading on respondents’ accounts and by making false statements and omitting material facts from the advice given to respondents. The complaint also alleged a RICO claim, 18 U. S. C. § 1962(c), and state law claims for fraud and breach of fiduciary duties. Relying on the customer agreements, petitioners moved to compel arbitration of the McMahons’ claims pursuant to § 3 of the Federal Arbitration Act, 9 U. S. C. §3. The District Court granted the motion in part. 618 F. Supp. 384 (1985). The court first rejected the McMahons’ contention that the arbitration agreements were unenforceable as contracts of 224 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. adhesion. It then found that the McMahons’ § 10(b) claims were arbitrable under the terms of the agreement, concluding that such a result followed from this Court’s decision in Dean Witter Reynolds Inc. v. Byrd, 470 U. S. 213 (1985), and the “strong national policy favoring the enforcement of arbitration agreements.” 618 F. Supp., at 388. The District Court also held that the McMahons’ state law claims were arbitrable under Dean Witter Reynolds Inc. v. Byrd, supra. It concluded, however, that the McMahons’ RICO claim was not arbitrable “because of the important federal policies inherent in the enforcement of RICO by the federal courts.” 618 F. Supp., at 387. The Court of Appeals affirmed the District Court on the state law and RICO claims, but it reversed on the Exchange Act claims. 788 F. 2d 94 (1986). With respect to the RICO claim, the Court of Appeals concluded that “public policy” considerations made it “inappropriat[e]” to apply the provisions of the Arbitration Act to RICO suits. Id., at 98. The court reasoned that RICO claims are “not merely a private matter.” Ibid. Because a RICO plaintiff may be likened to a “private attorney general” protecting the public interest, ibid., the Court of Appeals concluded that such claims should be adjudicated only in a judicial forum. It distinguished this Court’s reasoning in Mitsubishi Motors Corp. v. Soler Chrysler-Plymouth, Inc., 473 U. S. 614 (1985), concerning the arbitrability of antitrust claims, on the ground that it involved international business transactions and did not affect the law “as applied to agreements to arbitrate arising from domestic transactions.” 788 F. 2d, at 98. With respect to respondents’ Exchange Act claims, the Court of Appeals noted that under Wilko n. Swan, 346 U. S. 427 (1953), claims arising under § 12(2) of the Securities Act of 1933 (Securities Act), 48 Stat. 84, 15 U. S. C. §77Z(2), are not subject to compulsory arbitration. The Court of Appeals SHEARSON/AMERICAN EXPRESS INC. v. McMAHON 225 220 Opinion of the Court observed that it previously had extended the Wilko rule to claims arising under § 10(b) of the Exchange Act and Rule 10b-5. See, e. g., Allegaert v. Perot, 548 F. 2d 432 (CA2), cert, denied, 432 U. S. 910 (1977); Greater Continental Corp. v. Schechter, 422 F. 2d 1100 (CA2 1970). The court acknowledged that Scherk v. Alberto-Culver Co., 417 U. S. 506 (1974), and Dean Witter Reynolds Inc. v. Byrd, supra, had “cast some doubt on the applicability of Wilko to claims under § 10(b).” 788 F. 2d, at 97. The Court of Appeals nevertheless concluded that it was bound by the “clear judicial precedent in this Circuit,” and held that Wilko must be applied to Exchange Act claims. 788 F. 2d, at 98. We granted certiorari, 479 U. S. 812 (1986), to resolve the conflict among the Courts of Appeals regarding the arbitrability of § 10(b)1 and RICO2 claims. II The Federal Arbitration Act, 9 U. S. C. §1 et seq., provides the starting point for answering the questions raised in this case. The Act was intended to “revers[e] centuries of judicial hostility to arbitration agreements,” Scherk v. Alberto-Culver Co., supra, at 510, by “plac[ing] arbitration Compare 788 F. 2d 94 (CA2 1986) (case below); Jacobson v. Merrill Lynch, Pierce, Fenner & Smith, Inc., 797 F. 2d 1197 (CA3 1986), cert, pending, No. 86-487; King n. Drexel Burnham Lambert, Inc., 796 F. 2d 59 (CA5 1986), cert, pending, No. 86-282; Sterne v. Dean Witter Reynolds, Inc., 808 F. 2d 480 (CA6 1987); Conover v. Dean Witter Reynolds, Inc., 794 F. 2d 520 (CA9 1986), cert, pending, No. 86-321; and Wolfe v. E. F. Hutton & Co., 800 F. 2d 1032 (CA11 1986); with Page n. Moseley, Hall-garten, Estabrook & Weeden, Inc., 806 F. 2d 291 (CAI 1986); Phillips v. Merrill Lynch, Pierce, Fenner & Smith, Inc., 795 F. 2d 1393 (CA8 1986), cert, pending, No. 86-578. 2 Compare Page v. Moseley, Hallgarten, Estabrook & Weeden, supra; and 788 F. 2d 94 (CA2 1986) (case below), with Mayaja, Inc. v. Bodkin, 803 F. 2d 157 (CA5 1986). See also Jacobson v. Merrill Lynch, Pierce, Fenner & Smith, Inc., supra; Tashea v. Bache, Halsey, Stuart, Shields, Inc., 802 F. 2d 1337 (CA11 1986). 226 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. agreements ‘upon the same footing as other contracts.’” 417 U. S., at 511, quoting H. R. Rep. No. 96, 68th Cong., 1st Sess., 1, 2 (1924). The Arbitration Act accomplishes this purpose by providing that arbitration agreements “shall be valid, irrevocable, and enforceable, save upon such grounds as exist at law or in equity for the revocation of any contract.” 9 U. S. C. §2. The Act also provides that a court must stay its proceedings if it is satisfied that an issue before it is arbitrable under the agreement, § 3; and it authorizes a federal district court to issue an order compelling arbitration if there has been a “failure, neglect, or refusal” to comply with the arbitration agreement, § 4. The Arbitration Act thus establishes a “federal policy favoring arbitration,” Moses H. Cone Memorial Hospital n. Mercury Construction Corp., 460 U. S. 1, 24 (1983), requiring that “we rigorously enforce agreements to arbitrate.” Dean Witter Reynolds Inc. n. Byrd, supra, at 221. This duty to enforce arbitration agreements is not diminished when a party bound by an agreement raises a claim founded on statutory rights. As we observed in Mitsubishi Motors Corp. v. Soler Chrysler-Plymouth, Inc., “we are well past the time when judicial suspicion of the desirability of arbitration and of the competence of arbitral tribunals” should inhibit enforcement of the Act “‘in controversies based on statutes.’” 473 U. S., at 626-627, quoting Wilko v. Swan, supra, at 432. Absent a well-founded claim that an arbitration agreement resulted from the sort of fraud or excessive economic power that “would provide grounds ‘for the revocation of any contract,’” 473 U. S., at 627, the Arbitration Act “provides no basis for disfavoring agreements to arbitrate statutory claims by skewing the otherwise hospitable inquiry into arbitrability.” Ibid. The Arbitration Act, standing alone, therefore mandates enforcement of agreements to arbitrate statutory claims. Like any statutory directive, the Arbitration Act’s mandate may be overridden by a contrary congressional command. SHEARSON/AMERICAN EXPRESS INC. v. McMAHON 227 220 Opinion of the Court The burden is on the party opposing arbitration, however, to show that Congress intended to preclude a waiver of judicial remedies for the statutory rights at issue. See id., at 628. If Congress did intend to limit or prohibit waiver of a judicial forum for a particular claim, such an intent “will be deducible from [the statute’s] text or legislative history,” ibid., or from an inherent conflict between arbitration and the statute’s underlying purposes. See id., at 632-637; Dean Witter Reynolds Inc. v. Byrd, 470 U. S., at 217. To defeat application of the Arbitration Act in this case, therefore, the McMahons must demonstrate that Congress intended to make an exception to the Arbitration Act for claims arising under RICO and the Exchange Act, an intention discernible from the text, history, or purposes of the statute. We examine the McMahons’ arguments regarding the Exchange Act and RICO in turn. Ill When Congress enacted the Exchange Act in 1934, it did not specifically address the question of the arbitrability of § 10(b) claims. The McMahons contend, however, that congressional intent to require a judicial forum for the resolution of § 10(b) claims can be deduced from § 29(a) of the Exchange Act, 15 U. S. C. § 78cc(a), which declares void “[a]ny condition, stipulation, or provision binding any person to waive compliance with any provision of [the Act].” First, we reject the McMahons’ argument that § 29(a) forbids waiver of § 27 of the Exchange Act, 15 U. S. C. § 78aa. Section 27 provides in relevant part: “The district courts of the United States . . . shall have exclusive jurisdiction of violations of this title or the rules and regulations thereunder, and of all suits in equity and actions at law brought to enforce any liability or duty created by this title or the rules and regulations thereunder.” 228 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. The McMahons contend that an agreement to waive this jurisdictional provision is unenforceable because § 29(a) voids the waiver of “any provision” of the Exchange Act. The language of § 29(a), however, does not reach so far. What the antiwaiver provision of § 29(a) forbids is enforcement of agreements to waive “compliance” with the provisions of the statute. But § 27 itself does not impose any duty with which persons trading in securities must “comply.” By its terms, § 29(a) only prohibits waiver of the substantive obligations imposed by the Exchange Act. Because §27 does not impose any statutory duties, its waiver does not constitute a waiver of “compliance with any provision” of the Exchange Act under § 29(a). We do not read Wilko v. Swan, 346 U. S. 427 (1953), as compelling a different result. In Wilko, the Court held that a predispute agreement could not be enforced to compel arbitration of a claim arising under § 12(2) of the Securities Act, 15 U. S. C. § 77Z(2). The basis for the ruling was § 14 of the Securities Act, which, like § 29(a) of the Exchange Act, declares void any stipulation “to waive compliance with any provision” of the statute. At the beginning of its analysis, the Wilko Court stated that the Securities Act’s jurisdictional provision was “the kind of ‘provision’ that cannot be waived under § 14 of the Securities Act.” 346 U. S., at 435. This statement, however, can only be understood in the context of the Court’s ensuing discussion explaining why arbitration was inadequate as a means of enforcing “the provisions of the Securities Act, advantageous to the buyer.” Ibid. The conclusion in Wilko was expressly based on the Court’s belief that a judicial forum was needed to protect the substantive rights created by the Securities Act: “As the protective provisions of the Securities Act require the exercise of judicial direction to fairly assure their effectiveness, it seems to us that Congress must have intended § 14 . . .to apply to waiver of judicial trial and review.” Id., at 437. Wilko must be understood, therefore, as holding that the plaintiff’s waiver SHEARSON/AMERICAN EXPRESS INC. v. McMAHON 229 220 Opinion of the Court of the “right to select the judicial forum,” id., at 435, was unenforceable only because arbitration was judged inadequate to enforce the statutory rights created by § 12(2). Indeed, any different reading of Wilko would be inconsistent with this Court’s decision in Scherk v. Alberto-Culver Co., 417 U. S. 506 (1974). In Scherk, the Court upheld enforcement of a predispute agreement to arbitrate Exchange Act claims by parties to an international contract. The Scherk Court assumed for purposes of its opinion that Wilko applied to the Exchange Act, but it determined that an international contract “involve[d] considerations and policies significantly different from those found controlling in Wilko.” 417 U. S., at 515. The Court reasoned that arbitration reduced the uncertainty of international contracts and obviated the danger that a dispute might be submitted to a hostile or unfamiliar forum. At the same time, the Court noted that the advantages of judicial resolution were diminished by the possibility that the opposing party would make “speedy resort to a foreign court.” Id., at 518. The decision in Scherk thus turned on the Court’s judgment that under the circumstances of that case, arbitration was an adequate substitute for adjudication as a means of enforcing the parties’ statutory rights. Scherk supports our understanding that Wilko must be read as barring waiver of a judicial forum only where arbitration is inadequate to protect the substantive rights at issue. At the same time, it confirms that where arbitration does provide an adequate means of enforcing the provisions of the Exchange Act, § 29(a) does not void a predispute waiver of §27—Scherk upheld enforcement of just such a waiver. The second argument offered by the McMahons is that the arbitration agreement effects an impermissible waiver of the substantive protections of the Exchange Act. Ordinarily, “[b]y agreeing to arbitrate a statutory claim, a party does not forgo the substantive rights afforded by the statute; it only submits to their resolution in an arbitral, rather 230 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. than a judicial, forum.” Mitsubishi Motors Corp. v. Soler-Chrysler-Plymouth, Inc., 473 U. S., at 628. The McMahons argue, however, that § 29(a) compels a different conclusion. Initially, they contend that predispute agreements are void under § 29(a) because they tend to result from broker overreaching. They reason, as do some commentators, that Wilko is premised on the belief “that arbitration clauses in securities sales agreements generally are not freely negotiated.” See, e. g., Sterk, Enforceability of Agreements to Arbitrate: An Examination of the Public Policy Defense, 2 Cardozo L. Rev. 481, 519 (1981). According to this view, Wilko barred enforcement of predispute agreements because of this frequent inequality of bargaining power, reasoning that Congress intended for § 14 generally to ensure that sellers did not “maneuver buyers into a position that might weaken their ability to recover under the Securities Act.” 346 U. S., at 432. The McMahons urge that we should interpret § 29(a) in the same fashion. We decline to give Wilko a. reading so far at odds with the plain language of § 14, or to adopt such an unlikely interpretation of § 29(a). The concern that § 29(a) is directed against is evident from the statute’s plain language: it is a concern with whether an agreement “waive[s] compliance with [a] provision” of the Exchange Act. The voluntariness of the agreement is irrelevant to this inquiry: if a stipulation waives compliance with a statutory duty, it is void under § 29(a), whether voluntary or not. Thus, a customer cannot negotiate a reduction in commissions in exchange for a waiver of compliance with the requirements of the Exchange Act, even if the customer knowingly and voluntarily agreed to the bargain. Section 29(a) is concerned, not with whether brokers “maneuver[ed customers] into” an agreement, but with whether the agreement “weaken[s] their ability to recover under the [Exchange] Act.” 346 U. S., at 432. The former is grounds for revoking the contract under ordinary SHEARSON/AMERICAN EXPRESS INC. v. McMAHON 231 220 Opinion of the Court principles of contract law; the latter is grounds for voiding the agreement under § 29(a). The other reason advanced by the McMahons for finding a waiver of their § 10(b) rights is that arbitration does “weaken their ability to recover under the [Exchange] Act.” Ibid. That is the heart of the Court’s decision in Wilko, and respondents urge that we should follow its reasoning. Wilko listed several grounds why, in the Court’s view, the “effectiveness [of the Act’s provisions] in application is lessened in arbitration.” 346 U. S., at 435. First, the Wilko Court believed that arbitration proceedings were not suited to cases requiring “subjective findings on the purpose and knowledge of an alleged violator.” Id., at 435-436. Wilko also was concerned that arbitrators must make legal determinations “without judicial instruction on the law,” and that an arbitration award “may be made without explanation of [the arbitrator’s] reasons and without a complete record of their proceedings.” Id., at 436. Finally, Wilko noted that the “[p]ower to vacate an award is limited,” and that “interpretations of the law by the arbitrators in contrast to manifest disregard are not subject, in the federal courts, to judicial review for error in interpretation.” Id., at 436-437. Wilko concluded that in view of these drawbacks to arbitration, §12(2) claims “require[d] the exercise of judicial direction to fairly assure their effectiveness.” Id., at 437. As Justice Frankfurter noted in his dissent in Wilko, the Court’s opinion did not rest on any evidence, either “in the record . . . [or] in the facts of which [it could] take judicial notice,” that “the arbitral system . . . would not afford the plaintiff the rights to which he is entitled.” Id., at 439. Instead, the reasons given in Wilko reflect a general suspicion of the desirability of arbitration and the competence of arbitral tribunals—most apply with no greater force to the arbitration of securities disputes than to the arbitration of legal disputes generally. It is difficult to reconcile Wilko's mistrust of the arbitral process with this Court’s subsequent 232 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. decisions involving the Arbitration Act. See, e. g., Mitsubishi Motors Corp. v. Soler Chrysler-Plymouth, Inc., supra; Dean Witter Reynolds Inc. v. Byrd, 470 U. S. 213 (1985); Southland Corp. n. Keating, 465 U. S. 1 (1984); Moses H. Cone Memorial Hospital v. Mercury Construction Corp., 460 U. S. 1 (1983); Scherk v. Alberto-Culver Co., 417 U. S. 506 (1974). Indeed, most of the reasons given in Wilko have been rejected subsequently by the Court as a basis for holding claims to be nonarbitrable. In Mitsubishi, for example, we recognized that arbitral tribunals are readily capable of handling the factual and legal complexities of antitrust claims, notwithstanding the absence of judicial instruction and supervision. See 473 U. S., at 633-634. Likewise, we have concluded that the streamlined procedures of arbitration do not entail any consequential restriction on substantive rights. Id., at 628. Finally, we have indicated that there is no reason to assume at the outset that arbitrators will not follow the law; although judicial scrutiny of arbitration awards necessarily is limited, such review is sufficient to ensure that arbitrators comply with the requirements of the statute. See id., at 636-637, and n. 19 (declining to assume that arbitration will not be resolved in accordance with statutory law, but reserving consideration of “effect of an arbitral tribunal’s failure to take cognizance of the statutory cause of action on the claimant’s capacity to reinstate suit in federal court”). The suitability of arbitration as a means of enforcing Exchange Act rights is evident from our decision in Scherk. Although the holding in that case was limited to international agreements, the competence of arbitral tribunals to resolve § 10(b) claims is the same in both settings. Courts likewise have routinely enforced agreements to arbitrate § 10(b) claims where both parties are members of a securities exchange or the National Association of Securities Dealers (NASD), suggesting that arbitral tribunals are fully capable of handling such matters. See, e. g., Axelrod & Co. v. Kordich, Victor SHEARSON/AMERICAN EXPRESS INC. v. McMAHON 233 220 Opinion of the Court & Neufeld, 320 F. Supp. 193 (SDNY 1970), aff’d, 451 F. 2d 838 (CA2 1971); Brown v. Gilligan, Will & Co., 287 F. Supp. 766 (SDNY 1968). And courts uniformly have concluded that Wilko does not apply to the submission to arbitration of existing disputes, see, e. g., Gardner v. Shearson, Hammill & Co., 433 F. 2d 367 (CA5 1970); Moran v. Paine, Webber, Jackson & Curtis, 389 F. 2d 242 (CA3 1968), even though the inherent suitability of arbitration as a means of resolving § 10(b) claims remains unchanged. Cf. Mitsubishi, 473 U. S., at 633. Thus, the mistrust of arbitration that formed the basis for the Wilko opinion in 1953 is difficult to square with the assessment of arbitration that has prevailed since that time. This is especially so in light of the intervening changes in the regulatory structure of the securities laws. Even if Wilko’s assumptions regarding arbitration were valid at the time Wilko was decided, most certainly they do not hold true today for arbitration procedures subject to the SEC’s oversight authority. In 1953, when Wilko was decided, the Commission had only limited authority over the rules governing self-regulatory organizations (SROs)—the national securities exchanges and registered securities associations—and this authority appears not to have included any authority at all over their arbitration rules. See Brief for Securities and Exchange Commission as Amicus Curiae 14-15. Since the 1975 amendments to § 19 of the Exchange Act, however, the Commission has had expansive power to ensure the adequacy of the arbitration procedures employed by the SROs. No proposed rule change may take effect unless the SEC finds that the proposed rule is consistent with the requirements of the Exchange Act, 15 U. S. C. § 78s(b)(2); and the Commission has the power, on its own initiative, to “abrogate, add to, and delete from” any SRO rule if it finds such changes necessary or appropriate to further the objectives of the Act, 15 U. S. C. §78s(c). In short, the Commission has broad authority to oversee and to 234 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. regulate the rules adopted by the SROs relating to customer disputes, including the power to mandate the adoption of any rules it deems necessary to ensure that arbitration procedures adequately protect statutory rights.3 In the exercise of its regulatory authority, the SEC has specifically approved the arbitration procedures of the New York Stock Exchange, the American Stock Exchange, and the NASD, the organizations mentioned in the arbitration agreement at issue in this case. We conclude that where, as in this case, the prescribed procedures are subject to the Commission’s § 19 authority, an arbitration agreement does not effect a waiver of the protections of the Act. While stare decisis concerns may counsel against upsetting Wilko’s contrary conclusion under the Securities Act, we refuse to extend Wilko’s reasoning to the Exchange Act in light of these intervening regulatory developments. The McMahons’ agreement to submit to arbitration therefore is not tantamount to an impermissible waiver of the McMahons’ rights under § 10(b), and the agreement is not void on that basis under § 29(a). The final argument offered by the McMahons is that even if § 29(a) as enacted does not void predispute arbitration agreements, Congress subsequently has indicated that it desires § 29(a) to be so interpreted. According to the McMahons, Congress expressed this intent when it failed to make more 3 The McMahons contend that Securities Exchange Act Rei. No. 15984 (1979), [1979 Transfer Binder] CCH Fed. Sec. L. Rep. 182,122, and SEC Rule 15c2-2, 17 CFR § 240.15c2-2 (1986), provide authority for the view that § 29(a) bars enforcement of predispute arbitration agreements. We agree with the Commission, however, that its actions were not based on any independent analysis of § 29(a), but instead “were premised on the Commission’s assumption, based on court of appeals decisions following Wilko, . . . that agreements to arbitrate Rule 10b-5 claims were not, in fact, enforceable.” Brief for Securities and Exchange Commission as Amicus Curiae 18, n. 13 (citation omitted). The SEC’s actions therefore do not cast any additional light on the question of the arbitrability of Exchange Act claims. SHEARSON/AMERICAN EXPRESS INC. v. MCMAHON 235 220 Opinion of the Court extensive changes to § 28(b), 15 U. S. C. §78bb(b), in the 1975 amendments to the Exchange Act. Before its amendment, § 28(b) provided in relevant part: “Nothing in this chapter shall be construed to modify existing law (1) with regard to the binding effect on any member of any exchange of any action taken by the authorities of such exchange to settle disputes between its members, or (2) with regard to the binding effect of such action on any person who has agreed to be bound thereby, or (3) with regard to the binding effect on any such member of any disciplinary action taken by the authorities of the exchange.” 48 Stat. 903. The chief aim of this provision was to preserve the self-regulatory role of the securities exchanges, by giving the exchanges a means of enforcing their rules against their members. See, e. g., Tullis v. Kohlmeyer & Co., 551 F. 2d 632, 638 (CA5 1977) (“[P]reserv[ing] for the stock exchanges a major self-regulatory role ... is the basis of § 28(b)”); Axelrod & Co. v. Kordich, Victor & Neufeld, 451 F. 2d, at 840-841. In 1975, Congress made extensive revisions to the Exchange Act intended to “clarify the scope of the self-regulatory responsibilities of national securities exchanges and registered securities associations . . . and the manner in which they are to exercise those responsibilities.” S. Rep. No. 94-75, p. 22 (1975). In making these changes, the Senate Report observed: “The self-regulatory organizations must exercise governmental-type powers if they are to carry out their responsibilities under the Exchange Act. When a member violates the Act or a self-regulatory organization’s rules, the organization must be in a position to impose appropriate penalties or to revoke relevant privileges. ” Id., at 24. The amendments to § 28 reflect this objective. Paragraph (3) of § 28(b) was deleted and replaced with new § 28(c), which provided that the validity of any disciplinary action taken by an SRO would not be affected by a subsequent decision by the SEC to stay or modify the sanction. See 15 U. S. C. 236 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. §78bb(c). At the same time, § 28(b) was expanded to ensure that all SROs as well as the Municipal Securities Rulemaking Board had the power to enforce their substantive rules against their members. Section 28(b), as amended, provides: “Nothing in this chapter shall be construed to modify existing law with regard to the binding effect (1) on any member of or participant in any self-regulatory organization of any action taken by the authorities of such organization to settle disputes between its members or participants, (2) on any municipal securities dealer or municipal securities broker of any action taken pursuant to a procedure established by the Municipal Securities Rulemaking Board to settle disputes between municipal securities dealers and municipal securities brokers, or (3) of any action described in paragraph (1) or (2) on any person who has agreed to be bound thereby.” Thus, the amended version of § 28(b), like the original, mentions neither customers nor arbitration. It is directed at an entirely different problem: enhancing the self-regulatory function of the SROs under the Exchange Act. The McMahons nonetheless argue that we should find it significant that Congress did not take this opportunity to address the general question of the arbitrability of Exchange Act claims. Their argument is based entirely on a sentence from the Conference Report, which they contend amounts to a ratification of Wilko’s extension to Exchange Act claims. The Conference Report states: “The Senate bill amended section 28 of the Securities Exchange Act of 1934 with respect to arbitration proceedings between self-regulatory organizations and their participants, members, or persons dealing with members or participants. The House amendment contained no comparable provision. The House receded to the Senate. It was the clear understanding of the conferees that SHEARSON/AMERICAN EXPRESS INC. v. McMAHON 237 220 Opinion of the Court this amendment did not change existing law, as articulated in Wilko v. Swan, 346 U. S. 427 (1953), concerning the effect of arbitration proceedings provisions in agreements entered into by persons dealing with members and participants of self-regulatory organizations.” H. R. Conf. Rep. No. 94-229, p. Ill (1975). The McMahons contend that the conferees would not have acknowledged Wilko in a revision of the Exchange Act unless they were aware of lower court decisions extending Wilko to § 10(b) claims and intended to approve them. We find this argument fraught with difficulties. We cannot see how Congress could extend Wilko to the Exchange Act without enacting into law any provision remotely addressing that subject. See Train v. City of New York, 420 U. S. 35, 45 (1975). And even if it could, there is little reason to interpret the Report as the McMahons suggest. At the outset, the committee may well have mentioned Wilko for a reason entirely different from the one postulated by the McMahons—lower courts had applied § 28(b) to the Securities Act, see, e. g., Axelrod & Co. v. Kordich, Victor & Neufeld, supra, at 843, and the committee may simply have wished to make clear that the amendment to § 28(b) was not otherwise intended to affect Wilko’s construction of the Securities Act. Moreover, even if the committee were referring to the arbitrability of § 10(b) claims, the quoted sentence does not disclose what committee members thought “existing law” provided. The conference members might have had in mind the two Court of Appeals decisions extending Wilko to the Exchange Act, as the McMahons contend. See Greater Continental Corp. v. Schechter, 422 F. 2d 1100 (CA2 1970); Moran v. Paine, Webber, Jackson & Curtis, 389 F. 2d 242 (CA3 1968). It is equally likely, however, that the committee had in mind this Court’s decision the year before expressing doubts as to whether Wilko should be extended to § 10 (b) claims. See Scherk v. Alberto-Culver Co., 417 U. S., at 513 (“[A] colorable argument co^ld be made that even the 238 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. semantic reasoning of the Wilko opinion does not control [a case based on § 10(b)]”). Finally, even assuming the conferees had an understanding of existing law that all agreed upon, they specifically disclaimed any intent to change it. Hence, the Wilko issue was left to the courts: it was unaffected by the amendment to § 28(b). This statement of congressional inaction simply does not support the proposition that the 1975 Congress intended to engraft onto unamended § 29(a) a meaning different from that of the enacting Congress. We conclude, therefore, that Congress did not intend for § 29(a) to bar enforcement of all predispute arbitration agreements. In this case, where the SEC has sufficient statutory authority to ensure that arbitration is adequate to vindicate Exchange Act rights, enforcement does not effect a waiver of “compliance with any provision” of the Exchange Act under § 29(a). Accordingly, we hold the McMahons’ agreements to arbitrate Exchange Act claims “enforce[able] ... in accord with the explicit provisions of the Arbitration Act.” Scherk v. Alberto-Culver Co., supra, at 520. IV Unlike the Exchange Act, there is nothing in the text of the RICO statute that even arguably evinces congressional intent to exclude civil RICO claims from the dictates of the Arbitration Act. This silence in the text is matched by silence in the statute’s legislative history. The private trebledamages provision codified as 18 U. S. C. § 1964(c) was added to the House version of the bill after the bill had been passed by the Senate, and it received only abbreviated discussion in either House. See Sedima, S. P. R. L. v. Imrex Co., 473 U. S. 479, 486-488 (1985). There is no hint in these legislative debates that Congress intended for RICO trebledamages claims to be excluded from the ambit of the Arbitration Act. See Geneseo, Inc. v. T. Kakiuchi & Co., Ltd., SHEARSON/AMERICAN EXPRESS INC. v. McMAHON 239 220 Opinion of the Court 815 F. 2d 840, 850-851 (CA2 1987); Mayaja, Inc. v. Bodkin, 803 F. 2d 157, 164 (CA5 1986). Because RICO’s text and legislative history fail to reveal any intent to override the provisions of the Arbitration Act, the McMahons must argue that there is an irreconcilable conflict between arbitration and RICO’s underlying purposes. Our decision in Mitsubishi Motors Corp. v. Soler Chrysler-Plymouth, Inc., 473 U. S. 614 (1985), however, already has addressed many of the grounds given by the McMahons to support this claim. In Mitsubishi, we held that nothing in the nature of the federal antitrust laws prohibits parties from agreeing to arbitrate antitrust claims arising out of international commercial transactions. Although the holding in Mitsubishi was limited to the international context, see id., at 629, much of its reasoning is equally applicable here. Thus, for example, the McMahons have argued that RICO claims are too complex to be subject to arbitration. We determined in Mitsubishi, however, that “potential complexity should not suffice to ward off arbitration.” Id., at 633. Antitrust matters are every bit as complex as RICO claims, but we found that the “adaptability and access to expertise” characteristic of arbitration rebutted the view “that an arbitral tribunal could not properly handle an antitrust matter.” Id., at 633-634. Likewise, the McMahons contend that the “overlap” between RICO’s civil and criminal provisions renders § 1964(c) claims nonarbitrable. See Page v. Moseley, Hallgarten, Estabrook & Weeden, Inc., 806 F. 2d 291, 299, n. 13 (CAI 1986) (“[T]he makings of a ‘pattern of racketeering’ are not yet clear, but the fact remains that a ‘pattern’ for civil purposes is a ‘pattern’ for criminal purposes”). Yet § 1964(c) is no different in this respect from the federal antitrust laws. In Sedima, S. P. R. L. v. Imrex Co., supra, we rejected the view that § 1964(c) “provide[s] civil remedies for offenses criminal in nature.” See 473 U. S., at 492. In doing so, this Court observed: “[T]he fact that conduct can result in 240 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. both criminal liability and treble damages does not mean that there is not a bona fide civil action. The familiar provisions for both criminal liability and treble damages under the antitrust laws indicate as much.” Ibid. Mitsubishi recognized that treble-damages suits for claims arising under § 1 of the Sherman Act may be subject to arbitration, even though such conduct may also give rise to claims of criminal liability. See Mitsubishi Motors Corp. v. Soler Chrysler-Plymouth, Inc., supra. We similarly find that the criminal provisions of RICO do not preclude arbitration of bona fide civil actions brought under § 1964(c). The McMahons’ final argument is that the public interest in the enforcement of RICO precludes its submission to arbitration. Mitsubishi again is relevant to the question. In that case we thoroughly examined the legislative intent behind §4 of the Clayton Act in assaying whether the importance of the private treble-damages remedy in enforcing the antitrust laws precluded arbitration of § 4 claims. We found that “[notwithstanding its important incidental policing function, the treble-damages cause of action . . . seeks primarily to enable an injured competitor to gain compensation for that injury.” 473 U. S., at 635. Emphasizing the priority of the compensatory function of § 4 over its deterrent function, Mitsubishi concluded that “so long as the prospective litigant effectively may vindicate its statutory cause of action in the arbitral forum, the statute will continue to serve both its remedial and deterrent function.” Id., at 637. The legislative history of § 1964(c) reveals the same emphasis on the remedial role of the treble-damages provision. In introducing the treble-damages provision to the House Judiciary Committee, Representative Steiger stressed that “those who have been wronged by organized crime should at least be given access to a legal remedy.” Hearings on S. 30 and Related Proposals before Subcommittee No. 5 of the House Committee on the Judiciary, 91st Cong., 2d Sess., 520 (1970). The policing function of § 1964(c), although impor- SHEARSON/AMERICAN EXPRESS INC. v. McMAHON 241 220 Opinion of the Court tant, was a secondary concern. See ibid. (“In addition, the availability of such a remedy would enhance the effectiveness of title IX’s prohibitions”). During the congressional debates on § 1964(c), Representative Steiger again emphasized the remedial purpose of the provision: “It is the intent of this body, I am certain, to see that innocent parties who are the victims of organized crime have a right to obtain proper redress. ... It represents the one opportunity for those of us who have been seriously affected by organized crime activity to recover.” 116 Cong. Rec. 35346-35347 (1970). This focus on the remedial function of § 1964(c) is reinforced by the recurrent references in the legislative debates to §4 of the Clayton Act as the model for the RICO treble-damages provision. See, e. g., 116 Cong. Rec. 35346 (statement of Rep. Poff) (RICO provision “has its counterpart almost in haec verba in the antitrust statutes”); id., at 25190 (statement of Sen. McClellan) (proposed amendment would “authorize private civil damage suits based upon the concept of section 4 of the Clayton Antitrust Act”). See generally Sedima, S. P. R. L. v. Imrex Co., 473 U. S., at 489 (“The clearest current in [RICO’s] history is the reliance on the Clayton Act model”). Not only does Mitsubishi support the arbitrability of RICO claims, but there is even more reason to suppose that arbitration will adequately serve the purposes of RICO than that it will adequately protect private enforcement of the antitrust laws. Antitrust violations generally have a widespread impact on national markets as a whole, and the antitrust trebledamages provision gives private parties an incentive to bring civil suits that serve to advance the national interest in a competitive economy. See Lindsay, “Public” Rights and Private Forums: Predispute Arbitration Agreements and Securities Litigation, 20 Loyola (LA) L. Rev. 643, 691-692 (1987). RICO’s drafters likewise sought to provide vigorous incentives for plaintiffs to pursue RICO claims that would advance society’s fight against organized crime. See Sedima, 242 OCTOBER TERM, 1986 Opinion of Blackmun, J. 482 U. S. S. P. R. L. v. Imrex Co., supra, at 498. But in fact RICO actions are seldom asserted “against the archetypal, intimidating mobster.” Id., at 499; see also id., at 506 (Marshall, J., dissenting) (“[O]nly 9% of all civil RICO cases have involved allegations of criminal activity normally associated with professional criminals”). The special incentives necessary to encourage civil enforcement actions against organized crime do not support nonarbitrability of run-of-the-mill civil RICO claims brought against legitimate enterprises. The private attorney general role for the typical RICO plaintiff is simply less plausible than it is for the typical antitrust plaintiff, and does not support a finding that there is an irreconcilable conflict between arbitration and enforcement of the RICO statute. In sum, we find no basis for concluding that Congress intended to prevent enforcement of agreements to arbitrate RICO claims. The McMahons may effectively vindicate their RICO claim in an arbitral forum, and therefore there is no inherent conflict between arbitration and the purposes underlying § 1964(c). Moreover, nothing in RICO’s text or legislative history otherwise demonstrates congressional intent to make an exception to the Arbitration Act for RICO claims. Accordingly, the McMahons, “having made the bargain to arbitrate,” will be held to their bargain. Their RICO claim is arbitrable under the terms of the Arbitration Act. V Accordingly, the judgment of the Court of Appeals for the Second Circuit is reversed, and the case is remanded for further proceedings consistent with this opinion. It is so ordered. Justice Blackmun, with whom Justice Brennan and Justice Marshall join, concurring in part and dissenting in part. I concur in the Court’s decision to enforce the arbitration agreement with respect to respondents’ RICO claims and thus SHEARSON/AMERICAN EXPRESS INC. v. McMAHON 243 220 Opinion of Blackmun, J. join Parts I, II, and IV of the Court’s opinion. I disagree, however, with the Court’s conclusion that respondents’ § 10(b) claims also are subject to arbitration. Both the Securities Act of 1933 and the Securities Exchange Act of 1934 were enacted to protect investors from predatory behavior of securities industry personnel. In Wilko v. Swan, 346 U. S. 427 (1953), the Court recognized this basic purpose when it declined to enforce a predispute agreement to compel arbitration of claims under the Securities Act. Following that decision, lower courts extended Wilko's reasoning to claims brought under § 10(b) of the Exchange Act, and Congress approved of this extension. In today’s decision, however, the Court effectively overrules Wilko by accepting the Securities and Exchange Commission’s newly adopted position that arbitration procedures in the securities industry and the Commission’s oversight of the self-regulatory organizations (SROs) have improved greatly since Wilko was decided. The Court thus approves the abandonment of the judiciary’s role in the resolution of claims under the Exchange Act and leaves such claims to the arbitral forum of the securities industry at a time when the industry’s abuses towards investors are more apparent than ever. I At the outset, it is useful to review the manner by which the issue decided today has been kept alive inappropriately by this Court. As the majority explains, Wilko was limited to the holding “that a predispute agreement could not be enforced to compel arbitration of a claim arising under § 12(2) of the Securities Act.” Ante, at 228. Relying, however, on the reasoning of Wilko and the similarity between the pertinent provisions of the Securities Act and those of the Exchange Act, lower courts extended the Wilko holding to claims under the Exchange Act and refused to enforce predispute agreements to arbitrate them as well. See, e. g., Greater Continental Corp. v. Schechter, 422 F. 2d 1100, 1103 244 OCTOBER TERM, 1986 Opinion of Blackmun, J. 482 U. S. (CA2 1970) (dicta); Moran v. Paine, Webber, Jackson & Curtis, 389 F. 2d 242, 245-246 (CA3 1968). In Scherk n. Alberto-Culver Co., 417 U. S. 506 (1974), the Court addressed the question whether a particular predispute agreement to arbitrate § 10(b) claims should be enforced. Because that litigation involved international business concerns and because the case was decided on such grounds, the Court did not reach the issue of the extension of Wilko to § 10(b) claims. The Court, nonetheless, included in its opinion dicta noting that “a colorable argument could be made that even the semantic reasoning of the Wilko opinion does not control the case before us.” 417 U. S., at 513. There is no need to discuss in any detail that “colorable argument,” which rests on alleged distinctions between pertinent provisions of the Securities Act and those of the Exchange Act, because the Court does not rely upon it today.1 In fact, 1 The “colorable argument” amounted to a listing by the Scherk Court of the differences between a § 12(2) action, as it had been described by the Wilko Court, and a § 10(b) action under the Exchange Act. First, the Court noted that, while § 12(2) of the Securities Act provided an express cause of action, § 10(b) did not contain on its face such a cause of action, which, instead, had been implied from its language and that of Rule 10b-5. Scherk v. Alberto-Culver Co., 417 U. S. 506, 513 (1974). Second, the Court explained that the Exchange Act did not set forth the “special right” that the Wilko Court found established in § 12(2). 417 U. S., at 513-514; see also Wilko v. Swan, 346 U. S. 427, 431 (1953) (§ 12(2) right viewed as “special” because of differences between that right and a common-law cause of action, differences that favored the investor). Finally, the Court observed that the jurisdictional provisions of the two Acts were not the same. 417 U. S., at 514. Under § 22(a) of the Securities Act, 48 Stat. 86, as amended, 15 U. S. C. § 77v(a), suit could be brought in federal or state court, whereas, under § 27 of the Exchange Act, 48 Stat. 902, as amended, 15 U. S. C. § 78aa, suit could be brought only in federal court. In sum, the overall thrust of the “colorable argument,” as stated by the Court in Scherk, seemed to be as follows: The Wilko Court declined to enforce arbitration of § 12(2) claims because it found significant the special nature of that cause of action, but a similar concern does not apply to § 10(b) claims, which are neither “special” nor “express.” SHEARSON/AMERICAN EXPRESS INC. v. McMAHON 245 220 Opinion of Blackmun, J. the “argument” is important not so much for its substance2 as it is for its litigation role. It simply constituted a way of keeping the issue of the arbitrability of § 10(b) claims alive for those opposed to the result in Wilko. 2 That the Court passes over the “colorable argument” in silence, although petitioners have advanced it, see Brief for Petitioners 19-28, would appear to relegate that argument to its proper place in the graveyard of ideas. As the Commission explains in its brief, see Brief for Securities and Exchange Commission as Amicus Curiae 22-23, and nn. 18-19 (Brief), the procedural protections surrounding a § 10(b) action and its difference from a common-law action are as pronounced as those of a § 12(2) claim. More importantly, “Section 10(b) is just as much a ‘provision’ of the 1934 Act, with which persons trading in securities are required to ‘comply,’ as Section 12(2) is of the 1933 Act.” Brief 24. To state otherwise “might be interpreted as suggesting that the Section 10(b) implied right of action is somehow inferior to express rights,” which is “incompatible with the importance of the Section 10(b) remedy in the arsenal of securities law protections.” Id., at 26. And the difference in the jurisdictional provisions is not significant: as the Commission explains, the proper question is whether a § 10(b) or § 12(2) claimant is entitled to a judicial forum, not whether the claimant has a choice between judicial fora. Brief 22, n. 17. In fact, the limitation of § 10(b) actions to federal court argues against enforcing predispute arbitration agreements as to such actions. Because Congress gave the federal courts exclusive jurisdiction over § 10(b) claims, it may have intended them to develop an exclusive jurisprudence of § 10(b). See, e. g., Conover v. Dean Witter Reynolds, Inc., 794 F. 2d 520, 527 (CA9 1986), cert, pending, No. 86-321. Commentators, almost uniformly, have rejected the “colorable argument.” See, e. g., Comment, Predispute Arbitration Agreements Between Brokers and Investors: The Extension of Wilko to Section 10(b) Claims, 46 Md. L. Rev. 339, 364-366 (1987) (Maryland Comment); Brown, Shell, & Tyson, Arbitration of Customer-Broker Disputes Arising Under the Federal Securities Laws and RICO, 15 Sec. Reg. L. J. 3, 18-19 (1987) (Brown, Shell, & Tyson); Malcolm & Segall, The Arbitrability of Claims Arising Under Section 10(b) of the Securities Exchange Act: Should Wilko Be Extended?, 50 Albany L. Rev. 725, 748-751 (1986) (Malcolm & Segall); Note, Arbitrability of Claims Arising Under the Securities Exchange Act of 1934, 1986 Duke L. J. 548, 565-570 (Duke Note). But see Note, Arbitrability of Implied Rights of Action Under Section 10(b) of the Securities Exchange Act, 61 N. Y. U. L. Rev. 506, 520-526 (1986). 246 OCTOBER TERM, 1986 Opinion of Blackmun, J. 482 U. S. If, however, there could have been any doubts about the extension of Wilko’s holding to § 10(b) claims, they were undermined by Congress in its 1975 amendments to the Exchange Act. The Court questions the significance of these amendments, which, as it notes, concerned, among other things, provisions dealing with dispute resolution and disciplinary action by an SRO towards its own members. See ante, at 235-236. These amendments, however, are regarded as “the ‘most substantial and significant revision of this country’s Federal securities laws since the passage of the Securities Exchange Act in 1934.’” Herman & MacLean v. Huddleston, 459 U. S. 375, 384-385 (1983), quoting Securities Acts Amendments of 1975: Hearings on S. 249 before the Subcommittee on Securities of the Senate Committee on Banking, Housing and Urban Affairs, 94th Cong., 1st Sess., 1 (1975) (Hearings).3 More importantly, in enacting these amendments, Congress specifically was considering exceptions to § 29(a), 15 U. S. C. § 78cc, the nonwaiver provision of the Exchange Act, a provision primarily designed with the protection of investors in mind.4 The statement from the 8 Senator Williams, Chairman of the Subcommittee, observed: “This legislation represents the product of nearly 4 years of studies, investigations, and hearings. It has been carefully designed to improve the efficiency of the securities markets and to increase investor protection. It is reform legislation in the very best sense, for it will lay the foundation for a stronger and more profitable securities industry while assuring that investors are more economically and effectively served.” Hearings 1. 4 The text of one of the amendments suggests that Congress had investors in mind when making them. Although, as the Court observes, ante, at 235-236, § 28(b) deals only with disputes among securities-industry professionals, the amendment to § 15B, which permitted arbitration among municipal-securities brokers-dealers, provided that “no person other than a municipal securities broker, municipal securities dealer, or person associated with such a municipal securities broker or municipal securities dealer may be compelled to submit to such arbitration except at his instance and in accordance with section 29 of this title.” 89 Stat. 133, 15 U. S. C. § 78o-4(b)(2)(D); see also Brown, Shell, & Tyson, at 20. SHEARSON/AMERICAN EXPRESS INC. v. McMAHON 247 220 Opinion of Blackmun, J. legislative history, cited by the Court, ante, at 236-237, on its face indicates that Congress did not want the amendments to overrule Wilko. Moreover, the fact that this statement was made in an amendment to the Exchange Act suggests that Congress was aware of the extension of Wilko to § 10(b) claims. Although the remark does not necessarily signify Congress’ endorsement of this extension, in the absence of any prior congressional indication to the contrary, it implies that Congress was not concerned with arresting this trend.5 Such inaction during a wholesale revision of the securities laws, a revision designed to further investor protection, would argue in favor of Congress’ approval of Wilko and its extension to § 10(b) claims. See Wolfe v. E. F. Hutton & Co., 800 F. 2d 1032, 1037-1038 (CA11 1986) (en banc), cert. 6 Although I agree that the remark from the legislative history does not state expressly Congress’ approval of Wilko’s extension to Exchange Act claims, I do not believe that there are “difficulties,” as the Court suggests, in interpreting that remark to suggest such approval. See ante, at 237. Certainly, by the 1975 amendments dealing with exceptions to § 29(a) of the Exchange Act, Congress was enacting provisions directly related to the general subject of Wilko and its extension to Exchange Act claims —the scope of the nonwaiver provision—contrary to the Court’s flat statement that these provisions were not “remotely addressing that subject,” see ante, at 237. Moreover, understanding the remark to imply Congress’ affirmation of Wilko and an awareness of Wilko’s extension to § 10(b) claims is not incompatible with several of the concerns at the center of the Court’s “difficulties.” Thus, Congress’ concern that a possible misreading of § 28(b) might affect Wilko’s actual holding as to § 12(2) claims, see ante, at 237-238, is consistent with this understanding. In addition, the mention of “existing law” could very well have referred both to the Court’s decision in Scherk, where the Court assumed that Wilko could be applied to § 10(b) claims, see 417 U. S., at 515, and to holdings by the lower courts. I disagree with the Court’s assertion that Congress left the Wilko issue to the courts by way of its statement that it did not change existing law. Ante, at 238. Common sense suggests that, when Congress states that it is not changing the law, while at the same time undertaking extensive amendments to a particular area of the law, one can assume that Congress is approving the law in existence. See Herman & MacLean v. Huddleston, 459 U. S. 375, 384-386 (1983). 248 OCTOBER TERM, 1986 Opinion of Blackmun, J. 482 U. S. pending, No. 86-1218; cf. Herman & MacLean v. Huddleston, 459 U. S., at 384-386. One would have thought that, after these amendments, the matter of Wilko’s extension to Exchange Act claims at last would be uncontroversial. In the years following the Scherk decision, all the Courts of Appeals treating the issue so interpreted Wilko.6 In Dean Witter Reynolds Inc. v. Byrd, 470 U. S. 213 (1985), this Court declined to address the extension issue, which was not before it, but recognized the development in the case law. Id., at 215, n. 1. Yet, like a ghost reluctant to accept its eternal rest, the “colorable argument” surfaced again, this time in a concurring opinion. See id., at 224 (White, J.). That concurring opinion repeated the “argument,” but with no more development than the Scherk Court had given it.7 Where there had been uniformity in 6See Raiford v. Buslease Inc., 745 F. 2d 1419, 1421 (CA11 1984); Surman v. Merrill Lynch, Pierce, Fenner & Smith, 733 F. 2d 59, 61 (CA8 1984) (dictum); Ingbar v. Drexel Burnham Lambert Inc., 683 F. 2d 603, 605 (CAI 1982) (same); De Lande v. Birr, Wilson & Co., 648 F. 2d 1255, 1257-1259 (CA9 1981) (same); Mansbach v. Prescott, Ball & Turben, 598 F. 2d 1017, 1030 (CA6 1979); Merrill Lynch, Pierce, Fenner & Smith, Inc. v. Moore, 590 F. 2d 823, 827-829 (CAIO 1978); Weissbuch v. Merrill Lynch, Pierce, Fenner & Smith, Inc., 558 F. 2d 831, 833-836 (CA7 1977); Allegaert v. Perot, 548 F. 2d 432, 437-438 (CA2), cert, denied, 432 U. S. 910 (1977); Sibley v. Tandy Corp., 543 F. 2d 540, 543, and n. 3 (CA5 1976), cert, denied, 434 U. S. 824 (1977); Ayres v. Merrill Lynch, Pierce, Fenner & Smith, Inc., 538 F. 2d 532, 536-537 (CA3), cert, denied, 429 U. S. 1010 (1976). 7 Although in his concurrence, Justice White observed that the application of Wilko to § 10(b) claims was a “matter of substantial doubt,” Dean Witter Reynolds Inc. v. Byrd, 470 U. S., at 224, and stated that “the contrary holdings of the lower courts must be viewed with some doubt,” id., at 225, the only reasons offered for these assertions were those of the Scherk Court. The concurring opinion nowhere discussed the reasoning of the lower courts’ subsequent decisions, particularly their justification for the extension of Wilko because of the similar concern for the protection of investors that informed both the Securities Act and the Exchange Act. See, e. g., Weissbuch v. Merrill Lynch, Pierce, Fenner & Smith, Inc., 558 F. 2d, at 835. SHEARSON/AMERICAN EXPRESS INC. v. McMAHON 249 220 Opinion of Blackmun, J. the lower courts before Byrd, there now appeared disharmony on the issue of the arbitrability of § 10(b) claims.8 And, as the Court observes, see ante, at 225, we granted certiorari in this case to resolve this conflict among the Courts of Appeals. II There are essentially two problems with the Court’s conclusion that predispute agreements to arbitrate § 10(b) claims may be enforced. First, the Court gives Wilko an overly narrow reading so that it can fit into the syllogism offered by the Commission and accepted by the Court, namely, (1) Wilko 8 In the wake of the Byrd decision, the “colorable argument” took on another life as courts followed the suggestion of the concurrence. See, e. g., Page v. Moseley, Hallgarten, Estabrook & Weeden, Inc., 806 F. 2d 291, 296-298 (CAI 1986); Phillips v. Merrill Lynch, Pierce, Fenner & Smith, Inc., 795 F. 2d 1393, 1397-1398 (CA8 1986), cert, pending, No. 86-578; see also Duke Note 548, n. 7 (citing Federal District Court cases). It is somewhat curious that this “colorable argument” was taken up by many lower courts, often without any analysis on this point, even though the Court in Byrd specifically declined to address the issue, which was not before it. See 470 U. S., at 215, n. 1. Other courts reaffirmed their pre-Byrd holdings that § 10(b) claims were nonarbitrable. See Sterne v. Dean Witter Reynolds, Inc., 808 F. 2d 480, 483 (CA6 1987); Jacobson v. Merrill Lynch, Pierce, Fenner & Smith, Inc., 797 F. 2d 1197, 1202 (CA3 1986), cert, pending, No. 86-487; King v. Drexel Burnham Lambert, Inc., 796 F. 2d 59, 60 (CA5 1986), cert, pending, No. 86-282; 788 F. 2d 94, 98 (CA2 1986) (case below). Two courts, which reexamined the issue, came to the same result on the basis of the similarities between the provisions of both Acts and the policies underlying them. See Conover v. Dean Witter Reynolds, Inc., 794 F. 2d, at 527; Wolfe v. E. F. Hutton & Co., 800 F. 2d 1032, 1036-1037 (CA11 1986) (en banc), cert, pending, No. 86-1218. To a certain extent, the new popularity of the “colorable argument” was not unrelated to the belief that the judicial attitude toward arbitration had changed and that Wilko should be reconsidered because of this change. See Phillips v. Merrill Lynch, Pierce, Fenner & Smith, Inc., 795 F. 2d, at 1395, 1398, n. 16. One commentator observed: “The differences adduced by Justice White merely act as a wedge to hold the door open for this policy favoring arbitration.” Maryland Comment 356, n. 149. 250 OCTOBER TERM, 1986 Opinion of Blackmun, J. 482 U. S. was really a case concerning whether arbitration was adequate for the enforcement of the substantive provisions of the securities laws; (2) all of the Wilko Court’s doubts as to arbitration’s adequacy are outdated; (3) thus Wilko is no longer good law. See ante, at 228-229,232; Brief for Securities and Exchange Commission as Amicus Curiae 10. Second, the Court accepts uncritically petitioners’ and the Commission’s argument that the problems with arbitration, highlighted by the Wilko Court, either no longer exist or are not now viewed as problems by the Court. This acceptance primarily is based upon the Court’s belief in the Commission’s representations that its oversight of the SROs ensures the adequacy of arbitration. A I agree with the Court’s observation that, in order to establish an exception to the Arbitration Act, 9 U. S. C. §1 et seq., for a class of statutory claims, there must be “an intention discernible from the text, history, or purposes of the statute.” Ante, at 227. Where the Court first goes wrong, however, is in its failure to acknowledge that the Exchange Act, like the Securities Act, constitutes such an exception. This failure is made possible only by the unduly narrow reading of Wilko that ignores the Court’s determination there that the Securities Act was an exception to the Arbitration Act. The Court’s reading is particularly startling because it is in direct contradiction to the interpretation of Wilko given by the Court in Mitsubishi Motors Corp. n. Soler Chrysler-Plymouth, Inc., 473 U. S. 614 (1985), a decision on which the Court relies for its strong statement of a federal policy in favor of arbitration. But we observed in Mitsubishi: “Just as it is the congressional policy manifested in the Federal Arbitration Act that requires courts liberally to construe the scope of arbitration agreements covered by that Act, it is the congressional intention expressed in some other statute on which the courts must rely to identify any category of claims as to which agreements to ar- SHEARSON/AMERICAN EXPRESS INC. u McMAHON 251 220 Opinion of Blackmun, J. bitrate will be held unenforceable. See Wilko v. Swan, 346 U. S., at 434-435 .... We must assume that if Congress intended the substantive protection afforded by a given statute to include protection against waiver of the right to a judicial forum, that intention will be deducible from text or legislative history. See Wilko v. Swan, supra.” Id., at 627-628. Such language clearly suggests that, in Mitsubishi, we viewed Wilko as holding that the text and legislative history of the Securities Act—not general problems with arbitration-established that the Securities Act constituted an exception to the Arbitration Act. In a surprising display of logic, the Court uses Mitsubishi as support for the virtues of arbitration and thus as a means for undermining Wilko’s holding, but fails to take into account the most pertinent language in Mitsubishi. It is not necessary to rely just on the statement in Mitsubishi to realize that in Wilko the Court had before it the issue of congressional intent to exempt statutory claims from the reach of the Arbitration Act. One has only to reread the Wilko opinion without the constricted vision of the Court. The Court’s misreading is possible because, while extolling the policies of the Arbitration Act, it is insensitive to, and disregards the policies of, the Securities Act. This Act was passed in 1933, eight years after the Arbitration Act of 1925, see 43 Stat. 883, and in response to the market crash of 1929. The Act was designed to remedy abuses in the securities industry, particularly fraud and misrepresentation by securities-industry personnel, that had contributed to that disastrous event. See Malcolm & Segall 730-731. It had as its main goal investor protection, which took the form of an effort to place investors on an equal footing with those in the securities industry by promoting full disclosure of information on investments. See L. Loss, Fundamentals of Securities Regulation 36 (1983). 252 OCTOBER TERM, 1986 Opinion of Blackmun, J. 482 U. S. The Court in Wilko recognized the policy of investor protection in the Securities Act. It was this recognition that animated its discussion of whether § 14, 48 Stat. 84, 15 U. S. C. § 77n, the nonwaiver provision of the Securities Act, applied to § 22(a), 48 Stat. 86, as amended, 15 U. S. C. §77v(a), the provision that gave an investor a judicial forum for the resolution of securities disputes. In the Court’s words, the Securities Act, “[d]esigned to protect investors, . . . requires issuers, underwriters, and dealers to make full and fair disclosure of the character of securities sold in interstate and foreign commerce and to prevent fraud in their sale.” 346 U. S., at 431. The Court then noted that, to promote this policy in the Act, Congress had designed an elaborate statutory structure: it gave investors a “special right” of suit under § 12(2); they could bring the suit in federal or state court pursuant to § 22(a); and, if brought in federal court, there were numerous procedural advantages, such as nationwide service of process. Ibid. In reasoning that a predispute agreement to arbitrate §12(2) claims would constitute a “waiver” of a provision of the Act, i. e., the right to the judicial forum embodied in § 22(a), the Court specifically referred to the policy of investor protection underlying the Act: “While a buyer and seller of securities, under some circumstances, may deal at arm’s length on equal terms, it is clear that the Securities Act was drafted with an eye to the disadvantages under which buyers labor. Issuers of and dealers in securities have better opportunities to investigate and appraise the prospective earnings and business plans affecting securities than buyers. It is therefore reasonable for Congress to put buyers of securities covered by that Act on a different basis from other purchasers. “When the security buyer, prior to any violation of the Securities Act, waives his right to sue in courts, he gives up more than would a participant in other business transactions. The security buyer has a wider choice of courts SHEARSON/AMERICAN EXPRESS INC. v. McMAHON 253 220 Opinion of Blackmun, J. and venue. He thus surrenders one of the advantages the Act gives him and surrenders it at a time when he is less able to judge the weight of the handicap the Securities Act places upon his adversary.” Id., at 435. In the Court’s view, the express language, legislative history, and purposes of the Securities Act all made predispute agreements to arbitrate § 12(2) claims unenforceable despite the presence of the Arbitration Act.9 9 In discussing the similar nonwaiver provision under the Exchange Act, § 29(a), 48 Stat. 903, as amended, 15 U. S. C. § 78cc(a), the Court now suggests that it can be read only to mean that an investor cannot waive security-investment personnel’s “compliance” with a duty under the statute. See ante, at 228. The Court implies that the literal language of § 29(a) does not apply to an investor’s waiver of his own action. See ibid.; see also Brief for Petitioners 28-33; Fletcher, Privatizing Securities Disputes Through the Enforcement of Arbitration Agreements, 71 Minn. L. Rev. 393, 422-423 (1987) (Fletcher). It appears, however, that in Wilko the Court understood the nonwaiver provision also to mean that, at least in the predispute context, an investor could not waive his compliance with the provision for dispute resolution in the courts. This reading of the antiwaiver provision makes sense in terms of the policy of investor protection. To counteract the inherent superior position of the securities-industry professional, up to and including the time when a dispute might occur between a broker and the investor, Congress intended to place the investor on “a different basis from other purchasers.” 346 U. 8., at 435. Construing § 14 not to allow the investor to waive his right to a judicial forum in the predispute setting serves this congressional purpose of maintaining the investor in a special position. As one recognized commentator has noted, in the securities Acts “Congress did not take away from the citizen ‘his inalienable right to make a fool of himself.’ It simply attempted to prevent others from making a fool of him.” L. Loss, Fundamentals of Securities Regulation 36 (1983), quoting in part 1935 Report of the (Canadian) Royal Commission on Price Spreads 38. In Wilko, the Court did not discuss the situation where parties, after a dispute has arisen, enter into an agreement to arbitrate. 346 U. S., at 438 (Jackson, J., concurring). Courts have generally allowed enforcement of arbitration agreements in such circumstances despite the language of § 14, provided that the investor has made an informed waiver. See, e. g., Coenen v. R. W. Pressprich & Co., 453 F. 2d 1209, 1213 (CA2), cert, denied, 406 U. S. 949 (1972); Moran v. Paine, Webber, Jackson & Curtis, 389 254 OCTOBER TERM, 1986 Opinion of Blackmun, J. 482 U. S. Accordingly, the Court seriously errs when it states that the result in Wilko turned only on the perceived inadequacy of arbitration for the enforcement of § 12(2) claims. It is true that the Wilko Court discussed the inadequacies of this process, 346 U. S., at 435-437, and that this discussion constituted one ground for the Court’s decision. The discussion, however, occurred after the Court had concluded that the language, legislative history, and purposes of the Securities Act mandated an exception to the Arbitration Act for these securities claims. The Court’s decision in Scherk is consistent with this reading of Wilko, despite the Court’s suggestion to the contrary. See ante, at 229. Indeed, in reading Scherk as a case turning on the adequacy of arbitration, the Court completely ignores the central thrust of that decision. As the Court itself notes, ante, at 229, in Scherk the Court assumed that Wilko’s prohibition on enforcing predispute arbitration agreements ordinarily would extend to § 10(b) claims, such as those at issue in Scherk. The Scherk Court relied on a crucial difference between the international business situation presented to it and that before the Court in Wilko, where the laws of the United States, particularly the securities laws, clearly governed the dispute. Scherk, in contrast, presented F. 2d 242, 245-246 (CA3 1968); see also Duke Note 558, and nn. 59, 60. This distinction makes sense when one considers that the Court’s reading of § 14 to bar an investor’s “waiver” of the judicial forum in the predispute setting emphasized the moment when this waiver occurred—“at a time when he is less able to judge the weight of the handicap the Securities Act places upon his adversary.” 346 U. S., at 435. An investor would not be working under this disadvantage once a dispute has arisen. With the awareness—heightened by the reality of an actual dispute—of the possible benefits he would derive from proceeding in court and the possible burdens that his adversary would have to undergo, an investor might forgo the judicial forum for the quick resolution of the conflict in arbitration. He thus would remain master of the situation and in the special position Congress intended him to have. SHEARSON/AMERICAN EXPRESS INC. v. McMAHON 255 220 Opinion of Blackmun, J. a multinational conflict-of-laws puzzle.10 In such a situation, the Court observed, a contract provision setting forth a particular forum and the law to apply for possible disputes was “an almost indispensable precondition to achievement of the orderliness and predictability essential to any international business transaction.” 417 U. S., at 516. Indeed, the Court thought that failure to enforce such an agreement to arbitrate in this international context would encourage companies to file suits in countries where the law was most favorable to them, which “would surely damage the fabric of international commerce and trade, and imperil the willingness and ability of businessmen to enter into international commercial agreements.” Id., at 517. Accordingly, the Scherk decision turned on the special nature of agreements to arbitrate in the international commercial context.11 10 The Scherk Court observed: “Alberto-Culver is an American corporation with its principal place of business and the vast bulk of its activity in this country, while Scherk is a citizen of Germany whose companies were organized under the laws of Germany and Liechtenstein. The negotiations leading to the signing of the contract in Austria and to the closing in Switzerland took place in the United States, England, and Germany, and involved consultations with legal and trademark experts from each of those countries and from Liechtenstein. Finally, and most significantly, the subject matter of the contract concerned the sale of business enterprises organized under the laws of and primarily situated in European countries, whose activities were largely, if not entirely, directed to European markets.” 417 U. S., at 515. 11 This reading of Scherk is entirely consistent with our explanation of that decision in Mitsubishi Motors Corp. v. Soler Chrysler-Plymouth, Inc., 473 U. S. 614 (1985), a case that also involved an agreement to arbitrate in the international business context. There, citing Scherk, we concluded that “concerns of international comity, respect for the capacities of foreign and transnational tribunals, and sensitivity to the need of the international commercial system for predictability in the resolution of disputes require that we enforce the parties’ agreement, even assuming that a contrary result would be forthcoming in a domestic context.” 473 U. S., at 629. In discussing that case at length, we expressed our agreement with the remark in Scherk that such arbitration agreements constituted “ ‘a 256 OCTOBER TERM, 1986 Opinion of Blackmun, J. 482 U. S. In light of a proper reading of Wilko, the pertinent question then becomes whether the language, legislative history, and purposes of the Exchange Act call for an exception to the Arbitration Act for § 10(b) claims. The Exchange Act waiver provision is virtually identical to that of the Securities Act.12 More importantly, the same concern with investor protection that motivated the Securities Act is evident in the Exchange Act, although the latter, in contrast to the former, is aimed at trading in the secondary securities market. See Ernst & Ernst n. Hochfelder, 425 U. S. 185, 195 (1976). We have recognized that both Acts were designed with this common purpose in mind. See id., at 206 (“The 1933 and 1934 Acts constitute interrelated components of the federal regulatory scheme governing transactions in securities”). Indeed, the application of both Acts to the same conduct, see Brown, Shell, & Tyson 16, suggests that they have the same basic goal. And we have approved a cumulative construction of remedies under the securities Acts to promote the maximum possible protection of investors. See Herman & MacLean n. Huddleston, 459 U. S., at 384-385.13 In sum, the same reasons that led the Court to find an exception to the Arbitration Act for § 12(2) claims exist for specialized kind of forum-selection clause.’” 473 U. S., at 630, quoting Scherk, 417 U. S., at 519. 12 Compare 15 U. S. C. § 78cc(a) (“Any condition, stipulation, or provision binding any person to waive compliance with any provision of this chapter or of any rule or regulation thereunder, or of any rule of an exchange required thereby shall be void”) with 15 U. S. C. § 77n (“Any condition, stipulation, or provision binding any person acquiring any security to waive compliance with any provision of this subchapter or of the rules and regulations of the Commission shall be void”). 13 Courts that initially rejected the “colorable argument” after Scherk and approved of the extension of Wilko to Exchange Act claims acknowledged the similarity between the policies of the two Acts. See, e. g., Weissbuch v. Merrill Lynch, Pierce, Fenner & Smith Inc., 558 F. 2d, at 835. Courts that have rejected the “colorable argument” after Byrd have engaged in a similar analysis. See, e. g., Wolfe v. E. F. Hutton & Co., 800 F. 2d, at 1035. SHEARSON/AMERICAN EXPRESS INC. v. McMAHON 257 220 Opinion of Blackmun, J. § 10(b) claims as well. It is clear that Wilko, when properly read, governs the instant case and mandates that a predispute arbitration agreement should not be enforced as to § 10(b) claims. B Even if I were to accept the Court’s narrow reading of Wilko as a case dealing only with the inadequacies of arbitration in 1953,14 I do not think that this case should be resolved differently today so long as the policy of investor protection is given proper consideration in the analysis. Despite improvements in the process of arbitration and changes in the judicial attitude towards it, several aspects of arbitration that were seen by the Wilko court to be inimical to the policy of investor protection still remain. Moreover, I have serious reservations about the Commission’s contention that its oversight of the SROs’ arbitration procedures will ensure that the process is adequate to protect an investor’s rights under the securities Acts. As the Court observes, ante, at 231, in Wilko the Court was disturbed by several characteristics of arbitration that made such a process inadequate to safeguard the special position in which the Securities Act had placed the investor. The Court concluded that judicial review of the arbitrators’ application of the securities laws would be difficult because arbitrators were required neither to give the reasons for their decisions nor to make a complete record of their proceedings. See 346 U. S., at 436. The Court also observed that the grounds for vacating an arbitration award were limited. The Court noted that, under the Arbitration Act, there were only 14 This argument, in essence, is a functional one. It suggests that, although Congress intended to protect investors through the provision of a judicial forum for the enforcement of their rights under the securities Acts, this intention will not be contravened by sending these claims to arbitration because arbitration is now the “functional equivalent” of the courts. See Brief for Securities and Exchange Commission as Amicus Curiae 12; see also Maryland Comment 373. 258 OCTOBER TERM, 1986 Opinion of Blackmun, J. 482 U. S. four grounds for vacation of an award: fraud in procuring the award, partiality on the part of arbitrators, gross misconduct by arbitrators, and the failure of arbitrators to render a final decision. Id., at 436, n. 22, quoting 9 U. S. C. § 10 (1952 ed., Supp. V). The arbitrators’ interpretation of the law would be subject to judicial review only under the “manifest disregard” standard. 346 U. S., at 436. The Court today appears to argue that the Wilko Court’s assessment of arbitration’s inadequacy is outdated, first, because arbitration has improved since 1953, and second, because the Court no longer considers the criticisms of arbitration made in Wilko to be valid reasons why statutory claims, such as those under § 10(b), should not be sent to arbitration.15 It is true that arbitration procedures in the securities industry have improved since Wilko’s day. Of particular importance has been the development of a code of arbitration by the Commission with the assistance of representatives of the securities industry and the public. See Uniform Code of Arbitration, Exh. C, Fifth Report of the Securities Industry Conference on Arbitration 29 (Apr. 1986) (Fifth SICA Report).16 16 The Court does not mention specifically the improvements in arbitration as a reason for abandoning Wilko. This reason, however, is implied in the Court’s discussion of the Commission’s oversight of the SROs. See ante, at 233-234. 16 This Code has been used to harmonize the arbitration procedures among the SROs. See Katsoris, The Arbitration of a Public Securities Dispute, 53 Ford. L. Rev. 279, 283-284 (1984) (Katsoris). As the Commission explained: “[T]his [Code] marks a substantial improvement over the various arbitration procedures currently being utilized by the securities industry and represents an important step towards establishing a uniform system for resolving investor complaints through arbitration.” SEC Exchange Act Rei. No. 16390 (Nov. 30, 1979), 44 Fed. Reg. 70616, 70617. The rules of the Uniform Code provide for the selection of arbitrators and the manner in which the proceedings are conducted. See Fifth SICA Report; see also Code of Arbitration Procedure, CCH NASD Manual Ulf 3701-3744 (July 1986); Arbitration Rules 600-620, CCH American Stock Exchange Guide HU 9540-9551J (May 1986); Arbitration Rules 600-634, SHEARSON/AMERICAN EXPRESS INC. v. McMAHON 259 220 Opinion of Blackmun, J. Even those who favor the arbitration of securities claims do not contend, however, that arbitration has changed so significantly as to eliminate the essential characteristics noted by the Wilko Court. Indeed, proponents of arbitration would not see these characteristics as “problems,” because, in their view, the characteristics permit the unique “streamlined” nature of the arbitral process. As at the time of Wilko, preparation of a record of arbitration proceedings is not invariably required today.17 Moreover, arbitrators are not bound by precedent and are actually discouraged by their associations from giving reasons for a decision. See R. Coulson, Business Arbitration—What You Need to Know 29 (3d ed. 1986) (“Written opinions can be dangerous because they identify targets for the losing party to attack”); see also Duke Note 553; Fletcher 456-457. Judicial review is still substantially limited to the four grounds listed in § 10 of the Arbitration Act and to the concept of “manifest disregard” of the law. See, e. g., French n. Merrill Lynch, Pierce, Fenner & Smith, Inc., 784 F. 2d 902, 906 (CA9 1986), citing Swift Industries, Inc. v. Botany Industries, Inc., 466 F. 2d 1125, 1131 (CA3 1972) (an arbitrator’s decision must be upheld unless it is “‘completely irrational’”).18 CCH New York Stock Exchange Guide HI 2600-2634 (Mar. 1985). Some arbitration agreements permit arbitration before the American Arbitration Association, whose rules are similar to those in the above Codes. Brief for American Arbitration Association as Amicus Curiae 12-13, and App. B.; see also Fletcher 451. 17 Under the Uniform Code of Arbitration: “Unless fequested by the arbitrators or a party or parties to a dispute, no record of an arbitration proceeding shall be kept. If a record is kept, it shall be a verbatim record. If a party or parties to a dispute elect to have the record transcribed, the cost of such transcription shall be borne by the party or parties making the request.” Fifth SICA Report § 25, p. 36. 18 The Uniform Code of Arbitration and the SRO codes modeled upon it do provide for limited discovery, see Brief for Securities Industry Association, Inc., et al. as Amici Curiae 9, and the ability to subpoena witnesses, see Brief for American Arbitration Association as Amicus Curiae 13. Yet, by arbitrating their disputes, investors lose the wide choice of 260 OCTOBER TERM, 1986 Opinion of Blackmun, J. 482 U. S. The Court’s “mistrust” of arbitration may have given way recently to an acceptance of this process, not only because of the improvements in arbitration, but also because of the Court’s present assumption that the distinctive features of arbitration, its more quick and economical resolution of claims, do not render it inherently inadequate for the resolution of statutory claims. See Mitsubishi Motors Corp. n. Soler Chrysler-Plymouth, Inc., 473 U. S., at 633. Such reasoning, however, should prevail only in the absence of the congressional policy that places the statutory claimant in a special position with respect to possible violators of his statutory rights. As even the most ardent supporter of arbitration would recognize, the arbitral process at best places the investor on an equal footing with the securities-industry personnel against whom the claims are brought. Furthermore, there remains the danger that, at worst, compelling an investor to arbitrate securities claims puts him in a forum controlled by the securities industry. This result directly contradicts the goal of both securities Acts to free the investor from the control of the market professional. The Uniform Code provides some safeguards19 but despite them, and indeed because of the background of the arbitrators, the investor has the impression, frequently justified, that his claims are being judged by a forum composed of individuals sympathetic to the securities industry and not drawn venue and the extensive discovery provided by the courts. See Katsoris 287, n. 52. 19 The Uniform Code mandates that a majority of an arbitration panel, usually composed of between three to five arbitrators, be drawn from outside the industry. Fifth SICA Report §8(a), p. 31. Each arbitrator, moreover, is directed to disclose “any circumstances which might preclude such arbitrator from rendering an objective and impartial determination.” §11, p. 32. In addition, the parties are informed of the business associations of the arbitrators, § 9, and each party has the right to one peremptory challenge and to unlimited challenges for cause, § 10, p. 32. The arbitrators are usually individuals familiar with the federal securities laws. See Brener v. Becker Paribas Inc., 628 F. Supp. 442, 448 (SDNY 1985). SHEARSON/AMERICAN EXPRESS INC. v. McMAHON 261 220 Opinion of Blackmun, J. from the public. It is generally recognized that the codes do not define who falls into the category “not from the securities industry.” Brown, Shell, & Tyson 35, and n. 94; Katsoris 309-312. Accordingly, it is often possible for the “public” arbitrators to be attorneys or consultants whose clients have been exchange members or SROs. See Panel of Arbitrators 1987-1988, CCH American Stock Exchange Guide 158-160 (1987) (71 out of 116 “public” arbitrators are lawyers). The uniform opposition of investors to compelled arbitration and the overwhelming support of the securities industry for the process suggest that there must be some truth to the investors’ belief that the securities industry has an advantage in a forum under its own control. See N. Y. Times, Mar. 29, 1987, section 3, p. 8, col. 1 (statement of Sheldon H. Eisen, Chairman, American Bar Association Task Force on Securities Arbitration: “The houses basically like the present system because they own the stacked deck”).20 More surprising than the Court’s acceptance of the present adequacy of arbitration for the resolution of securities claims is its confidence in the Commission’s oversight of the arbitration procedures of the SROs to ensure this adequacy. Such confidence amounts to a wholesale acceptance of the Commission’s present position that this oversight undermines the force of Wilko and that arbitration therefore should be compelled because the Commission has supervisory authority 20 Commentators have argued that more public participation in the SRO arbitration procedures is needed to give investors the impression that they are not in a forum biased in favor of the securities industry. See, e. g., Katsoris 313. The amici in support of petitioners and some commentators argue that the statistics concerning the results of arbitration show that the process is not weighted in favor of the securities industry. See Brief for Securities Industry Association, Inc., et al. as Amici Curiae 9; Brief for American Arbitration Association as Amicus Curiae 17; Fletcher 452. Such statistics, however, do not indicate the damages received by customers in relation to the damages to which they believed they were entitled. It is possible for an investor to “prevail” in arbitration while recovering a sum considerably less than the damages he actually incurred. 262 OCTOBER TERM, 1986 Opinion of Blackmun, J. 482 U. S. over the SROs’ arbitration procedures. The Court, however, fails to acknowledge that, until it filed an amicus brief in this case, the Commission consistently took the position that § 10(b) claims, like those under § 12(2), should not be sent to arbitration, that predispute arbitration agreements, where the investor was not advised of his right to a judicial forum, were misleading, and that the very regulatory oversight upon which the Commission now relies could not alone make securities-industry arbitration adequate.21 It is most questionable, then, whether the Commission’s recently adopted position is entitled to the deference that the Court accords it. The Court is swayed by the power given to the Commission by the 1975 amendments to the Exchange Act in order to permit the Commission to oversee the rules and procedures of the SROs, including those dealing with arbitration. See ante, at 233-234. Subsequent to the passage of these amendments, however, the Commission has taken the consistent position that predispute arbitration agreements, 21 The Court accepts the argument, put forward now by the Commission, see Brief 18, n. 13, that its prior position was based solely on the Wilko decision and the decisions in the Courts of Appeals extending Wilko to § 10(b) claims, and not on its independent assessment of the adequacy of arbitration or its awareness of the possible abuses to which predispute agreements to arbitrate were subject. See ante, at 234, n. 3. Suffice it to say that the Commission’s opposition to predispute agreements that might mislead an investor into giving up statutory rights even predates Wilko. In a release discussing proposed Rule 15c2-2, which prohibited the use of clauses purporting to bind investors to arbitrate future disputes, the Commission observed that, at least since 1951, it had opposed provisions in agreements whose result or purpose was to have investors give up rights or remedies under the securities Acts. See Disclosure Regarding Recourse to the Federal Courts Notwithstanding Arbitration Clauses in Broker-Dealer Customer Agreements, SEC Exchange Act Rei. No. 19813 (May 26, 1983), [1982-1983 Transfer Binder] CCH Fed. Sec. L. Rep. 183,356, p. 85,967, n. 6. SHEARSON/AMERICAN EXPRESS INC. v. McMAHON 263 220 Opinion of Blackmun, J. which did not disclose to an investor that he has a right to a judicial forum, were misleading and possibly actionable under the securities laws.22 The Commission remained dissatis 22 The Commission, in a release issued in 1979, explained its opposition to predispute arbitration agreements: “It is the Commission’s view that it is misleading to customers to require execution of any customer agreement which does not provide adequate disclosure about the meaning and effect of its terms, particularly any provision which might lead a customer to believe that he or she has waived prospectively rights under the federal securities laws, rules thereunder, or certain rules of any self-regulatory organization. Customers should be made aware prior to signing an agreement containing an arbitration clause that such a prior agreement does not bar a cause of action arising under the federal securities laws. If a broker-dealer customer’s agreement contains an arbitration clause, it must be consistent with current judicial decisions regarding the application of the federal securities laws to predispute arbitration agreements. “The Commission is especially concerned that arbitration clauses continue to be part of form agreements widely used by broker-dealers, despite the number of cases in which these clauses have been held to be unenforceable in whole or in part. Requiring the signing of an arbitration agreement without adequate disclosure as to its meaning and effect violates standards of fair dealing with customers and constitutes conduct that is inconsistent with just and equitable principles of trade. In addition, it may raise serious questions of compliance with the anti-fraud provisions of the federal securities laws.” Broker-Dealers Concerning Clauses in Customer Agreements Which Provide for Arbitration of Future Disputes, SEC Exchange Act Rei. No. 15984 (July 2, 1979), 44 Fed. Reg. 40462, 40464 (footnotes omitted). As the quoted material suggests, the Commission was aware of the court cases concerning such arbitration agreements. In the release, the Commission discussed at length this Court’s Wilko decision and cases in which courts had extended it to § 10(b) claims. See 44 Fed. Reg., at 40463. The thrust of the release is that the Commission not only accepted the case law but also, for its own reasons, thought that the arbitration agreements in the predispute context were inappropriate and misleading. See, e. g., Implementation of an Investor Dispute Resolution System, SEC Exchange Act Rei. No. 13470 (Apr. 26, 1977), [1977-1978 Transfer Binder] CCH Fed. Sec. L. Rep. 1181,136, p. 87,907 (“Customer agreements to arbitrate, at the instance of a firm, in margin agreements or elsewhere, should be pro 264 OCTOBER TERM, 1986 Opinion of Blackmun, J. 482 U. S. fied with the continued use of these arbitration agreements and eventually it proposed a rule to prohibit them, explaining that such a prohibition was not inconsistent with its support of arbitration for resolving securities disputes, particularly existing ones. See Disclosure Regarding Recourse to the Federal Courts Notwithstanding Arbitration Clauses in Broker-Dealer Customer Agreements, SEC Exchange Act Rei. No. 19813 (May 26, 1983), [1982-1983 Transfer Binder] CCH Fed. Sec. L. Rep. If 83,356, p. 85,967. While emphasizing the Court’s Wilko decision as a basis for its proposed rule, the Commission noted that its proposal also was in line with its own understanding of the problems with such agreements and with the “[c]ongressional determination that public investors should also have available the special protection of the federal courts for resolution of disputes arising under the federal securities laws.” Id., at p. 85,968. Although the rule met with some opposition,23 it was adopted and remains in force todays hibited”). The Commission acknowledges that in 1975 it even filed an amicus brief in Ayres n. Merrill Lynch, Pierce, Fenner & Smith, Inc., 538 F. 2d 532 (CA3), cert, denied, 429 U. S. 1010 (1976), in which it supported the extension of Wilko to § 10(b) claims. See Brief 18, n. 13. 23 The Commission rejected commentators’ suggestions that the refusal to compel arbitration of securities disputes on the basis of the predispute agreements “‘rests on questionable legal ground.’” See Recourse to the Courts Notwithstanding Arbitration Clauses in Broker-Dealer Customer Agreements, SEC Exchange Act Rei. No. 20397 (Nov. 18, 1983), [1983— 1984 Transfer Binder] CCH Fed. Sec. L. Rep. 5183,452, p. 86,357, n. 6, quoting comments of the Securities Industry Association. 24 This rule provides in pertinent part: “It shall be a fraudulent, manipulative or deceptive act or practice for a broker or dealer to enter into an agreement with any public customer which purports to bind the customer to the arbitration of future disputes between them arising under the Federal securities laws, or to have in effect such an agreement, pursuant to which it effects transactions with or for a customer.” Rule 15c2-2, 17 CFR §240.15c2-2(a) (1986). SHEARSON/AMERICAN EXPRESS INC. v. McMAHON 265 220 Opinion of Blackmun, J. Moreover, the Commission’s own description of its enforcement capabilities contradicts its position that its general overview of SRO rules and procedures can make arbitration adequate for resolving securities claims. The Commission does not pretend that its oversight consists of anything other than a general review of SRO rules and the ability to require that an SRO adopt or delete a particular rule. It does not contend that its “sweeping authority,” Brief 16, includes a review of specific arbitration proceedings. It thus neither polices nor monitors the results of these arbitrations for possible misapplications of securities laws or for indications of how investors fare in these proceedings. Given, in fact, the present constraints on the Commission’s resources in this time of market expansion, see General Accounting Office, Report to the Chairman, Subcommittee on Telecommunications, Consumer Protection, and Finance of the House Committee on Energy and Commerce: Securities Regulation—Securities and Exchange Commission Oversight of Self-Regulation 60 (1986) (Report), it is doubtful whether the Commission could undertake to conduct any such review.25 Finally, the Court’s complacent acceptance of the Commission’s oversight is alarming when almost every day brings another example of illegality on Wall Street. See, e. g., N. Y. Times, Jan. 2, 1987, p. B6, col. 3. Many of the abuses re 25 Even those who would agree with the Commission that its general oversight of SRO arbitration procedures has bettered the adequacy of arbitration recognize that improvements in this oversight still are needed. For example, commentators have suggested that the Commission should revise the Uniform Code of Arbitration in order to ensure that predispute arbitration agreements are displayed prominently, that the reference to a person drawn from “outside the securities industry” be more specifically defined, and that arbitrators be required to give a more detailed statement of their reasoning. See Brown, Shell, & Tyson 34-36. Congress could give to the Commission specific rulemaking authority in the area of arbitration with the goal of preventing abuses in the process that have surfaced in recent years. Id., at 34. 266 OCTOBER TERM, 1986 Opinion of Blackmun, J. 482 U. S. cently brought to light, it is true, do not deal with the question of the adequacy of SRO arbitration. They, however, do suggest that the industry’s self-regulation, of which the SRO arbitration is a part, is not functioning acceptably. See Report 63. Moreover, these abuses have highlighted the difficulty experienced by the Commission, at a time of growth in the securities market and a decrease in the Commission’s staff, see id., at 60-61, to carry out its oversight task. Such inadequacies on the part of the Commission strike at the very heart of the reasoning of the Court, which is content to accept the soothing assurances of the Commission without examining the reality behind them. Indeed, while the amici cite the number of arbitrations of securities disputes as a sign of the success of this process in the industry, see Brief for Securities Industry Association, Inc., et al. as Amici Curiae 10-11, these statistics have a more portentous meaning. In this era of deregulation, the growth in complaints about the securities industry, many of which find their way to arbitration, parallels the increase in securities violations and suggests a market not adequately controlled by the SROs. See General Accounting Office, Report to the Chairman, Subcommittee on Oversight and Investigation of the House Committee on Energy and Commerce: Statistics on SEC’s Enforcement Program 3-4 (1985). In such a time, one would expect more, not less, judicial involvement in resolution of securities disputes. Ill There is, fortunately, a remedy for investors. In part as a result of the Commission’s position in this case, Congress has begun to look into the adequacy of the self-regulatory arbitration and the Commission’s oversight of the SROs. In a letter dated February 11, 1987, Representative Dingell, Chairman of the House Subcommittee on Oversight and Investigations, notified the Chairman of the Commission that the Subcommittee is “conducting an inquiry into the adequacy of the current self-regulatory system and the Commis- SHEARSON/AMERICAN EXPRESS INC. v. McMAHON 267 220 Opinion of Blackmun, J. sion’s oversight thereof in connection with complaints against broker-dealers for securities-law violations.” Letter, p. 1, enclosed with Letter from Theodore G. Eppenstein, counsel for respondents, to Joseph F. Spaniol, Jr., Clerk of this Court (Mar. 2, 1987). Representative Dingell noted that his Subcommittee was “particularly concerned about increasing numbers of complaints in connection with churning and violations of suitability requirements, as well as complaints that arbitration procedures are rife with conflicts of interest (since the arbitrators are peers of the brokerage firm being sued) and are inadequate to enforce the statutory rights of customers against broker-dealers.” Ibid. To justify this inquiry, he cited several well-publicized examples of abuse of investors by securities-industry personnel and a General Accounting Office report on the increase in securities-law violations by brokers that went undetected by the SROs. In concluding the letter, Representative Dingell expressed his surprise at the Commission’s position in the present case. In his view, that position was at odds with the one the Commission consistently had taken before the Subcommittee, which stressed the limitations on the Commission’s authority over the SROs in general, and over arbitrations in particular. Id., at 3. Thus, there is hope that Congress will give investors the relief that the Court denies them today. In the meantime, the Court leaves lower courts with some authority, albeit limited, to protect investors before Congress acts. Courts should take seriously their duty to review the results of arbitration to the extent possible under the Arbitration Act. As we explained in Mitsubishi Motors Corp. v. Soler Chrysler-Plymouth, Inc., “courts should remain attuned to well-supported claims that the agreement to arbitrate resulted from the sort of fraud or overwhelming economic power that would provide grounds ‘for the revocation of any contract.’” 473 U. S., at 627, quoting 9 U. S. C. §2. Indeed, in light of today’s decision compelling the enforcement of predispute arbitration agreements, it is likely 268 OCTOBER TERM, 1986 Opinion of Stevens, J. 482 U. S. that investors will be inclined, more than ever, to bring complaints to federal courts that arbitrators were partial or acted in “manifest disregard” of the securities laws. See Brown, Shell, & Tyson 36. It is thus ironic that the Court’s decision, no doubt animated by its desire to rid the federal courts of these suits, actually may increase litigation about arbitration. I therefore respectfully dissent in part. Justice Stevens, concurring in part and dissenting in part. Gaps in the law must, of course, be filled by judicial construction. But after a statute has been construed, either by this Court or by a consistent course of decision by other federal judges and agencies, it acquires a meaning that should be as clear as if the judicial gloss had been drafted by the Congress itself. This position reflects both respect for Congress’ role, see Boys Market, Inc. n. Retail Clerks, 398 U. S. 235, 257-258 (1970) (Black, J., dissenting), and the compelling need to preserve the courts’ limited resources, see B. Cardozo, The Nature of the Judicial Process 149 (1921). During the 32 years immediately following this Court’s decision in Wilko n. Swan, 346 U. S. 427 (1953), each of the eight Circuits that addressed the issue concluded that the holding of Wilko was fully applicable to claims arising under the Securities Exchange Act of 1934? See ante, at 248, n. 6 (opinion of Blackmun, J.). This longstanding interpretation2 creates a strong presumption, in my view, that any mis- 1 It was only after Justice White’s concurrence in Dean Witter Reynolds Inc. v. Byrd, 470 U. S. 213, 224 (1985), indicating his “substantial doubt” about Wilko's applicability to the 1934 Act, that two Circuits held it to be inapplicable. See ante, at 249, n. 8 (opinion of Blackmun, J.). 2 Because I have never been convinced that the antifraud provisions of the federal securities laws were intended to apply to private transactions negotiated between fully informed parties of relatively equal bargaining strength, see Landreth Timber Co. v. Landreth, 471 U. S. 681, 697 (1985) (Stevens, J., dissenting), I was not at all surprised by the Court’s decision in Scherk v. Alberto-Culver Co., 417 U. S. 506 (1974), refusing to apply the Wilko rule to such a case. See Alberto-Culver Co. v. Scherk, 484 F. 2d SHEARSON/AMERICAN EXPRESS INC. v. McMAHON 269 220 Opinion of Stevens, J. take that the courts may have made in interpreting the statute is best remedied by the Legislative, not the Judicial, Branch. The history set forth in Part I of Justice Blackmun’s opinion adds special force to that presumption in this case. For this reason, I respectfully dissent from the portion of the Court’s judgment that holds Wilko inapplicable to the 1934 Act. Like Justice Blackmun, however, I join Parts I, II, and IV of the Court’s opinion. 611, 615-620 (CA7 1973) (Stevens, J., dissenting). As Justice Blackmun has demonstrated, that refusal was not predicated on any perceived difference between the 1933 Act and the 1934 Act, and it is thus fair to state that the decision the Court announces today changes a settled construction of the relevant statute. 270 OCTOBER TERM, 1986 Syllabus 482 U. S. INTERSTATE COMMERCE COMMISSION v. BROTHERHOOD OF LOCOMOTIVE ENGINEERS et al. CERTIORARI TO THE UNITED STATES COURT OF APPEALS FOR THE DISTRICT OF COLUMBIA CIRCUIT No. 85-792. Argued November 10, 1986-Decided June 8, 1987* In October 1982, petitioner Interstate Commerce Commission (ICC or Commission) issued an order, which, inter alia, granted petitioner Missouri-Kansas-Texas Railroad Co. and another railroad the right to conduct operations using the tracks of a third, newly consolidated carrier. In April 1983, respondent Brotherhood of Locomotive Engineers (BLE) filed a “Petition for Clarification” asking the ICC to declare that the earlier order did not authorize the tenant railroads to use their own crews on routes they had not previously served. In an order served on May 18, 1983, the ICC denied the petition, ruling that its prior decision did not require clarification since the tenant railroads’ trackage rights applications had proposed that they use their own crews and the Commission’s approval of the applications authorized such operations. BLE and respondent United Transportation Union then filed timely petitions for reconsideration of the May 18 order, contending, inter alia, that the tenant railroads’ crewing procedures violated employee protections that had been included in the original order. In an order served on October 25, 1983, the ICC denied these petitions, responding in detail to the unions’ contentions. On respondent unions’ petitions for review, the Court of Appeals vacated the ICC orders of May 18 and October 25, rejecting the threshold claim that its review was time barred, and ruling for the unions on the merits. Held: 1. Although respondent unions’ petitions for Court of Appeals review of the ICC’s October 25, 1983, order were timely filed, they should have been dismissed since the order itself, whereby the Commission refused to reconsider its May 18, 1983, order refusing to clarify its prior approval order, is unreviewable. Pp. 277-284. (a) Since an ICC order in a rail proceeding is “final on the date on which it is served,” the unions’ petitions for review of the October 25 order, which petitions were filed with the Court of Appeals on December *Together with No. 85-793, Missouri-Kansas-Texas Railroad Co. v. Brotherhood of Locomotive Engineers et al., also on certiorari to the same court. ICC v. LOCOMOTIVE ENGINEERS 271 270 Syllabus 16 and 23, 1983, were timely under the Hobbs Act, which requires that any person aggrieved by an ICC final order file a review petition within 60 days after the order’s entry. P. 277. (b) Only when a petition to reopen and reconsider an agency order alleges new evidence or changed circumstances is the agency’s refusal to reopen subject to judicial review, and then only as to whether such refusal was arbitrary, capricious, or an abuse of discretion. Where, as here, the petition to reopen is based not on new data but only on alleged “material error,” the agency’s denial of the petition is not subject to judicial review. Such an appeal would place before the court precisely the same substance that could have been brought before it by timely appeal from the original order. Review in that circumstance would serve no purpose if the reconsideration petition were filed within the agency’s discretionary review period (and within the period for judicial review of the original order), since the petition would toll the judicial review period and thereby preserve a direct appeal of the original order until reconsideration was denied. On the other hand, if the reconsideration petition were untimely filed, judicial review would serve only the peculiar purpose of extending indefinitely the time within which seriously mistaken agency orders could be judicially overturned, since it would have to be shown not only that the original agency decision was unlawful, but that it was so unlawful that refusal to reconsider it is an abuse of discretion. It is irrelevant that the ICC’s order discussed respondent unions’ substantive claims at length, since it is the Commission’s formal action, rather than its discussion, that is dispositive as to whether reconsideration was in fact granted or denied. Pp. 277-281. (c) In addition to being implicit in the Hobbs Act’s 60-day limitation on seeking judicial review, nonreviewability of refusals to reopen for material error is established by 5 U. S. C. § 701(a)(2), which precludes judicial review of action “committed to agency discretion by law.” Refusals to reopen have traditionally been reviewed only when new evidence or changed circumstances are alleged, and it is impossible to devise an adequate standard of review when only material error is alleged. Application of the ordinary standards for reviewing errors of law and fact would entirely frustrate the Hobbs Act’s time limitation, and the only other alternative-some form of “clearly erroneous” standard—would produce the strange result that only really bad mistakes (whatever that term might mean when considerable discretion is already afforded to agencies) would escape the time limitation. Nor does agency action become reviewable merely because the agency gives a “reviewable” reason— i- e., a reason that courts are well qualified to consider—for that action. Pp. 281-284. 272 OCTOBER TERM, 1986 Syllabus 482 U. S. 2. Although the petition for Court of Appeals review of the ICC’s May 18, 1983, order was timely filed, it should have been dismissed since the order itself, which denied BLE’s petition for clarification of the earlier trackage rights approval order, is not appealable. Pp. 284-287. (a) Even though the petition for judicial review was filed more than 60 days after the May 18 order was served, it was nonetheless effective because respondent unions’ timely petitions for administrative reconsideration stayed the running of the Hobbs Act’s filing period until the ICC denied reconsideration. Although a contrary conclusion is suggested by the language of 49 U. S. C. § 10327(i), which provides that, notwithstanding the statutory provision authorizing ICC reconsideration of its orders, a Commission action is final and can be appealed on the day it is served, in view of prior constructions of similar language in 5 U. S. C. § 704 that language must be construed merely to relieve parties from the requirement of petitioning for reconsideration before seeking judicial review, but not to prevent reconsideration petitions that are actually filed from rendering orders under reconsideration nonfinal. Pp. 284-285. (b) If BLE’s petition is treated as a genuine “Petition for Clarification”— i. e., as seeking only a specification, one way or the other, of what the original order meant—its denial is unappealable because BLE could not be considered “aggrieved” under the Hobbs Act by the ICC’s unambiguous explanation in the order of what the earlier order meant. If, as is more likely, BLE’s petition was actually treated by the parties as a petition to reopen based on what BLE felt to be a serious error of law, the petition’s denial is unreviewable under the Court’s analysis, supra, since BLE put forth no new evidence or changed circumstances. There can be no exception to the rule of nonreviewability even where an order is ambiguous and thereby causes a party to think that its interests are not infringed, since the Hobbs Act’s time limit would then be held hostage to ever-present ambiguities. The remedy for such ambiguity is to petition the Commission for reconsideration within the 60-day period, thereby enabling judicial review to be pursued (if ICC resolution of the ambiguity is adverse) after disposition of that petition. Pp. 285-286. 245 U. S. App. D. C. 311, 761 F. 2d 714, vacated and remanded. Scalia, J., delivered the opinion of the Court, in which Rehnquist, C. J., and White, Powell, and O’Connor, JJ., joined. Stevens, J., filed an opinion concurring in the judgment, in which Brennan, Marshall, and Blackmun, JJ., joined, post, p. 287. Henri F. Rush argued the cause for petitioner in No. 85-792. With him on the briefs were Solicitor General Fried, Deputy Solicitor General Cohen, Robert S. Burk, and ICC v. LOCOMOTIVE ENGINEERS 273 270 Opinion of the Court Sidney L. Strickland, Jr. Joseph L. Manson III, argued the cause for petitioner in No. 85-793. With him on the briefs was Michael E. Roper. Harold A. Ross argued the cause and filed a brief for respondent Brotherhood of Locomotive Engineers in both cases. John O’B. Clarke, Jr., argued the cause for respondent United Transportation Union in both cases. With him on the brief was Robert L. Hart. Charles A. Miller, Gregg H. Levy, and Mark B. Goodwin filed a brief for respondents Union Pacific Railroad Co. et al.t Justice Scalia delivered the opinion of the Court. On September 15, 1980, Union Pacific Railroad Co. (UP) and Missouri Pacific Railroad Co. (MP) and their respective corporate parents filed a joint application with the Interstate Commerce Commission (ICC or Commission) seeking permission for UP to acquire control of MP. The same day, a similar but separate application was jointly filed by UP and the Western Pacific Railroad Co. (WP). In a consolidated proceeding, the control applications were opposed by a number of labor organizations, including respondents Brotherhood of Locomotive Engineers (BLE) and United Transportation Union (UTU), as well as several competing railroads, including petitioner Missouri-Kansas-Texas Railroad Co. (MKT) and the Denver and Rio Grande Western Railroad Co. (DRGW). MKT and DRGW, in addition to opposing the mergers, filed responsive applications seeking the right to conduct operations using the track of the new consolidated carrier in the event that the control applications were approved. MKT’s request for trackage rights specified that “MKT, with its own employees, and at its sole cost and expense, shall operate its engines, cars and trains on and along Joint Track.” Proposed Trackage Rights Agreement § 5, Fi- Richard T. Conway, William F. Sheehan, and Kenneth P. Kolson filed a brief for the Association of American Railroads et al. as amici curiae urging reversal. 274 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. nance Docket No. 30,000 (Sub.-No. 25). DRGW’s application indicated that it “may, at its option, elect to employ its own crews for the movement of its trains, locomotives and cars to points on or over the Joint Track.” Proposed Trackage Rights Agreement § 6(c)(3), Finance Docket No. 30,000 (Sub.-No. 18). On October 20, 1982, the ICC approved UP’s control acquisitions and granted MKT’s application for trackage rights over 200 miles of MP and UP track in four States and DRGW’s application for rights over 619 miles of MP track between Pueblo and Kansas City. See Union Pacific Corp., Pacific Rail System, Inc. & Union Pacific R. Co.—Control— Missouri Pacific Corp. & Missouri Pacific R. Co., 366 I. C. C. 459 (1982), aff’d sub nom. Southern Pacific Transportation Co. n. ICC, 237 U. S. App. D. C. 99, 736 F. 2d 708 (1984), cert, denied, 469 U. S. 1208 (1985). The approved trackage rights were to become effective “immediately upon consummation of the consolidations.” 366 I. C. C., at 590. It is the Commission’s standard practice, in pursuit of its statutory responsibility to shield railroad employees from dislocations resulting from actions that it approves, see 49 U. S. C. § 11347, to impose on trackage rights transactions a set of employee protections known as the “NW-BN-Mendocino” conditions. See Norfolk and Western R. Co.— Trackage Rights—Burlington Northern, Inc., 354 I. C. C. 605 (1978), modified, Mendocino Coast R. Co.—Lease and Operate—California Western R. Co., 360 I. C. C. 653 (1980), aff’d sub nom. Railway Labor Executives’ Assn. v. United States, 219 U. S. App. D. C. 23, 675 F. 2d 1248 (1982). These provide, inter alia, for “the selection of forces from all employees involved,” 354 I. C. C., at 610, in transactions involving the dismissal or displacement of employees, and for retention of “[t]he rates of pay, rules, working conditions and all collective bargaining and other rights, privileges and benefits . . . unless changed by future collective bargaining agreements or applicable statutes.” Ibid. The ICC’s Octo- ICC v. LOCOMOTIVE ENGINEERS 275 270 Opinion of the Court ber 20, 1982, order indicated, without discussion, that approval of the trackage rights applications was “subject to the imposition of employee protective conditions to the extent specified in [NW-BN and Mendocino].” 366 I. C. C., at 654. See also id., at 471, 622. The control transactions among UP, MP, and WP were consummated on December 22, 1982, at which point the grants of trackage rights also became effective. MKT commenced its operations, using its own crews, on or about January 6, 1983; and DRGW shortly thereafter entered into an agreement with MP providing “for using MP crews on [DRGW] trains for a temporary, interim period, after which [DRGW] will operate the trains with [its] own crews.” App. in Nos. 83-2290 and 83-2317 (CADO), p. 6. Although numerous parties, including BLE, had petitioned for review of the Commission’s October 20, 1982, order (which was affirmed in most respects some 18 months later, see Southern Pacific Transportation Co. v. ICC, supra), no question concerning the crewing of MKT or DRGW trains was raised at that time. However, on April 4, 1983, BLE filed with the Commission a “Petition for Clarification,” contending that the Commission had no jurisdiction to, and as a matter of consistent practice did not, inject itself into labor matters such as crew selection, and asking the Commission to declare that its October 20, 1982, order did not have the intent or effect of authorizing the tenant carriers to use their own crews on routes that they had not previously served.1 In a brief order served May 18, 1983, the Commission denied the petition, ruling that its prior decision “does not require clarification.” App. to Pet. for Cert, in No. 85-793, p. A38. The 1 Ordinarily, a petition for relief from a final order is denominated a “petition to reopen.” See 49 CFR §§ 1115.3(a), 1115.4 (1986). A Commission rule, however, permits parties to “see[k] relief not provided for in any other rule,” 49 CFR § 1117.1 (1986), including clarification of the terms of a prior order. See Burlington Northern Inc. v. United States, 459 U. S. 131, 136 (1982). 276 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. tenant railroads, it said, had proposed to use their own crews in their trackage rights applications, and “our approval of the applications authorizes such operations.” Ibid. Within the period prescribed by Commission rules for filing petitions for administrative review, see 49 CFR § 1115.3(e) (1986), both BLE and UTU sought “reconsideration” of the Commission’s denial. In addition to repeating BLE’s earlier arguments, the unions contended that the tenant railroads’ crewing procedures constituted a unilateral change in working conditions forbidden by the NW-BN-Mendocino labor protective conditions, by the Railway Labor Act, 45 U. S. C. § 151 et seq. (RLA), and by collective-bargaining agreements, and that the Commission had made no findings that would justify exempting the trackage rights transactions from applicable labor laws. In a lengthy order served on October 25, 1983, responding in some detail to all of the major contentions, the Commission denied the petitions. In particular, the Commission emphasized its reliance on 49 U. S. C. § 11341(a), which provides that a carrier participating in a consolidation approved by the Commission “is exempt from the antitrust laws and from all other law ... as necessary to let that person carry out the transaction . . . .” The Commission concluded that the exemption provided by this section extends to the RLA and is self-executing, requiring no findings by the Commission to make it effective. On December 16, 1983, BLE petitioned for judicial review of the May 18, 1983, and October 25, 1983, orders; UTU petitioned for review of the latter order on December 23, 1983. The cases were consolidated, and the United States Court of Appeals for the District of Columbia Circuit vacated both orders. 245 U. S. App. D. C. 311, 761 F. 2d 714 (1985). The court rejected the threshold claim that the appeals were time barred, and concluded on the merits that if the ICC intended to exempt the railroads from the requirements of the RLA, it was required to explain, as it had not done, why that exemp- ICC v. LOCOMOTIVE ENGINEERS 277 270 Opinion of the Court tion was necessary to effectuate the transactions it approved. The dissent disagreed on both counts. MKT and the Commission filed petitions for certiorari on the question of the proper construction of § 11341(a), which we granted and consolidated for argument. See 475 U. S. 1081 (1986). We now conclude that the petitions for review must be dismissed. I The petitions for review and the Court of Appeals’ order encompass both the May 18, 1983, order refusing to clarify the Commission’s prior approval order, and the October 25, 1983, order refusing to reconsider that refusal to clarify. We consider first the appeal of the latter order. With certain exceptions not relevant here, see 28 U. S. C. § 1336(b), judicial review of final orders of the ICC is governed by the Hobbs Act, 28 U. S. C. §2341 et seq., which provides that any party aggrieved by a “final order” of the Commission “may, within 60 days after its entry, file a petition to review the order in the court of appeals wherein venue lies.” §2344. A Commission order in a rail proceeding is “final on the date on which it is served.” 49 U. S. C. § 10327(i). The Commission’s order refusing to reconsider its refusal to clarify thus became final on October 25, 1983, and the unions’ petitions for review—filed on December 16, 1983, and December 23, 1983 —were therefore timely for purposes of reviewing that order (though they obviously were not timely for purposes of reviewing the original order of October 20, 1982). However, although the timeliness requirements of the Hobbs Act were satisfied, the order from which the unions have appealed is unreviewable. The Commission’s authority to reopen and reconsider its prior actions stems from 49 U. S. C. § 10327(g), which provides: “The Commission may, at any time on its own initiative because of material error, new evidence, or substantially changed circumstances — 278 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. “(A) reopen a proceeding; “(B) grant rehearing, reargument, or reconsideration of an action of the Commission; and “(C) change an action of the Commission. “An interested party may petition to reopen and reconsider an action of the Commission under this paragraph under regulations of the Commission.” When the Commission reopens a proceeding for any reason and, after reconsideration, issues a new and final order setting forth the rights and obligations of the parties, that order—even if it merely reaffirms the rights and obligations set forth in the original order—is reviewable on its merits. See, e. g., United States v. Seatrain Lines, Inc., 329 U. S. 424 (1947). Where, however, the Commission refuses to reopen a proceeding, what is reviewable is merely the lawfulness of the refusal. Absent some provision of law requiring a reopening (which is not asserted to exist here), the basis for challenge must be that the refusal to reopen was “arbitrary, capricious, [or] an abuse of discretion.” 5 U. S. C. §706 (2)(A). We have said that overturning the refusal to reopen requires “a showing of the clearest abuse of discretion,” United States v. Pierce Auto Freight Lines, Inc., 327 U. S. 515, 534-535 (1946), and we have actually reversed the ICC only once, see Atchison, T. & S. F. R. Co. v. United States, 284 U. S. 248 (1932), in a decision that was “promptly restricted ... to its special facts, . . . and . . . stands virtually alone.” ICC v. Jersey City, 322 U. S. 503, 515 (1944). See also Bowman Transportation, Inc. n. Arkansas-Best Freight System, Inc., 419 U. S. 281, 295-296 (1974). More importantly for present purposes, all of our cases entertaining review of a refusal to reopen appear to have involved petitions alleging “new evidence” or “changed circumstances” that rendered the agency’s original order inappropriate. See id., at 295, and cases cited therein; Jersey City, supra, at 514-518, and cases cited therein. We know of no case in which we have reviewed the denial of a petition to reopen ICC v. LOCOMOTIVE ENGINEERS 279 270 Opinion of the Court based upon no more than “material error” in the original agency decision. There is good reason for distinguishing between the two. If review of denial to reopen for new evidence or changed circumstances is unavailable, the petitioner will have been deprived of all opportunity for judicial consideration-even on a “clearest abuse of discretion” basis—of facts which, through no fault of his own, the original proceeding did not contain. By contrast, where no new data but only “material error” has been put forward as the basis for reopening, an appeal places before the courts precisely the same substance that could have been brought there by appeal from the original order—but asks them to review it on the strange, one-step-removed basis of whether the agency decision is not only unlawful, but so unlawful that the refusal to reconsider it is an abuse of discretion. Such an appeal serves no purpose whatever where a petition for reconsideration has been filed within a discretionary review period specifically provided by the agency2 (and within the period allotted for judicial review of the original order), since in that situation the petition tolls the period for judicial review of the original order, which can therefore be appealed to the courts directly after the petition for reconsideration is denied. And where the petition is filed outside that period (and outside the 2 The ICC’s regulations, for example, provide as follows: “(a) A discretionary appeal is permitted. It will be designated a ‘petition for administrative review,’ except that, when it is related to an action of the entire Commission in the first instance, it must be designated a ‘petition to reopen.’ “(b) The petition will be granted only upon a showing of one or more of the following points: “(3) The prior action was taken by the entire Commission in the first instance, and involves material error. “(e) Petitions must be filed within 20 days after the service of the action or within any further period (not to exceed 20 days) as a division or the Commission may authorize.” 49 CFR § 1115.3 (1986). 280 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. period for judicial review of the original order) judicial review would serve only the peculiar purpose of extending indefinitely the time within which seriously mistaken agency orders can be judicially overturned. That is to say, the Hobbs Act’s 60-day limitation provision would effectively be subjected to a proviso that reads: “Provided, however, that if the agency error is so egregious that refusal to correct it would be an abuse of discretion, judicial review may be sought at any time.” For these reasons, we agree with the conclusion reached in an earlier case by the Court of Appeals that, where a party petitions an agency for reconsideration on the ground of “material error,” i. e., on the same record that was before the agency when it rendered its original decision, “an order which merely denies rehearing of. . . [the prior] order is not itself reviewable.” Microwave Communications, Inc. v. FCC, 169 U. S. App. D. C. 154, 156, n. 7, 515 F. 2d 385, 387, n. 7 (1974). See also SEC v. Louisiana Public Service Common, 353 U. S. 368, 371-372 (1957); National Bank of Davis v. Comptroller of Currency, 233 U. S. App. D. C. 284, 285, and n. 3, 725 F. 2d 1390, 1391, and n. 3 (1984); 5 U. S. C. § 701(a)(2). This rule is familiar from other contexts. If a judicial panel or an en banc court denies rehearing, no one supposes that that denial, as opposed to the panel opinion, is an appealable action (though the filing of a timely rehearing petition, like the filing of a timely petition for agency reconsideration, extends the time for appealing from the original decision). It is irrelevant that the Commission’s order refusing reconsideration discussed the merits of the unions’ claims at length. Where the Commission’s formal disposition is to deny reconsideration, and where it makes no alteration in the underlying order, we will not undertake an inquiry into whether reconsideration “in fact” occurred. In a sense, of course, it always occurs, since one cannot intelligently rule upon a petition to reconsider without reflecting upon, among ICC v. LOCOMOTIVE ENGINEERS 281 270 Opinion of the Court other things, whether clear error was shown. It would hardly be sensible to say that the Commission can genuinely deny reconsideration only when it gives the matter no thought; nor to say that the character of its action (as grant or denial) depends upon whether it chooses to disclose its reasoning. Rather, it is the Commission’s formal action, rather than its discussion, that is dispositive. Accordingly, the petitions for review of the Commission’s October 25, 1983, order refusing to reconsider its May 18, 1983, refusal to clarify should have been dismissed. That portion of the concurrence which deals with the issue of jurisdiction (Part I) consists largely of the citation and discussion of numerous cases affording judicial review of agency refusals to reopen. In the third from last paragraph, however, one finds the acknowledgment that all these cases involved refusals “based upon new evidence or changed circumstances” rather than upon “material error,” post, at 293-294, and are therefore fully in accord with the principle set forth in the present opinion. The only point of dispute between this opinion and the concurrence is whether separate treatment of refusals to reopen based on material error has some basis in statute or is rather, as the concurrence would have it, “a pure creature of judicial invention.” Post, at 294. Even if our search for statutory authorization were limited to the text of the Hobbs Act, it seems to us not inventiveness but the most plebeian statutory construction to find implicit in the 60-day limit upon judicial review a prohibition against the agency’s permitting, or a litigant’s achieving, perpetual availability of review by the mere device of filing a suggestion that the agency has made a mistake and should consider the matter again. Substantial disregard of the Hobbs Act is effected, not by our opinion, but by what the concurrence delivers (after having rejected our views) in Part II of its opinion: on-the-merits review of an agency decision of law rendered 14 months before the petition for review was filed, using the same standard of review that would have been 282 482 U. S. OCTOBER TERM, 1986 Opinion of the Court applied had appeal been filed within the congressionally prescribed 60-day period. Statutory authority for preventing this untoward result need not be sought solely in the Hobbs Act, however. While the Hobbs Act specifies the form of proceeding for judicial review of ICC orders, see 5 U. S. C. § 703, it is the Administrative Procedure Act (APA) that codifies the nature and attributes of judicial review, including the traditional principle of its unavailability “to the extent that. . . agency action is committed to agency discretion by law.” 5 U. S. C. §701 (a)(2). We have recently had occasion to apply this limitation to the general grant of jurisdiction contained in 28 U. S. C. § 1331, see Heckler n. Chaney, 470 U. S. 821 (1985); it applies to the general grant of jurisdiction of the Hobbs Act as well. In Chaney we found that the type of agency decision in question “has traditionally been ‘committed to agency discretion,’ and . . . that the Congress enacting the APA did not intend to alter that tradition.” Id., at 832. As discussed above, we perceive that a similar tradition of nonreviewability exists with regard to refusals to reconsider for material error, by agencies as by lower courts; and we believe that to be another tradition that 5 U. S. C. § 701(a)(2) was meant to preserve. We are confirmed in that view by the impossibility of devising an adequate standard of review for such agency action. One is driven either to apply the ordinary standards for reviewing errors of fact or law (in which event the time limitation of the Hobbs Act—or whatever other time limitation applies to the particular case—will be entirely frustrated); or else to adopt some “clearly erroneous” standard (which produces the strange result that only really bad mistakes escape the time limitation—whatever “really bad” might mean in this context where great deference is already accorded to agency action). The concurrence chooses to impale itself upon the first horn of this dilemma. The concurrence’s effort to bring SEC v. Chenery Corp., 332 U. S. 194 (1947), into the present discussion is mis- ICC V. LOCOMOTIVE ENGINEERS 283 270 Opinion of the Court guided. That case pertains to the basis that a court may use for the affirmance of agency action that is reviewable. (It may not affirm on a basis containing any element of discretion-including discretion to find facts and interpret statutory ambiguities — that is not the basis the agency used, since that would remove the discretionary judgment from the agency to the court.) Chenery has nothing whatever to do with whether agency action is reviewable. It does not establish, as the concurrence evidently believes, the principle that if the agency gives a “reviewable” reason for otherwise unreviewable action, the action becomes reviewable. To demonstrate the falsity of that proposition it is enough to observe that a common reason for failure to prosecute an alleged criminal violation is the prosecutor’s belief (sometimes publicly stated) that the law will not sustain a conviction. That is surely an eminently “reviewable” proposition, in the sense that courts are well qualified to consider the point; yet it is entirely clear that the refusal to prosecute cannot be the subject of judicial review. Finally, we may note that the concurrence’s solution to review of denials of reconsideration, in addition to nullifying limitation periods, is simply not workable (or not workable on any basis the concurrence has explained) in the vast majority of cases. The concurrence reviews the Commission’s stated conclusions regarding 49 U. S. C. § 11341 on the usual basis applicable to agency conclusions of law, and reviews the Commission’s stated refusal to consider the newly raised (though previously available) issue of RLA crewing rights on an “abuse of discretion” standard. The vast majority of denials of reconsideration, however, are made without statement of reasons, since 5 U. S. C. § 555(e) exempts from the normal APA requirement of “a brief statement of the grounds for denial” agency action that consists of “affirming a prior denial.” One wonders how, in this more normal context, the concurrence would go about determining what answer Chenery supplies to the question of reviewability—and, 284 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. if the answer permits review, what standard to apply. U nder the proper analysis, the solution is clear: If the petition that was denied sought reopening on the basis of new evidence or changed circumstances review is available and abuse of discretion is the standard; otherwise, the agency’s refusal to go back over ploughed ground is nonre viewable. II There remains BLE’s appeal from the May 18, 1983, order denying its petition for clarification. While the petition for review was filed more than 60 days after that order was served, we conclude that it was nonetheless effective, because the timely petition for administrative reconsideration stayed the running of the Hobbs Act’s limitation period until the petition had been acted upon by the Commission. A contrary conclusion is admittedly suggested by the language of the Hobbs Act and of 49 U. S. C. § 10327(i), which provides that, “[notwithstanding” the provision authorizing the Commission to reopen and reconsider its orders (§ 10327(g)), “an action of the Commission ... is final on the date on which it is served, and a civil action to enforce, enjoin, suspend, or set aside the action may be filed after that date.” This would seem to mean that the pendency of reconsideration motions does not render Commission orders nonfinal for purposes of triggering the Hobbs Act limitations period. The same argument could be made, however, with respect to a similar provision of the APA, 5 U. S. C. §704, which reads in relevant part: “Except as otherwise expressly required by statute, agency action otherwise final is final for the purposes of this section [entitled ‘Actions Reviewable’] whether or not there has been presented or determined an application for . . . any form of reconsiderations, or, unless the agency otherwise requires by rule and provides that the action meanwhile is inoperative, for an appeal to superior agency authority.” That language has long been construed by this and other courts merely to relieve ICC v. LOCOMOTIVE ENGINEERS 285 270 Opinion of the Court parties from the requirement of petitioning for rehearing before seeking judicial review (unless, of course, specifically required to do so by statute—see, e. g., 15 U. S. C. §§717r, 3416(a)), but not to prevent petitions for reconsideration that are actually filed from rendering the orders under reconsideration nonfinal. See American Farm Lines v. Black Ball Freight Service, 397 U. S. 532, 541 (1970) (dictum); CAB v. Delta Air Lines, Inc., 367 U. S. 316, 326-327 (1961) (dictum); id., at 339-343 (Whittaker, J., dissenting); Outland n. CAB, 109 U. S. App. D. C. 90, 93, 284 F. 2d 224, 227 (1960). We can find no basis for distinguishing the language of § 10327(i) from that of § 704. The appeal from the denial of clarification was therefore timely. As with the Commission’s denial of reconsideration, however, its denial of clarification was not an appealable order. If BLE’s motion is treated as a genuine “Petition for Clarification”— i. e., as seeking nothing more than specification, one way or the other, of what the original order meant with regard to crewing rights — then the denial is unappealable because BLE was not “aggrieved” by it within the meaning of the Hobbs Act. BLE could have been aggrieved by a refusal to clarify in this narrow sense only if the refusal left it uncertain as to the Commission’s view of its rights or obligations, which plainly was not the case. Though the May 18, 1983, order denied the petition for clarification, the text of the denial made it unmistakably clear that the Commission interpreted the October 20, 1982, order as authorizing MKT and DRGW to use their own crews. BLE could, of course, disagree with that construction, but it could hardly complain that the clarification it sought had not been provided. In fact, however, we think that BLE’s petition was understood by all of the parties to be in effect a petition to reopen. BLE did not merely ask the Commission for clarification; it asked for clarification “in the manner set forth [in the petition],” App. in Nos. 83-2290 and 83-2317 (CADC), p. 5, i. e., for “clarification” that the October 20, 1982, order meant 286 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. what the unions believed it to mean. Most precisely described, the petition sought, in the alternative, clarification or (in the event clarification would be contrary to the union’s interpretation) reopening of the earlier order. Even when the petition is viewed as a petition to reopen, however, the Commission’s denial is no more reviewable than is its denial of the petitions for reconsideration discussed earlier. BLE brought forth no new evidence or changed circumstances; it merely urged the Commission to correct what BLE thought to be a serious error of law. That should have been sought many months earlier, by an appeal from the original order. We are not prepared to acknowledge an exception to that requirement where an order is ambiguous, so that a party might think that its interests are not infringed. The remedy for such ambiguity is to petition the Commission for reconsideration within the 60-day period, enabling judicial review to be pursued (if Commission resolution of the ambiguity is adverse) after disposition of that petition. Otherwise, the time limits of the Hobbs Act would be held hostage to everpresent ambiguities. If, of course, the ICC’s action here had gone beyond what was (at most) clarification of an ambiguity, and in the guise of interpreting the original order in fact revised it, that would have been a new order immediately appealable. It is impossible to make such a contention here. It was, at the very most, arguable that the Commission’s routine reference to the general NW-BN-Mendocino conditions was meant to cause those conditions to supersede, in the event of conflict, the specific terms of the trackage rights applications that the Commission generally approved—and also far from certain that any conflict between the two existed, since it was unclear whether the protective conditions extended to the situation in which the landlord and tenant railroads conduct no joint operations and the tenant carries only traffic for its own account. ICC v. LOCOMOTIVE ENGINEERS 287 270 Stevens, J., concurring in judgment The case is remanded to the Court of Appeals with instructions to dismiss the petitions for lack of jurisdiction. Vacated and remanded. Justice Stevens, with whom Justice Brennan, Justice Marshall, and Justice Blackmun join, concurring in the judgment. Congress has authorized interested parties to petition the Interstate Commerce Commission (ICC or Commission) to reopen a proceeding, grant rehearing or reconsideration, or change an action of the Commission “because of material error, new evidence, or substantially changed circumstances.” 49 U. S. C. § 10327(g). The statute draws no distinction between petitions alleging that the action was based on “material error” and petitions alleging “new evidence, or substantially changed circumstances.” Nor does the Hobbs Act, 28 U. S. C. § 2341 et seq., which affords judicial review of agency orders, recognize any such distinction. Yet, solely because it believes that such a distinction is desirable as a matter of policy, the Court today fashions a new rule of “jurisdiction.” The Court holds that agency decisions denying petitions for reopening based on “material error” are not subject to judicial review, even if the denial is arbitrary or contrary to law. Because this holding is not supported by the relevant statutes, and is contrary to longstanding principles of administrative law, I cannot subscribe to it. Addressing the merits, I conclude that the Court of Appeals erred in reversing the ICC’s decisions not to reopen the matter before it. To the extent that the Commission based those decisions on its reading of the statute, its reading was correct; to the extent that it refused to consider issues not raised previously, it did not abuse its discretion. I The Administrative Orders Review Act, 28 U. S. C. § 2341 et seq., commonly known as the Hobbs Act, is quite plain in 288 OCTOBER TERM, 1986 Stevens, J., concurring in judgment 482 U. S. spelling out when and where parties may file petitions for judicial review of agency decisions. After stating that the courts of appeals have exclusive jurisdiction to “enjoin, set aside, suspend (in whole or in part), or to determine the validity of all rules, regulations, or final orders of the Interstate Commerce Commission,” 28 U. S. C. §2342 (except for certain orders not relevant here which are reviewed in the district courts, see 28 U. S. C. §2321), the Act provides: “Any party aggrieved by the final order may, within 60 days after its entry, file a petition to review the order in the court of appeals where venue lies.” 28 U. S. C. §2344. The Court appears to agree that all of these elements were satisfied here. The denials of reopening and reconsideration constituted final orders; respondents were aggrieved parties; the petitions were filed within the required time period; and venue was appropriate in the United States Court of Appeals for the District of Columbia Circuit. Nonetheless, the Court concludes that the Court of Appeals should have dismissed the petitions for want of jurisdiction. I agree with the Court that the only agency actions properly before the Court of Appeals were the ICC’s denial of clarification served on May 18, 1983 (which the Court properly treats as a motion to reopen), and the ICC’s denial of reconsideration served on October 25, 1983. I also agree that, depending on an agency’s regulations, a decision whether to reopen may be based on discretionary factors that will virtually always survive judicial review. But this does not divest the courts of appeals of jurisdiction; it simply means that an agency’s decision based on such discretionary considerations will almost always, if not always, be upheld as not “arbitrary, capricious, an abuse of discretion, or otherwise not in accordance with law.” 5 U. S. C. §706(2)(A). See United States v. Pierce Auto Freight Lines, Inc., 327 U. S. 515, 535 (1946). Not every denial of a petition for reopening is premised on discretionary considerations. In many contexts the agency’s ICC v. LOCOMOTIVE ENGINEERS 289 270 Stevens, J., concurring in judgment authority to exercise its discretion is limited by regulation, and the agency may choose to justify its denial of the petition on the ground that those criteria have not been satisfied. Here, for example, the regulations allowed the Commission to grant a petition for reopening of a decision reached by the entire Commission only if the petitioner could show either that “[t]he prior action . . . will be affected materially because of new evidence or changed circumstances,” or that the prior action “involves material error.” 49 CFR §1115.3 (1986). Similarly, a statute provides that the Commission may only supplement a prior order approving consolidations and related transactions if “cause exists,” 49 U. S. C. § 11351, a term that has been interpreted narrowly by some courts. See Illinois v. ICC, 713 F. 2d 305, 310 (CA7 1983); Greyhound Corp. v. ICC, 215 U. S. App. D. C. 322, 330, 668 F. 2d 1354, 1362 (1981). When an agency denies rehearing on a legal ground, such as when it premises its decision on its reading of the statute, the reviewing court has a significant role to play; it must decide whether that reading can be sustained, or whether it is contrary to law.1 As we explained in Pierce Auto Freight Lines, the court’s job is to ascertain 1 In Heckler v. Chaney, 470 U. S. 821 (1985), the Court held that agency decisions not to take enforcement action are generally committed to agency discretion by law under 5 U. S. C. § 701(a)(2). Even so, the Court stressed that it was not dealing with “a refusal by the agency to institute proceedings based solely on the belief that it lacked jurisdiction.” Id., at 833, n. 4; see also NAACP v. FPC, 425 U. S. 662 (1976) (reviewing agency s refusal to promulgate rule because refusal was based on agency’s contention that it lacked jurisdiction); International Union, United Automobile, Aerospace, & Agricultural Implement Workers v. Brock, 251 U. S. App. D. C. 239, 247, 783 F. 2d 237, 245 (1986) (distinguishing decisions announcing that agency will not take enforcement action from decisions announcing substantive statutory interpretations). Of course, an agency may, if it chooses, deny the petition on discretionary grounds without considering whether the petition makes a showing sufficient to support reopening under the statute or regulations. See INS v. Bagamasbad, 429 U. S. 24, 26 (1976). 290 OCTOBER TERM, 1986 Stevens, J., concurring in judgment 482 U. S. “whether there is warrant in the law and the facts for what the Commission has done.” 327 U. S., at 536. One of the basic tenets of judicial review of agency decisions is that “an agency’s order must be upheld, if at all, ‘on the same basis articulated in the order by the agency itself.’ ” FPC n. Texaco, Inc., 417 U. S. 380, 397 (1974), quoting Burlington Truck Lines, Inc. v. United States, 371 U. S. 156, 168-169 (1962). As the Court explained in SEC v. Chenery Corp., 332 U. S. 194 (1947): “[A] simple but fundamental rule of administrative law . . . is . . . that a reviewing court, in dealing with a determination or judgment which an administrative agency alone is authorized to make, must judge the propriety of such action solely by the grounds invoked by the agency. If those grounds are inadequate or improper, the court is powerless to affirm the administrative action.” Id., at 196.2 Consequently, when an agency explains that it has denied a petition for reopening based on its understanding of the underlying statute, a reviewing court may only uphold the agency decision if that reasoning withstands review. To be sure, an agency may announce that it has declined to consider the petition for reopening because it is duplicative, because it was filed so late in the day, or for some other discretionary consideration. If the agency explains its decision on such grounds, its reasoning must be scrutinized under an abuse-of-discretion standard.3 But if the agency chooses not to pro- 2 Contrary to the Court’s suggestion, ante, at 282-283, I do not believe that Chenery establishes whether agency action is reviewable. Chenery does, however, powerfully rebut the Court’s contention that there is nothing to review when an agency bases a decision on an erroneous legal ground when it could have denied relief on discretionary grounds. 3 The Court ignores this fact when it claims that litigants will routinely disregard the Hobbs Act’s 60-day limit on judicial review unless denials of reopening are declared nonreviewable. Because an agency has the right to reject the petition on discretionary grounds, without reaching the mate- ICC v. LOCOMOTIVE ENGINEERS 291 270 Stevens, J., concurring in judgment ceed in that fashion and to base its decision on its reading of a statute or a constitutional provision, its decision cannot be sustained on the conjecture that it has the discretion to deny reopening on a variety of grounds. If the court of appeals finds legal error, it must remand the case to the agency, which may then consider whether to exercise its discretion to grant the petition for reopening.4 This is the lesson of Chenery and its progeny—a lesson the Court ignores today when it claims that “it is the Commission’s formal action, rather than its discussion, that is dispositive.” Ante, at 281. The Court argues that it would be more prudent to require the parties to seek review in the court of appeals immediately after an agency’s initial decision. This may or may not be true.5 But our view of efficient procedure does not give us the power to rewrite the United States Code or the Code of Federal Regulations. Nor does it justify ignoring this Court’s decisions explicitly holding that a denial of a petition for reopening is reviewable, see Giova v. Rosenberg, 379 U. S. 18 (1964); reversing an agency for failing to reopen a rial error alleged by the party, see n. 1, supra, litigants have the strongest of incentives to seek timely review of the agency’s original decision. 4 That many or even most denials of petitions for reopening are made without statements of reasons does not make judicial review unworkable. Of course, to the extent that 5 U. S. C. § 555(e) allows the agency to announce its denial without a statement, it may continue to do so. See Vermont Yankee Nuclear Power Corp. v. Natural Resources Defense Council, Inc., 435 U. S. 519 (1978). In such a case, the reviewing court may properly presume that the decision was reached on ordinary discretionary considerations, and may review the decision on that basis to ascertain whether it constitutes an abuse of discretion. ■ In some cases, an agency might base its order denying reopening on a new legal ground which it had not included in its original decision. Alternatively, the agency may clarify its earlier decision in a manner that gives rise to new legal challenges. In either event, it is surely unfair to deprive the aggrieved party of judicial review of an agency interpretation which he or she had no prior opportunity to challenge. The Court seems to agree that such an order would be reviewable, see ante, at 286, but, of course, cannot square this position with its newfound rule of “jurisdiction.” 292 OCTOBER TERM, 1986 Stevens, J., concurring in judgment 482 U. S. matter, see Atchison, T. & S. F. R. Co. v. United States, 284 U. S. 248 (1932);6 or reviewing denials of petitions for reopening, see, e. g., Radio Corporation of America v. United States, 341 U. S. 412, 420-421 (1951); Acker v. United States, 298 U. S. 426, 432-433 (1936); St. Joseph Stock Yards Co. v. United States, 298 U. S. 38, 47-49 (1936); United States v. Northern Pacific R. Co., 288 U. S. 490, 492-494 (1933); see also INS v. Rios-Pineda, 471 U. S. 444 (1985); INS v. Jong Ha Wang, 450 U. S. 139 (1981). In many of these cases we have stressed the limited role courts play in reviewing agency denials of reopening; but, until today, we have never held that the courts lack jurisdiction to review the decisions at all.7 As Judge Friendly put it: 6 The Court is correct in observing that Atchison, T. & S. F. R. Co. “was ‘promptly restricted ... to its special facts, . . . and . . . stands virtually alone.’” Ante, at 278, quoting ICC v. Jersey City, 322 U. S. 503, 515 (1944). But the unique aspect of the case is that it reversed the agency action—not that it reviewed the agency action. 7 In Califano v. Sanders, 430 U. S. 99 (1977), the Court held that judicial review was unavailable from the Secretary of Health, Education, and Welfare’s final decision not to reopen a claim for benefits. The Court rested this holding on the statutory language there, which provides that an individual may seek review by commencing a civil action within 60 days “after any final decision of the Secretary made after a hearing.” 42 U. S. C. § 405(g). The Court explained that “[t]his provision clearly limits judicial review to a particular type of agency action, a ‘final decision of the Secretary made after a hearing.’ ” 430 U. S., at 108. A denial of a petition for reopening did not satisfy this language, and the Court felt compelled “to respect” Congress’ choice. Ibid. Similarly, in SEC v. Louisiana Public Service Comm’n, 353 U. S. 368 (1957), the Court held that the specific judicial review language in 15 U. S. C. § 79k(b), which dealt with Commission orders that “revoke or modify” a previous order, restricted judicial review to the orders listed in the statute. Neither Sanders nor Louisiana Public Service Comm’n is relevant here for no such restrictive language applies to review of ICC decisions under the Hobbs Act. The Court’s citation of Court of Appeals decisions supporting its nonreview position, ante, at 280, is not at all conclusive. Many decisions have held that such petitions are indeed reviewable. See Carter/Mondale Presidential Committee, Inc. v. FEC, 249 U. S. App. D. C. 349, 353-354, 775 ICC v. LOCOMOTIVE ENGINEERS 293 270 Stevens, J., concurring in judgment “The fact that reopening is a matter of agency discretion to a considerable extent. . . does not lead inevitably to a conclusion that such an exercise of administrative power is wholly immune from judicial examination; § 10 (e) of the APA expressly authorizes the courts to set aside any administrative decision constituting an abuse of discretion. The question is whether the Secretary in deciding not to reopen enjoys absolute discretion .... Absent any evidence to the contrary, Congress may rather be presumed to have intended that the courts should fulfill their traditional role of defining and maintaining the proper bounds of administrative discretion.” Cappadora n. Celebrezze, 356 F. 2d 1, 5-6 (CA2 1966).8 The Court brushes off the many cases reviewing denials of petition for reopening by distinguishing between petitions for reopening based upon new evidence or changed circumstances and those based upon claims of material error. F. 2d 1182, 1186-1187 (1985); Wausau v. United States, 703 F. 2d 1042 (CA7 1983); Virginia Appalachian Lumber Corp. v. ICC, 197 U. S. App. D. C. 13, 19, 606 F. 2d 1385, 1391 (1979); Provisioners Frozen Express, Inc. v. ICC, 536 F. 2d 1303, 1305 (CA9 1976); B. F. Goodrich Co. v. Northwest Industries, Inc., 424 F. 2d 1349, 1355 (CA3), cert, denied, 400 U. S. 822 (1970); Northeast Broadcasting, Inc. v. FCC, 130 U. S. App. D. C. 278, 287-288, 400 F. 2d 749, 758-759 (1968). Cases reviewing denials of reopening in the immigration context are too common to require citation. In Provisioners Frozen Express, supra, the court dealt with a petition for review which, like the one here, had been filed more than 60 days after the initial decision, but within 60 days of the petition for reopening. The court held that the Commission’s refusal to entertain the petition for reopening “did not create a new final order which would give the Court jurisdiction to review some five years of proceedings in this matter.” Id., at 1305. But, the court hastened to add, it did have jurisdiction to determine whether the Commission abused its discretion in rejecting [the] petition to reopen.” Ibid. 8 Although, given the intervening decision in Sanders, supra, this case is no longer good law with respect to review of Social Security determinations, Judge Friendly’s discussion on the general issue of denials of petitions to reopen continues to merit our respect. 294 OCTOBER TERM, 1986 Stevens, J., concurring in judgment 482 U. S. Ante, at 278-280. According to the Court, denials of petitions for reopening that involve allegations of “new evidence” or “changed circumstances” are reviewable. Similarly, decisions reached after reopening are reviewable, even if the agency merely reaffirms its earlier decision. But, the Court proclaims, denials of petitions to reopen that allege only material error are not reviewable. Whether or not such a distinction might be reasonable, it is nevertheless a pure creature of judicial invention. I am unable to join such a creative reading of the plain language of the Hobbs Act.9 The Court’s reliance on 5 U. S. C. § 701(a)(2) for the proposition that denials of petitions for reopening are nonreview-able because they are “committed to agency discretion by law” is conclusively rebutted by the Court’s own analysis. If denials of petitions to reopen based on material error are discretionary then so are denials of petitions to reopen based on new facts or changed circumstances. Yet the Court concedes, as it must, that denials of petitions claiming new facts or changed circumstances are reviewable, notwithstanding the discretionary element. Heckler v. Chaney, 470 U. S. 821 (1985), does not speak to this issue. Unlike a prosecutor’s or an agency’s decision not to take enforcement action, there is no dearth of “judicially manageable standards” in this context. Nor is there any basis for the Court’s ability to “perceive that a . . . tradition of nonreviewability exists with regard to refusals to reconsider for material error.” Ante, at 282. I am not sure what the Court’s source for this tradition is, but it is surely not to be found in the decisions of the courts or in anything that Congress has said. It seems clear to me that neither the Hobbs Act nor the Administrative Procedure Act recognizes the distinction that 9 The only statute other than the Hobbs Act that the Court refers to on this matter is 49 U. S. C. § 10327(i) which specifies when an ICC order becomes final. That provision does not speak to the question of what types of orders are reviewable, and I do not read the Court’s opinion as relying on it. ICC v. LOCOMOTIVE ENGINEERS 295 270 Stevens, J., concurring in judgment the Court creates today. Thus, I am compelled to conclude, as the ICC itself recognizes, see Supplemental Memorandum for ICC 4, that the petitions for review in this case were timely, but only for the purpose of reviewing the Commission’s orders denying clarification and reconsideration. II The Commission’s decisions denying the petitions for clarification and reconsideration in this case were based partly on discretionary considerations and partly on the Commission’s interpretation of the relevant law. The Commission read the statute as not requiring it to make a “necessity” finding (which it had not made), and concluded that any deficiencies in the “public interest” findings that it was required to make (and had made) were attributable to the unions’ failure to raise the issue in the initial proceeding. I believe that the Commission’s legal conclusion was sound, and that it did not abuse its discretion in refusing to consider facts and theories that the unions could have brought up in their original objections to the transactions. The petitions for clarification and reconsideration apparently were prompted by the unions’ belated realization that the Missouri-Kansas-Texas Railroad Co. (MKT) and the Denver and Rio Grande Western Railroad Co. (DRGW) believed that they had the right to use their own crews to operate trains pursuant to their newly granted trackage rights over the Union Pacific Railroad Company’s (UP) and Missouri Pacific Railroad Company’s (MP) tracks. The unions argued that the Railway Labor Act (RLA), certain provisions of the Interstate Commerce Act (ICA), and their collectivebargaining agreements, gave UP and MP employees certain rights with respect to working those tracks. The MKT and DRGW, on the other hand, argued that even if such rights existed, the ICC had approved MKT’s and DRGW’s applications for trackage rights, and the applications had specified that the two railroads had the right to utilize their own 296 OCTOBER TERM, 1986 Stevens, J., concurring in judgment 482 U. S. crews. In light of the ICC’s approval, the railroads concluded, they were exempt under 49 U. S. C. § 11341 from any conflicting law on the question of who would crew the trains. That statute provides: “A carrier, corporation, or person participating in [an] approved or exempted transaction is exempt from the antitrust laws and from all other law, including State and municipal law, as necessary to let that person carry out the transaction, hold, maintain, and operate property, and exercise control or franchises acquired through the transaction.” In its denials of the petitions for clarification and reconsideration, the ICC rejected the unions’ argument that in order for the exemption provision of § 11341 to come into play, the ICC must make an actual, on the record, determination that the exemption is “necessary” to the transaction. The Commission explained: “The terms of Section 11341 immunizing an approved transaction from any other laws are self-executing and there is no need for us expressly to order or to declare that a carrier is specifically relieved from certain restraints. See Brotherhood of Loc. Eng. [ v. Chicago & N. W. R. Co., 314 F. 2d 424, 430-431 (CA8), cert denied, 375 U. S. 819 (1963)], citing Chicago, St. P., M. & 0. Ry. Lease, 295 I. C. C. 696, at 432 (1958). “In evaluating a transaction under the criteria of 49 U. S. C. 11344, we must consider the policies of statutes other than the Interstate Commerce Act to the extent that those policies are relevant to the determination of whether a proposal is consistent with the public interest. For example, the public interest evaluation must include consideration of the policies of the antitrust laws. See McLean Trucking Co. v. United States, 321 U. S. 67, 87 (1944). ICC v. LOCOMOTIVE ENGINEERS 297 270 Stevens, J., concurring in judgment “While the RLA, like the antitrust laws, embodies certain public policy considerations, the Interstate Commerce Act also specifies that interests of affected employees must be considered. In these proceedings, we gave full consideration to the impact of the consolidation on railroad employees in accordance with our established policies. 366 I. C. C. 618-22. “The record in these proceedings is devoid of any suggestion by BLE, UTU, or any other party that the approval of the responsive trackage rights applications, subject to the usual labor protective conditions, would be in any way inconsistent with the policies of the RLA. In these circumstances, we can find no merit in UTU’s argument that we improperly failed to reconcile the policies of the RLA with our decision.” App. to Pet. for Cert, in No. 85-793, pp. A44-A45. The Court of Appeals reversed, rejecting the ICC’s argument that § 11341 is self-executing and that the Commission need not make any explicit necessity findings. 245 U. S. App. D. C. 311, 320, 761 F. 2d 714, 723 (1985). The court explained that “Congress has given ICC broad powers to immunize transactions from later legal obstacles, but this delegation by Congress is explicitly qualified by a necessity component. ... In exercising its waiver authority ICC must do more than shake a wand to make a law go away. It must supply a reasoned basis for that exercise of its statutory authority.” Ibid. Because it did not believe that the ICC had supplied justification for the necessity of waiving any RLA provisions regarding crew selection, the Court remanded the case for the ICC to determine whether it should “exercise its exemption authority.” Id., at 322, 761 F. 2d, at 725. Judge MacKinnon dissented, arguing that the statute is self-executing, and that the ICC is not required to make 298 OCTOBER TERM, 1986 Stevens, J., concurring in judgment 482 U. S. any finding of necessity. Id., at 326-332, 761 F. 2d, at 729-735.10 The Court of Appeals’ holding was based on a misunderstanding of § 11341. That statute, as its plain language indicates, does not condition exemptions on the ICC’s announcing that a particular exemption is necessary to an approved transaction. Rather, § 11341 automatically exempts a person from “other laws” whenever an exemption is “necessary to let that person carry out the transaction, hold, maintain, and operate property, and exercise control or franchises acquired through the transaction.” The breadth of the exemption is defined by the scope of the approved transaction,11 and no explicit announcement of exemption is required to make the statute applicable. As this Court explained with reference to § 11341’s predecessor: “[A]pproval of a voluntary railroad merger which is within the scope of the Act is dependent upon three, and upon only three, considerations: First, a finding that it ‘will be consistent with the public interest.’ (§ 5 (2)(b).) Second, a finding that, subject to any modification made by the Commission, it is ‘just and reasonable.’ (§5 (2) (b).) Third, assent of a ‘majority ... of the holders of the shares entitled to vote.’ (§5 (11).) When these “Judge MacKinnon also dissented from the Court of Appeals’ holding that the petition was filed within sufficient time as to call into question the ICC’s initial decision. 245 U. S. App. D. C., at 322-326, 761 F. 2d, at 725-729. 11 There may, of course, be some limitations on the types of matters the ICC can approve as part of a transaction. See Palestine v. United States, 559 F. 2d 408, 414 (CA5 1977), cert, denied, 435 U. S. 950 (1978) (looking at whether ICC approval was “germane to the success of the . • . transaction”). The Commission stated in this case that “[p]rovisions of trackage rights agreements designating which carrier’s employees will perform trackage rights operations are material terms of the agreement.” App. to Pet. for Cert, in No. 85-793, p. A50. In his dissent, Judge MacKinnon elaborated on this point. 245 U. S. App. D. C., at 329, 761 F. 2d, at 732. ICC v. LOCOMOTIVE ENGINEERS 299 270 Stevens, J., concurring in judgment conditions have been complied with, the Commission-approved transaction goes into effect without the need for invoking any approval under state authority, and the parties are relieved of ‘restraints, limitations, and prohibitions of law, Federal, State, or municipal, insofar as may be necessary to enable them to carry into effect the transaction so approved....’ (§5 (11).).” Schwabacher v. United States, 334 U. S. 182, 194 (1948). See also Seaboard Air Line R. Co. v. Daniel, 333 U. S. 118, 124 (1948) (not necessary for Commission’s order to manifest “a clear purpose to authorize the exemption”); 49 U. S. C. § 11344(b) (listing items that Commission must consider in evaluating mergers). The present statute is no different in this respect; the exemption is self-executing.12 12 The legislative history of § 11341 supports the meaning derived from its plain language. Section 407 of the Transportation Act of 1920, ch. 91, 41 Stat. 480, authorized Commission approvals of consolidations, and included a section providing: “The carriers affected by any order made under the foregoing provisions of this section . . . shall be, and they are hereby, relieved from the operation of the ‘antitrust laws,’. . . and of all other restraints or prohibitions by law, State, or Federal, in so far as may be necessary to enable them to do anything authorized or required by an order made under and pursuant to the foregoing provisions of this section.” 41 Stat. 482, amending § 5(8) of the Interstate Commerce Act. The operative language, with its self-executing phraseology, was incorporated into the 1940 amendment and recodification of the Transportation Act. See § 7 of the Transportation Act of 1940, ch. 722, 54 Stat. 908-909, amending §5 (11). Again, when the Act was recodified as 49 U. S. C. § 11341(a), no substantive change was affected. Pub. L. 95-473, § 3(a), 92 Stat. 1466. See H. R. Rep. No. 95-1395, pp. 158-168 (1978). In addition, the Commission has consistently treated the exemption as automatically flowing from an approval. See, e. g., Railway Express Agency, Inc., Notes, 348 I. C. C. 157, 215 (1975); Ex Parte No. 260, Revised Regulations Governing Interlocking Officers, 336 I. C. C. 679, 681-682 (1970); Texas Turnpike Authority Abandonment by St. Louis Southwestern R. Co., 328 I. C. C. 42, 46 (1965); Chicago, St. P., M. & O. R. Co. Lease, 295 I. C. C. 696, 702 (1958); Control of Central Pacific by Southern Pacific, 76 I. C. C. 508, 515-517 (1923). A number of Courts of 300 OCTOBER TERM, 1986 Stevens, J., concurring in judgment 482 U. S. Of course, as the Commission explained, in conducting the public interest inquiry under 49 U. S. C. §11344 the Commission must consider that the legal consequence of approving the transaction as proposed will be to exempt the parties from the dictates of “other laws” to the extent necessary to carry out the transaction.13 See McLean Trucking Co. v. United States, 321 U. S. 67, 79-88 (1944). But this is a far cry from requiring the Commission to predict exactly what type of exemptions will be required and, for each one, whether the transaction could survive absent the exemption.14 Appeals have agreed. See Missouri Pacific R. Co. v. United Transportation Union, 782 F. 2d 107, 111 (CA8 1986), cert, pending, No. 85-1054; Brotherhood of Locomotive Engineers v. Chicago & N. W. R. Co., 314 F. 2d 424 (CAS), cert, denied, 375 U. S. 819 (1963). 18 This does not mean, as respondents fear, that a party claiming an exemption on the basis of § 11341 need merely assert that its conduct is “necessary” in order to prevail in its claim. Any tribunal that is faced with a claim that a party is violating some “other law” has the responsibility of determining whether an exemption is “necessary to let that person carry out the transaction, hold, maintain, and operate property, and exercise control or franchises acquired through the transaction.” 49 U. S. C. § 11341. See Railway Express Agency, supra, at 215-218. 14 The Court of Appeals appeared to recognize the problems of its suggested approach when it agreed that the ICC need not “enumerate every legal obstacle that is waived in its approval,” since to require the Commission to contemplate unforeseeable legal obstacles to fruition of the transaction would “undermine the approval authority’s purpose of ‘facilitating] merger and consolidation in the national transportation system.’” 245 U. S. App. D. C. 311, 320, n. 4, 761F. 2d 714, 723, n. 4 (1985), quoting County of Marin v. United States, 356 U. S. 412, 416 (1958). But, the court nonetheless held that the “ICC’s decisionmaking process, either in the approval or in a later proceeding, must reveal evidence supporting a conclusion that waiver of a particular legal obstacle is necessary to effectuate the transaction.” 245 U. S. App. D. C., at 320, n. 4, 761 F. 2d, at 723, n. 4. The idea of having parties repeatedly return to the ICC for decisions on the necessity of an exemption is without basis in the statutory scheme, and would clearly not mitigate the delay and confusion surrounding consolidations. ICC v. LOCOMOTIVE ENGINEERS 301 270 Stevens, J., concurring in judgment In its decision on the petition for reconsideration the Commission stated that it had considered the effect of the transaction on rail labor in its public interest calculations. Indeed, the original decision approving the transactions included a four-page discussion about its effect on labor, see Union Pacific—Control—Missouri Pacific, Western Pacific, 366 I. C. C. 462, 618-622 (1982)—issues the Commission is explicitly required to consider pursuant to 49 U. S. C. § 11344(b), and 49 U. S. C. §11347 (1982 ed., Supp. III). Generally, the Commission concluded that the “transactions will directly benefit labor both by producing a substantial net increase in existing and future employment opportunities and by increasing the job security of their present employees.” 366 I. C. C., at 619. To be sure, as the Commission itself recognized in its denial of the motion for reconsideration, it never explicitly considered the public interest ramifications of displacing any RLA provisions involving whose crew should be used in the newly granted trackage rights. But the ICC explained that it had not considered that issue because none of the unions had ever suggested that anything proposed by the railroads, including the MKT’s and DRGW’s use of their own crews, would be inconsistent with the RLA. It is on this last point that the special considerations involving review of an agency’s refusal to reopen a matter come into play. The agency clearly has the right to deny reopening to a party who failed to present evidence or to raise an objection that could have been presented or raised before the agency’s initial decision was reached.15 Indeed, even at the 161 find no merit in respondents’ suggestion that they had no notice that the Commission might consider allowing the MKT and DRGW to use their own crews in connection with their newly acquired trackage rights. Although the Commission’s approval may have been ambiguous on this issue (the Commission does not think it was), the trackage rights proposals submitted by the MKT and DRGW explicitly stated that the railroads wished to be allowed to use their own crews. See ante, at 273-274; see also Missouri Pacific R. Co., supra, at 112; App. to Pet. for Cert, in No. 85-793, pp. A45-A50. The ICC’s conclusion that the unions should have been 302 OCTOBER TERM, 1986 Stevens, J., concurring in judgment 482 U. S. time of the petition for clarification, the Commission did not believe that the unions had adequately supported their contention that the RLA had anything to say about the trackage rights issue.16 Similarly, the Commission declined to address the other issues that the unions raised, such as whether MKT’s and DRGW’s use of their own crews violated the terms of the labor protective conditions that the Commission had imposed in the approval order. See App. to Pet. for Cert, in No. 85-793, pp. A45-A50 (discussing the terms imposed pursuant to New York Dock R. Co.—Control—Brooklyn Eastern Dist., 360 I. C. C. 60 (1979), and Norfolk & Western R. Co.—Trackage Rights—BN, 354 I. C. C. 605 (1978), as modified by Mendocino Coast R. Co. —Lease and Operate, 360 I. C. C. 653, 664 (1980)). Surveying the many opportunities that the unions had to raise objections to the trackage rights proposals during the proceedings, the Commission concluded: “BLE, UTU, and various other railway labor organizations participated in these proceedings, and none made any argument or presented any evidence that the responsive trackage rights proposals would violate any applicable labor agreement. Rather, the record supports the conclusion that the trackage rights operations, using the tenants’ crews, could be implemented as approved without raising any dispute over crew assignments between the employees of different railroads.” App. to Pet for Cert, in No. 85-793, p. A45. It is thus clear that the agency’s refusal to take action based on the unions’ new claim that the use of the tenants’ crews conflicted with various laws was based on the premise that aware of the terms of the proposals is entitled to substantial deference, resting as it does on the intricacies of practice before the Commission. 16 The Commission prefaced its discussion of the §11341 issue by concluding that the unions had not adequately demonstrated that “the trackage rights agreements . . . involve a change in UP-MP employees’ working conditions in a manner contrary to RLA requirements.” Id., at A43. ICC v. LOCOMOTIVE ENGINEERS 303 270 Stevens, J., concurring in judgment the unions had, so to speak, procedurally defaulted on those claims. There is no basis for concluding that this decision constituted an abuse of discretion. I would therefore reverse the Court of Appeals on these grounds, not because it lacked jurisdiction. 304 OCTOBER TERM, 1986 Syllabus 482 U. S. FIRST ENGLISH EVANGELICAL LUTHERAN CHURCH OF GLENDALE v. COUNTY OF LOS ANGELES, CALIFORNIA APPEAL FROM THE COURT OF APPEAL OF CALIFORNIA, SECOND APPELLATE DISTRICT No. 85-1199. Argued January 14, 1987—Decided June 9, 1987 In 1957, appellant church purchased land on which it operated a campground, known as “Lutherglen,” as a retreat center and a recreational area for handicapped children. The land is located in a canyon along the banks of a creek that is the natural drainage channel for a watershed area. In 1978, a flood destroyed Lutherglen’s buildings. In response to the flood, appellee Los Angeles County, in 1979, adopted an interim ordinance prohibiting the construction or reconstruction of any building or structure in an interim flood protection area that included the land on which Lutherglen had stood. Shortly after the ordinance was adopted, appellant filed suit in a California court, alleging, inter alia, that the ordinance denied appellant all use of Lutherglen, and seeking to recover damages in inverse condemnation for such loss of use. The court granted a motion to strike the allegation, basing its ruling on Agins v. Tiburon, 24 Cal. 3d 266, 598 P. 2d 25, aff’d on other grounds, 447 U. S. 255, in which the California Supreme Court held that a landowner may not maintain an inverse condemnation suit based upon a “regulatory” taking, and that compensation is not required until the challenged regulation or ordinance has been held excessive in an action for declaratory relief or a writ of mandamus and the government has nevertheless decided to continue the regulation in effect. Because appellant alleged a regulatory taking and sought only damages, the trial court deemed the allegation that the ordinance denied all use of Lutherglen to be irrelevant. The California Court of Appeal affirmed. Held: 1. The claim that the Agins case improperly held that the Just Compensation Clause of the Fifth Amendment does not require compensation as a remedy for “temporary” regulatory takings—those regulatory takings which are ultimately invalidated by the courts—is properly presented in this case. In earlier cases, this Court was unable to reach the question because either the regulations considered to be in issue by the state courts did not effect a taking, or the factual disputes yet to be resolved by state authorities might still lead to the conclusion that no taking had occurred. Here, the California Court of Appeal assumed FIRST LUTHERAN CHURCH v. LOS ANGELES COUNTY 305 304 Syllabus that the complaint sought damages for the uncompensated “taking” of all use of Lutherglen by the ordinance, and relied on the California Supreme Court’s Agins decision for the conclusion that the remedy for the taking was limited to nonmonetary relief, thus isolating the remedial question for this Court’s consideration. MacDonald, Sommer & Frates v. Yolo County, 477 U. S. 340; Williamson County Regional Planning Comm’n v. Hamilton Bank, 473 U. S. 172; San Diego Gas & Electric Co. v. San Diego, 450 U. S. 621; and Agins, all distinguished. Pp. 311-313. 2. Under the Just Compensation Clause, where the government has “taken” property by a land-use regulation, the landowner may recover damages for the time before it is finally determined that the regulation constitutes a “taking” of his property. The Clause is designed not to limit the governmental interference with property rights per se, but rather to secure compensation in the event of otherwise proper interference amounting to a taking. A landowner is entitled to bring an action in inverse condemnation as a result of the self-executing character of the constitutional provision with respect to compensation. While the typical taking occurs when the government acts to condemn property in the exercise of its power of eminent domain, the doctrine of inverse condemnation is predicated on the proposition that a taking may occur without such formal proceedings. “Temporary” regulatory takings which, as here, deny a landowner all use of his property, are not different in kind from permanent takings for which the Constitution clearly requires compensation. Once a court determines that a taking has occurred, the government retains the whole range of options already available— amendment of the regulation, withdrawal of the invalidated regulation, or exercise of eminent domain. But where the government’s activities have already worked a taking of all use of property, no subsequent action by the government can relieve it of the duty to provide compensation for the period during which the taking was effective. Invalidation of the ordinance without payment of fair value for the use of the property during such period would be a constitutionally insufficient remedy. Pp. 314-322. Reversed and remanded. Rehnquist, C. J., delivered the opinion of the Court, in which Brennan, White, Marshall, Powell, and Scalia, JJ., joined. Stevens, J., filed a dissenting opinion, in Parts I and III of which Blackmun and O’Connor, JJ., joined, post, p. 322. Michael M. Berger argued the cause for appellant. With him on the briefs was Jerrold A. Fadem. 306 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. Jack R. White argued the cause for appellee. With him on the brief were DeWitt W. Clinton, Charles J. Moore, and Darlene B. Fischer. * Chief Justice Rehnquist delivered the opinion of the Court. In this case the California Court of Appeal held that a landowner who claims that his property has been “taken” by a land-use regulation may not recover damages for the time be- * Briefs of amici curiae urging reversal were filed for the American College of Real Estate Lawyers by Robert 0. Hetlage, David A. Richards, Eugene J. Morris, and John P. Trevaskis, Jr.; for the California Association of Realtors by William M. Pfeiffer; for the California Building Industry Association by Gideon Kanner; for the National Association of Home Builders by Kenneth B. Bley and Gus Bauman; for the National Association of Realtors by William D. North; and for the Pacific Legal Foundation et al. by Ronald A. Zumbrun and Robert K. Best. Briefs of amici curiae urging affirmance were filed for the United States by Solicitor General Fried, Assistant Attorney General Habicht, Deputy Solicitor General Ayer, Deputy Assistant Attorneys General Marzulla, Hookano, and Kmiec, and Edwin S. Kneedler and Peter R. Steenland, Jr.; for the State of California et al. by John K. Van de Kamp, Attorney General of California, Andrea Sheridan Ordin, Chief Assistant Attorney General, Richard C. Jacobs, N. Gregory Taylor, and Theodora Berger, Assistant Attorneys General, and Craig C. Thompson and Richard M. Frank, Deputy Attorneys General, joined by the Attorneys General for their respective States as follows: Harold M. Brown of Alaska, John Steven Clark of Arkansas, Jim Smith of Florida, Corinne K. A. Watanabe of Hawaii, Neil F. Hartigan of Illinois, James E. Tierney of Maine, Francis X. Bellotti of Massachusetts, Hubert H. Humphrey III of Minnesota, Edwin L. Pittman of Mississippi, William L. Webster of Missouri, Stephen E. Merrill of New Hampshire, Robert Abrams of New York, Nicholas J. Spaeth of North Dakota, Michael Turpin of Oklahoma, T. Travis Medlock of South Carolina, Mark V. Meierhenry of South Dakota, Jim Maddox of Texas, David L. Wilkinson of Utah, Jeffrey L. Amestoy of Vermont, Mary Sue Terry of Virginia, Kenneth 0. Eikenberry of Washington, Archie G. McClintock of Wyoming, and Hector Rivera Cruz of Puerto Rico; for the city of Los Angeles et al. by Gary R. Netzer, Claudia McGee Henry, and Anthony Saul Alperin; for the National Association of Counties et al. by Benna Ruth Solomon, Joyce Holmes Benjamin, and Beate Bloch; and for the Conservation Foundation et al. by Fred P. Bosselman and Elizabeth S. Merritt. FIRST LUTHERAN CHURCH v. LOS ANGELES COUNTY 307 304 Opinion of the Court fore it is finally determined that the regulation constitutes a “taking” of his property. We disagree, and conclude that in these circumstances the Fifth and Fourteenth Amendments to the United States Constitution would require compensation for that period. In 1957, appellant First English Evangelical Lutheran Church purchased a 21-acre parcel of land in a canyon along the banks of the Middle Fork of Mill Creek in the Angeles National Forest. The Middle Fork is the natural drainage channel for a watershed area owned by the National Forest Service. Twelve of the acres owned by the church are flat land, and contained a dining hall, two bunkhouses, a caretaker’s lodge, an outdoor chapel, and a footbridge across the creek. The church operated on the site a campground, known as “Lutherglen,” as a retreat center and a recreational area for handicapped children. In July 1977, a forest fire denuded the hills upstream from Lutherglen, destroying approximately 3,860 acres of the watershed area and creating a serious flood hazard. Such flooding occurred on February 9 and 10, 1978, when a storm dropped 11 inches of rain in the watershed. The runoff from the storm overflowed the banks of the Mill Creek, flooding Lutherglen and destroying its buildings. In response to the flooding of the canyon, appellee County of Los Angeles adopted Interim Ordinance No. 11,855 in January 1979. The ordinance provided that “[a] person shall not construct, reconstruct, place or enlarge any building or structure, any portion of which is, or will be, located within the outer boundary lines of the interim flood protection area located in Mill Creek Canyon . . . .” App. to Juris. Statement A31. The ordinance was effective immediately because the county determined that it was “required for the immediate preservation of the public health and safety . . . .” Id., at A32. The interim flood protection area described by the ordinance included the flat areas on either side of Mill Creek on which Lutherglen had stood. 308 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. The church filed a complaint in the Superior Court of California a little more than a month after the ordinance was adopted. As subsequently amended, the complaint alleged two claims against the county and the Los Angeles County Flood Control District. The first alleged that the defendants were liable under Cal. Govt. Code Ann. §835 (West 1980)1 for dangerous conditions on their upstream properties that contributed to the flooding of Lutherglen. As a part of this claim, appellant also alleged that “Ordinance No. 11,855 denies [appellant] all use of Lutherglen.” App. 12, 49. The second claim sought to recover from the Flood Control District in inverse condemnation and in tort for engaging in cloud seeding during the storm that flooded Lutherglen. Appellant sought damages under each count for loss of use of Lutherglen. The defendants moved to strike the portions of the complaint alleging that the county’s ordinance denied all use of Lutherglen, on the view that the California Supreme Court’s decision in Agins v. Tiburon, 24 Cal. 3d 266, 598 P. 2d 25 (1979), aff’d on other grounds, 447 U. S. 255 (1980), rendered the allegation “entirely immaterial and irrele-vant[, with] no bearing upon any conceivable cause of action herein.” App. 22. See Cal. Civ. Proc. Code Ann. §436(a) (West Supp. 1987) (“The court may. . . [s]trike out any irrelevant, false, or improper matter inserted in any pleading”). In Agins v. Tiburon, supra, the California Supreme Court decided that a landowner may not maintain an inverse condemnation suit in the courts of that State based upon a “regulatory” taking. 24 Cal. 3d, at 275-277, 598 P. 2d, at 29-31. In the court’s view, maintenance of such a suit would allow a landowner to force the legislature to exercise its power of eminent domain. Under this decision, then, compensation is not required until the challenged regulation or ordinance has been held excessive in an action for declaratory 1 Section 835 of the California Government Code establishes conditions under which a public entity may be liable “for injury caused by a dangerous condition of its property. ...” FIRST LUTHERAN CHURCH v. LOS ANGELES COUNTY 309 304 Opinion of the Court relief or a writ of mandamus and the government has nevertheless decided to continue the regulation in effect. Based on this decision, the trial court in the present case granted the motion to strike the allegation that the church had been denied all use of Lutherglen. It explained that “a careful rereading of the Agins case persuades the Court that when an ordinance, even a non-zoning ordinance, deprives a person of the total use of his lands, his challenge to the ordinance is by way of declaratory relief or possibly mandamus.” App. 26. Because the appellant alleged a regulatory taking and sought only damages, the allegation that the ordinance denied all use of Lutherglen was deemed irrelevant.2 On appeal, the California Court of Appeal read the complaint as one seeking “damages for the uncompensated taking of all use of Lutherglen by County Ordinance No. 11,855 . . . .” App. to Juris. Statement A13-A14. It too relied on the California Supreme Court’s decision in Agins in rejecting the cause of action, declining appellant’s invitation to reevaluate Agins in light of this Court’s opinions in San Diego Gas & Electric Co. n. San Diego, 450 U. S. 621 (1981). The court found itself obligated to follow Agins “because the United States Supreme Court has not yet ruled on the question of whether a state may constitutionally limit the remedy for a taking to nonmonetary relief . . . .” App. to Juris. Statement A16. It accordingly affirmed the trial court’s decision to strike the allegations concerning appellee’s ordinance.3 The California Supreme Court denied review. 2 The trial court also granted defendants’ motion for judgment on the pleadings on the second cause of action, based on cloud seeding. It limited trial on the first cause of action for damages under Cal. Govt. Code Ann. § 835 (West 1980), rejecting the inverse condemnation claim. At the close of plaintiff’s evidence, the trial court granted a nonsuit on behalf of defend- ants, dismissing the entire complaint. 8 The California Court of Appeal also affirmed the lower court’s orders limiting the issues for trial on the first cause of action, granting a nonsuit on the issues that proceeded to trial, and dismissing the second cause of action—based on cloud seeding—to the extent it was founded on a theory 310 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. This appeal followed, and we noted probable jurisdiction. 478 U. S. 1003 (1986). Appellant asks us to hold that the California Supreme Court erred in Agins v. Tiburon in determining that the Fifth Amendment, as made applicable to the States through the Fourteenth Amendment, does not require compensation as a remedy for “temporary” regulatory takings—those regulatory takings which are ultimately invalidated by the courts.4 Four times this decade, we have considered similar claims and have found ourselves for one reason or another unable to consider the merits of the Agins rule. See MacDonald, Sommer & Frates v. Yolo County, 477 U. S. 340 (1986); Williamson County Regional Planning Common w Hamilton Bank, 473 U. S. 172 (1985); San Diego Gas & Electric Co., supra; Agins v. Tiburon, supra. For the reasons explained below, however, we find the constitutional claim properly presented in this case, and hold that of strict liability in tort. The court reversed the trial court’s ruling that the second cause of action could not be maintained against the Flood Control District under the theory of inverse condemnation. The case was remanded for further proceedings on this claim. These circumstances alone, apart from the more particular issues presented in takings cases and discussed in the text, require us to consider whether the pending resolution of further liability questions deprives us of jurisdiction because we are not presented with a “final judgmen[t] or de-cre[e]” within the meaning of 28 U. S. C. § 1257. We think that this case is fairly characterized as one “in which the federal issue, finally decided by the highest court in the State [in which a decision could be had], will survive and require decision regardless of the outcome of future state-court proceedings.” Cox Broadcasting Corp. v. Cohn, 420 U. S. 469, 480 (1975). As we explain infra, at 311-313, the California Court of Appeal rejected appellant’s federal claim that it was entitled to just compensation from the county for the taking of its property; this distinct issue of federal law will survive and require decision no matter how further proceedings resolve the issues concerning the liability of the Flood Control District for its cloud seeding operation. 4 The Fifth Amendment provides “nor shall private property be taken for public use, without just compensation,” and applies to the States through the Fourteenth Amendment. See Chicago, B. & Q. R. Co. v. Chicago, 166 U. S. 226 (1897). FIRST LUTHERAN CHURCH v. LOS ANGELES COUNTY 311 304 Opinion of the Court on these facts the California courts have decided the compensation question inconsistently with the requirements of the Fifth Amendment. I Concerns with finality left us unable to reach the remedial question in the earlier cases where we have been asked to consider the rule of Agins. See MacDonald, Sommer & Frates, supra, at 351 (summarizing cases). In each of these cases, we concluded either that regulations considered to be in issue by the state court did not effect a taking, Agins v. Tiburon, 447 U. S., at 263, or that the factual disputes yet to be resolved by state authorities might still lead to the conclu-ion that no taking had occurred. MacDonald, Sommer & Frates, supra, at 351-353; Williamson County, supra, at 188-194; San Diego Gas & Electric Co., supra, at 631-632. Consideration of the remedial question in those circumstances, we concluded, would be premature. The posture of the present case is quite different. Appellant’s complaint alleged that “Ordinance No. 11,855 denies [it] all use of Lutherglen,” and sought damages for this deprivation. App. 12, 49. In affirming the decision to strike this allegation, the Court of Appeal assumed that the complaint sought “damages for the uncompensated taking of all use of Lutherglen by County Ordinance No. 11,855.” App. to Juris. Statement A13-A14 (emphasis added). It relied on the California Supreme Court’s Agins decision for the conclusion that “the remedy for a taking [is limited] to nonmonetary relief . . . .” App. to Juris. Statement A16 (emphasis added). The disposition of the case on these grounds isolates the remedial question for our consideration. The rejection of appellant’s allegations did not rest on the view that they were false. Cf. MacDonald, Sommer & Frates, supra, at 352-353, n. 8 (California court rejected allegation in the complaint that appellant was deprived of all beneficial use of its property); Agins v. Tiburon, supra, at 259, n. 6 (same). Nor did the court rely on the theory that regulatory measures such as 312 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. Ordinance No. 11,855 may never constitute a taking in the constitutional sense. Instead, the claims were deemed irrelevant solely because of the California Supreme Court’s decision in Agins that damages are unavailable to redress a “temporary” regulatory taking.5 The California Court of Appeal has thus held that, regardless of the correctness of appellant’s claim that the challenged ordinance denies it “all use of Lutherglen,” appellant may not recover damages until the ordinance is finally declared unconstitutional, and then only for any period after that declaration for which the county seeks to enforce it. The constitutional question pretermitted in our earlier cases is therefore squarely presented here.6 We reject appellee’s suggestion that, regardless of the state court’s treatment of the question, we must independently evaluate the adequacy of the complaint and resolve the 6 It has been urged that the California Supreme Court’s discussion of the compensation question in Agins v. Tiburon was dictum, because the court had already decided that the regulations could not work a taking. See Martino n. Santa Clara Valley Water District, 703 F. 2d 1141, 1147 (CA9 1983) (“extended dictum”). The Court of Appeal in this case considered and rejected the possibility that the compensation discussion in Agins was dictum. See App. to Juris. Statement A14-A15, quoting Aptos Seascape Corp. v. County of Santa Cruz, 138 Cal. App. 3d 484, 493, 188 Cal. Rptr. 191, 195 (1982) (“[I]t is apparent that the Supreme Court itself did not intend its discussion [of inverse condemnation as a remedy for a taking] to be considered dictum . . . and it has not been treated as such in subsequent Court of Appeal cases”). Whether treating the claim as a takings claim is inconsistent with the first holding of Agins is not a matter for our concern. It is enough that the court did so for us to reach the remedial question. 6 Our cases have also required that one seeking compensation must “seek compensation through the procedures the State has provided for doing so” before the claim is ripe for review. Williamson County Regional Planning Comm’n v. Hamilton Bank, 473 U. S. 172, 194 (1985). It is clear that appellant met this requirement. Having assumed that a taking occurred, the California court’s dismissal of the action establishes that “the inverse condemnation procedure is unavailable . . . .” Id., at 197. The compensation claim is accordingly ripe for our consideration. FIRST LUTHERAN CHURCH v. LOS ANGELES COUNTY 313 304 Opinion of the Court takings claim on the merits before we can reach the remedial question. However “cryptic”—to use appellee’s description—the allegations with respect to the taking were, the California courts deemed them sufficient to present the issue. We accordingly have no occasion to decide whether the ordinance at issue actually denied appellant all use of its property7 or whether the county might avoid the conclusion that a compensable taking had occurred by establishing that the denial of all use was insulated as a part of the State’s authority to enact safety regulations. See, e. g., Goldblatt v. Hempstead, 369 U. S. 590 (1962); Hadacheck v. Sebastian, 239 U. S. 394 (1915); Mugler n. Kansas, 123 U. S. 623 (1887). These questions, of course, remain open for decision on the remand we direct today. We now turn to the question whether the Just Compensation Clause requires the government to pay for “temporary” regulatory takings.8 7 Because the issue was not raised in the complaint or considered relevant by the California courts in their assumption that a taking had occurred, we also do not consider the effect of the county’s permanent ordinance on the conclusions of the courts below. That ordinance, adopted in 1981 and reproduced at App. to Juris. Statement A32-A33, provides that “[a] person shall not use, erect, construct, move onto, or . . . alter, modify, enlarge or reconstruct any building or structure within the boundaries of a flood protection district except . . . [a]ccessory buildings and structures that will not substantially impede the flow of water, including sewer, gas, electrical, and water systems, approved by the county engineer . . . ; [a]utomobile parking facilities incidental to a lawfully established use; [and] [f]lood-control structures approved by the chief engineer of the Los Angeles County Flood Control District.” County Code §22.44.220. 8 In addition to challenging the finality of the takings decision below, appellee raises two other challenges to our jurisdiction. First, going to both the appellate and certiorari jurisdiction of this Court under 28 U. S. C. § 1257, appellee alleges that appellant has failed to preserve for review any claim under federal law. Though the complaint in this case invoked only the California Constitution, appellant argued in the Court of Appeal that “recent Federal decisions . . . show the Federal Constitutional error in. . . Agins[ v. Tiburon, 24 Cal. 3d 266, 598 P. 2d 25 (1979)].” App. to Appellant’s Opposition to Appellee’s Second Motion to Dismiss A13. The Court of Appeal, by applying the state rule of Agins to dismiss appel- 314 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. II Consideration of the compensation question must begin with direct reference to the language of the Fifth Amendment, which provides in relevant part that “private property [shall not] be taken for public use, without just compensation.” As its language indicates, and as the Court has frequently noted, this provision does not prohibit the taking of private property, but instead places a condition on the exercise of that power. See Williamson County, 473 U. S., at 194; Hodel n. Virginia Surface Mining & Reclamation Assn., Inc., 452 U. S. 264, 297, n. 40 (1981); Hurley n. lant’s action, rejected on the merits the claim that the rule violated the United States Constitution. This disposition makes irrelevant for our purposes any deficiencies in the complaint as to federal issues. Where the state court has considered and decided the constitutional claim, we need not consider how or when the question was raised. Manhattan Life Ins. Co. n. Cohen, 234 U. S. 123, 134 (1914). Having succeeded in bringing the federal issue into the case, appellant preserved this question on appeal to the California Supreme Court, see App. to Appellant’s Opposition to Appellee’s Second Motion to Dismiss A14-A22, which declined to review its Agins decision. Accordingly, we find that the issue urged here was both raised and passed upon below. Second, appellee challenges our appellate jurisdiction on the grounds that the case below did not draw “in question the validity of a statute of any state . . . .” 28 U. S. C. § 1257(2). There is, of course, no doubt that the ordinance at issue in this case is “a statute of [a] state” for purposes of § 1257. See Erznoznik v. City of Jacksonville, 422 U. S. 205, 207, n. 3 (1975). As construed by the state courts, the complaint in this case alleged that the ordinance, by denying all use of the property, worked a taking without providing for just compensation. We have frequently treated such challenges to zoning ordinances as challenges to their validity under the Federal Constitution, and see no reason to revise that approach here. See, e. g., MacDonald, Sommer & Frates v. Yolo County, 477 U. S. 340 (1986); Loretto v. Teleprompter Manhattan CATV Corp., 458 U. S. 419 (1982); Agins v. Tiburon, 447 U. S. 255 (1980); Penn Central Transportation Co. v. New York City, 438 U. S. 104 (1978). By holding that the failure to provide compensation was not unconstitutional, moreover, the California courts upheld the validity of the ordinance against the particular federal constitutional question at issue here—just compensation—and the case is therefore within the terms of § 1257(2). FIRST LUTHERAN CHURCH v. LOS ANGELES COUNTY 315 304 Opinion of the Court Kincaid, 285 U. S. 95, 104 (1932); Monongahela Navigation Co. v. United States, 148 U. S. 312, 336 (1893); United States v. Jones, 109 U. S. 513, 518 (1883). This basic understanding of the Amendment makes clear that it is designed not to limit the governmental interference with property rights per se, but rather to secure compensation in the event of otherwise proper interference amounting to a taking. Thus, government action that works a taking of property rights necessarily implicates the “constitutional obligation to pay just compensation.” Armstrong n. United States, 364 U. S. 40, 49 (1960). We have recognized that a landowner is entitled to bring an action in inverse condemnation as a result of “‘the selfexecuting character of the constitutional provision with respect to compensation . . . United States v. Clarke, 445 U. S. 253, 257 (1980), quoting 6 P. Nichols, Eminent Domain §25.41 (3d rev. ed. 1972). As noted in Justice Brennan’s dissent in San Diego Gas & Electric Co., 450 U. S., at 654-655, it has been established at least since Jacobs n. United States, 290 U. S. 13 (1933), that claims for just compensation are grounded in the Constitution itself: “The suits were based on the right to recover just compensation for property taken by the United States for public use in the exercise of its power of eminent domain. That right was guaranteed by the Constitution. The fact that condemnation proceedings were not instituted and that the right was asserted in suits by the owners did not change the essential nature of the claim. The form of the remedy did not qualify the right. It rested upon the Fifth Amendment. Statutory recognition was not necessary. A promise to pay was not necessary. Such a promise was implied because of the duty to pay imposed by the Amendment. The suits were thus founded upon the Constitution of the United States.” Id., at 16. (Emphasis added.) 316 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. Jacobs, moreover, does not stand alone, for the Court has frequently repeated the view that, in the event of a taking, the compensation remedy is required by the Constitution. See, e. g., Kirby Forest Industries, Inc. v. United States, 467 U. S. 1, 5 (1984); United States v. Causby, 328 U. S. 256, 267 (1946); Seaboard Air Line R. Co. v. United States, 261 U. S. 299, 304-306 (1923); Monongahela Navigation, supra, at 327.9 It has also been established doctrine at least since Justice Holmes’ opinion for the Court in Pennsylvania Coal Co. v. Mahon, 260 U. S. 393 (1922), that “[t]he general rule at least is, that while property may be regulated to a certain extent, if regulation goes too far it will be recognized as a taking.” Id., at 415. While the typical taking occurs when the government acts to condemn property in the exercise of its power of eminent domain, the entire doctrine of inverse condemnation is predicated on the proposition that a taking may occur without such formal proceedings. In Pumpelly v. Green Bay Co., 13 Wall. 166, 177-178 (1872), construing a provision in the Wisconsin Constitution identical to the Just Compensation Clause, this Court said: “It would be a very curious and unsatisfactory result, if ... it shall be held that if the government refrains from the absolute conversion of real property to the uses of 9 The Solicitor General urges that the prohibitory nature of the Fifth Amendment, see supra, at 314, combined with principles of sovereign immunity, establishes that the Amendment itself is only a limitation on the power of the Government to act, not a remedial provision. The cases cited in the text, we think, refute the argument of the United States that “the Constitution does not, of its own force, furnish a basis for a court to award money damages against the government.” Brief for United States as Amicus Curiae 14. Though arising in various factual and jurisdictional settings, these cases make clear that it is the Constitution that dictates the remedy for interference with property rights amounting to a taking. See San Diego Gas & Electric Co. v. San Diego, 450 U. S. 621, 655, n. 21 (1981) (Brennan, J., dissenting), quoting United States v. Dickinson, 331 U. S. 745, 748 (1947). FIRST LUTHERAN CHURCH v. LOS ANGELES COUNTY 317 304 Opinion of the Court the public it can destroy its value entirely, can inflict irreparable and permanent injury to any extent, can, in effect, subject it to total destruction without making any compensation, because, in the narrowest sense of that word, it is not taken for the public use.” Later cases have unhesitatingly applied this principle. See, e. g., Kaiser Aetna n. United States, 444 U. S. 164 (1979); United States v. Dickinson, 331 U. S. 745, 750 (1947); United States v. Causby, supra. While the California Supreme Court may not have actually disavowed this general rule in Agins, we believe that it has truncated the rule by disallowing damages that occurred prior to the ultimate invalidation of the challenged regulation. The California Supreme Court justified its conclusion at length in the Agins opinion, concluding that: “In combination, the need for preserving a degree of freedom in the land-use planning function, and the inhibiting financial force which inheres in the inverse condemnation remedy, persuade us that on balance mandamus or declaratory relief rather than inverse condemnation is the appropriate relief under the circumstances.” 24 Cal. 3d, at 276-277, 598 P. 2d, at 31. We, of course, are not unmindful of these considerations, but they must be evaluated in the light of the command of the Just Compensation Clause of the Fifth Amendment. The Court has recognized in more than one case that the government may elect to abandon its intrusion or discontinue regulations. See, e. g., Kirby Forest Industries, Inc. v. United States, supra; United States v. Dow, 357 U. S. 17, 26 (1958). Similarly, a governmental body may acquiesce in a judicial declaration that one of its ordinances has effected an unconstitutional taking of property; the landowner has no right under the Just Compensation Clause to insist that a “temporary” taking be deemed a permanent taking. But we have 318 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. not resolved whether abandonment by the government requires payment of compensation for the period of time during which regulations deny a landowner all use of his land. In considering this question, we find substantial guidance in cases where the government has only temporarily exercised its right to use private property. In United States v. Dow, supra, at 26, though rejecting a claim that the Government may not abandon condemnation proceedings, the Court observed that abandonment “results in an alteration in the property interest taken—from [one of] full ownership to one of temporary use and occupation. ... In such cases compensation would be measured by the principles normally governing the taking of a right to use property temporarily. See Kimball Laundry Co. v. United States, 338 U. S. 1 [1949]; United States v. Petty Motor Co., 327 U. S. 372 [1946]; United States v. General Motors Corp., 323 U. S. 373 [1945].” Each of the cases cited by the Dow Court involved appropriation of private property by the United States for use during World War II. Though the takings were in fact “temporary,” see United States v. Petty Motor Co., 327 U. S. 372, 375 (1946), there was no question that compensation would be required for the Government’s interference with the use of the property; the Court was concerned in each case with determining the proper measure of the monetary relief to which the property holders were entitled. See Kimball Laundry Co. n. United States, 338 U. S. 1, 4-21 (1949); Petty Motor Co., supra, at 377-381; United States v. General Motors Corp., 323 U. S. 373, 379-384 (1945). These cases reflect the fact that “temporary” takings which, as here, deny a landowner all use of his property, are not different in kind from permanent takings, for which the Constitution clearly requires compensation. Cf. San Diego Gas & Electric Co., 450 U. S., at 657 (Brennan, J., dissenting) (“Nothing in the Just Compensation Clause suggests that Takings’ must be permanent and irrevocable”). It is axiomatic that the Fifth Amendment’s just compensation provision is “designed to bar Government from forcing some FIRST LUTHERAN CHURCH v. LOS ANGELES COUNTY 319 304 Opinion of the Court people alone to bear public burdens which, in all fairness and justice, should be borne by the public as a whole.” Armstrong v. United States, 364 U. S., at 49. See also Penn Central Transportation Co. v. New York City, 438 U. S. 104, 123-125 (1978); Monongahela Navigation Co. v. United States, 148 U. S., at 325. In the present case the interim ordinance was adopted by the County of Los Angeles in January 1979, and became effective immediately. Appellant filed suit within a month after the effective date of the ordinance and yet when the California Supreme Court denied a hearing in the case on October 17, 1985, the merits of appellant’s claim had yet to be determined. The United States has been required to pay compensation for leasehold interests of shorter duration than this. The value of a leasehold interest in property for a period of years may be substantial, and the burden on the property owner in extinguishing such an interest for a period of years may be great indeed. See, e. g., United States v. General Motors, supra. Where this burden results from governmental action that amounted to a taking, the Just Compensation Clause of the Fifth Amendment requires that the government pay the landowner for the value of the use of the land during this period. Cf. United States v. Causby, 328 U. S., at 261 (“It is the owner’s loss, not the taker’s gain, which is the measure of the value of the property taken”). Invalidation of the ordinance or its successor ordinance after this period of time, though converting the taking into a “temporary” one, is not a sufficient remedy to meet the demands of the Just Compensation Clause. Appellee argues that requiring compensation for denial of all use of land prior to invalidation is inconsistent with this Court’s decisions in Danforth n. United States, 308 U. S. 271 (1939), and Agins v. Tiburon, 447 U. S. 255 (1980). In Danforth, the landowner contended that the “taking” of his property had occurred prior to the institution of condemnation proceedings, by reason of the enactment of the Flood Control Act itself. He claimed that the passage of that Act had di 320 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. minished the value of his property because the plan embodied in the Act required condemnation of a flowage easement across his property. The Court held that in the context of condemnation proceedings a taking does not occur until compensation is determined and paid, and went on to say that “[a] reduction or increase in the value of property may occur by reason of legislation for or the beginning or completion of a project,” but “[s]uch changes in value are incidents of ownership. They cannot be considered as a ‘taking’ in the constitutional sense.” Danforth, supra, at 285. Agins likewise rejected a claim that the city’s preliminary activities constituted a taking, saying that “[m]ere fluctuations in value during the process of governmental decisionmaking, absent extraordinary delay, are ‘incidents of ownership.’” See 447 U. S., at 263, n. 9. But these cases merely stand for the unexceptional proposition that the valuation of property which has been taken must be calculated as of the time of the taking, and that depreciation in value of the property by reason of preliminary activity is not chargeable to the government. Thus, in Agins, we concluded that the preliminary activity did not work a taking. It would require a considerable extension of these decisions to say that no compensable regulatory taking may occur until a challenged ordinance has ultimately been held invalid.10 10 Williamson County Regional Planning Comm’n, is not to the contrary. There, we noted that “no constitutional violation occurs until just compensation has been denied.” 473 U. S., at 194, n. 13. This statement, however, was addressed to the issue whether the constitutional claim was ripe for review and did not establish that compensation is unavailable for government activity occurring before compensation is actually denied. Though, as a matter of law, an illegitimate taking might not occur until the government refuses to pay, the interference that effects a taking might begin much earlier, and compensation is measured from that time. See Kirby Forest Industries, Inc. v. United States, 467 U. S. 1, 5 (1984) (Where Government physically occupies land without condemnation proceedings, “the owner has a right to bring an ‘inverse condemnation’ suit to FIRST LUTHERAN CHURCH v. LOS ANGELES COUNTY 321 304 Opinion of the Court Nothing we say today is intended to abrogate the principle that the decision to exercise the power of eminent domain is a legislative function “ ‘for Congress and Congress alone to determine.’ ” Hawaii Housing Authority v. Midkiff, 467 U. S. 229, 240 (1984), quoting Berman v. Parker, 348 U. S. 26, 33 (1954). Once a court determines that a taking has occurred, the government retains the whole range of options already available—amendment of the regulation, withdrawal of the invalidated regulation, or exercise of eminent domain. Thus we do not, as the Solicitor General suggests, “permit a court, at the behest of a private person, to require the . . . Government to exercise the power of eminent domain . . . .” Brief for United States as Amicus Curiae 22. We merely hold that where the government’s activities have already worked a taking of all use of property, no subsequent action by the government can relieve it of the duty to provide compensation for the period during which the taking was effective. We also point out that the allegation of the complaint which we treat as true for purposes of our decision was that the ordinance in question denied appellant all use of its property. We limit our holding to the facts presented, and of course do not deal with the quite different questions that would arise in the case of normal delays in obtaining building permits, changes in zoning ordinances, variances, and the like which are not before us. We realize that even our present holding will undoubtedly lessen to some extent the freedom and flexibility of land-use planners and governing bodies of municipal corporations when enacting land-use regulations. But such consequences necessarily flow from any decision upholding a claim of constitutional right; many of the provisions of the Constitution are designed to limit the flexibility and freedom of governmental authorities, and the Just Compensation Clause of the Fifth Amendment is one of them. As Justice Holmes aptly noted more than 50 years ago, “a strong public recover the value of the land on the date of the intrusion by the Government”). (Emphasis added.) 322 OCTOBER TERM, 1986 Stevens, J., dissenting 482 U. S. desire to improve the public condition is not enough to warrant achieving the desire by a shorter cut than the constitutional way of paying for the change.” Pennsylvania Coal Co. n. Mahon, 260 U. S., at 416. Here we must assume that the Los Angeles County ordinance has denied appellant all use of its property for a considerable period of years, and we hold that invalidation of the ordinance without payment of fair value for the use of the property during this period of time would be a constitutionally insufficient remedy. The judgment of the California Court of Appeal is therefore reversed, and the case is remanded for further proceedings not inconsistent with this opinion. It is so ordered. Justice Stevens, with whom Justice Blackmun and Justice O’Connor join as to Parts I and III, dissenting. One thing is certain. The Court’s decision today will generate a great deal of litigation. Most of it, I believe, will be unproductive. But the mere duty to defend the actions that today’s decision will spawn will undoubtedly have a significant adverse impact on the land-use regulatory process. The Court has reached out to address an issue not actually presented in this case, and has then answered that self-imposed question in a superficial and, I believe, dangerous way. Four flaws in the Court’s analysis merit special comment. First, the Court unnecessarily and imprudently assumes that appellant’s complaint alleges an unconstitutional taking of Lutherglen. Second, the Court distorts our precedents in the area of regulatory takings when it concludes that all ordinances which would constitute takings if allowed to remain in effect permanently, necessarily also constitute takings if they are in effect for only a limited period of time. Third, the Court incorrectly assumes that the California Supreme Court has already decided that it will never allow a state court to grant monetary relief for a temporary regulatory taking, and FIRST LUTHERAN CHURCH v. LOS ANGELES COUNTY 323 304 Stevens, J., dissenting then uses that conclusion to reverse a judgment which is correct under the Court’s own theories. Finally, the Court errs in concluding that it is the Takings Clause, rather than the Due Process Clause, which is the primary constraint on the use of unfair and dilatory procedures in the land-use area. I In the relevant portion of its complaint for inverse condemnation, appellant alleged: “16 “On January 11, 1979, the County adopted Ordinance No. 11,855, which provides: “‘Section 1. A person shall not construct, reconstruct, place or enlarge any building or structure, any portion of which is, or will be, located within the outer boundary lines of the interim flood protection area located in Mill Creek Canyon, vicinity of Hidden Springs, as shown on Map No. 63 ML 52, attached hereto and incorporated herein by reference as though fully set forth.’ “17 “Lutherglen is within the flood protection area created by Ordinance No. 11,855. “18 “Ordinance No. 11,855 denies First Church all use of Lutherglen.” App. 49. Because the Church sought only compensation, and did not request invalidation of the ordinance, the Superior Court granted a motion to strike those three paragraphs, and consequently never decided whether they alleged a “taking.”1 'The Superior Court’s entire explanation for its decision to grant the motion to strike reads as follows: ‘However a careful rereading of the Agins case persuades the Court that when an ordinance, even a non-zoning ordinance, deprives a person of the total use of his lands, his challenge to the ordinance is by way of declaratory relief or possibly mandamus.” App. 26. 324 OCTOBER TERM, 1986 Stevens, J., dissenting 482 U. S. The Superior Court granted the motion to strike on the basis of the rule announced in Agins v. Tiburon, 24 Cal. 3d 266, 598 P. 2d 25 (1979). Under the rule of that case, a property owner who claims that a land-use restriction has taken property for public use without compensation must file an action seeking invalidation of the regulation, and may not simply demand compensation. The Court of Appeal affirmed on the authority of Agins alone,2 also without holding that the complaint had alleged a violation of either the California Constitution or the Federal Constitution. At most, it assumed, arguendo, that a constitutional violation had been alleged. This Court clearly has the authority to decide this case by ruling that the complaint did not allege a taking under the Federal Constitution,3 and therefore to avoid the novel con- 2 The Court of Appeal described the Agins case in this way: “In Agins v. City of Tiburon (1979) 24 Cal. 3d 266, the plaintiffs filed an action for damages in inverse condemnation and for declaratory relief against the City of Tiburon, which had passed a zoning ordinance in part for ‘open space’ that would have permitted a maximum of five or a minimum of one dwelling units on the plaintiffs’ five acres. A demurrer to both causes of action was sustained, and a judgment of dismissal was entered. The California Supreme Court affirmed the dismissal, finding that the ordinance did not on its face ‘deprive the landowner of substantially all reasonable use of his property,’ (Agins, supra, 24 Cal. 3d, at p. 277), and did not ‘unconstitutionally interfere with plaintiff’s entire use of the land or impermissibly decrease its value’ (ibid.). The Supreme Court further said that ‘mandamus or declaratory relief rather than inverse condemnation [was] the appropriate relief under the circumstances.’ (Ibid.)” App. to Juris. Statement A14. 8 “The familiar rule of appellate court procedure in federal courts [is] that, without a cross-petition or appeal, a respondent or appellee may support the judgment in his favor upon grounds different from those upon which the court below rested its decision.” McGoldrick v. Compagnie Generate, 309 U. S. 430, 434 (1940), citing United States v. American Railway Express Co., 265 U. S. 425,435 (1924); see also Dandridge v. 'Williams, 397 U. S. 471, 475-476, n. 6 (1970). It is also well settled that this Court is not bound by a state court’s determination (much less an assumption) that a complaint states a federal claim. See Staub v. City of Baxley, 355 U. S. 313, 318 (1958); First National Bank of Guthrie Center n. Anderson, FIRST LUTHERAN CHURCH v. LOS ANGELES COUNTY 325 304 Stevens, J., dissenting stitutional issue that it addresses. Even though I believe the Court’s lack of self-restraint is imprudent, it is imperative to stress that the Court does not hold that appellant is entitled to compensation as a result of the flood protection regulation that the county enacted. No matter whether the regulation is treated as one that deprives appellant of its property on a permanent or temporary basis, this Court’s precedents demonstrate that the type of regulatory program at issue here cannot constitute a taking. “Long ago it was recognized that ‘all property in this country is held under the implied obligation that the owner’s use of it shall not be injurious to the community.’” Keystone Bituminous Coal Assn. n. DeBenedictis, 480 U. S. 470, 491-492 (1987), quoting Mugler v. Kansas, 123 U. S. 623, 665 (1887). Thus, in order to protect the health and safety of the community,4 government may condemn unsafe structures, 269 U. S. 341, 346 (1926). Especially in the takings context, where the details of the deprivation are so significant, the economic drain of litigation on public resources is “too great to permit cases to go forward without a more substantial indication that a constitutional violation may have occurred.” Pace Resources, Inc. v. Shrewsbury Township, 808 F. 2d 1023, 1026 (CA3), cert, denied, post, p. 906. 4 See Keystone Bituminous Coal Assn. v. DeBenedictis, 480 U. S. 470, 485-493 (1987) (coal mine subsidence); Goldblatt v. Hempstead, 369 U. S. 590 (1962) (rock quarry excavation); Miller v. Schoene, 276 U. S. 272 (1928) (infectious tree disease); Hadacheck v. Sebastian, 239 U. S. 394 (1915) (emissions from factory); Mugler v. Kansas, 123 U. S. 623 (1887) (intoxicating liquors); see also Penn Central Transportation Co. v. New York City, 438 U. S. 104,145 (1978) (Rehnquist, J., dissenting) (“The question is whether the forbidden use is dangerous to the safety, health, or welfare of others”). Many state courts have reached the identical conclusion. See Keystone Bituminous, supra, at 492, n. 22 (citing cases). In Keystone Bituminous we explained that one of the justifications for the rule that health and safety regulation cannot constitute a taking is that individuals hold their property subject to the limitation that they not use it in dangerous or noxious ways. 480 U. S., at 491, n. 20. The Court’s recent decision in United States v. Cherokee Nation of Oklahoma, 480 U. S. 700 (1987), adds support to this thesis. There, the Court reaffirmed the traditional rule that when the United States exercises its power to assert a 326 OCTOBER TERM, 1986 Stevens, J., dissenting 482 U. S. may close unlawful business operations, may destroy infected trees, and surely may restrict access to hazardous areas—for example, land on which radioactive materials have been discharged, land in the path of a lava flow from an erupting volcano, or land in the path of a potentially life-threatening flood.5 When a governmental entity imposes these types of health and safety regulations, it may not be “burdened with the condition that [it] must compensate such individual owners for pecuniary losses they may sustain, by reason of their not being permitted, by a noxious use of their property, to inflict injury upon the community.” Mugler, supra, at 668-669; see generally Keystone Bituminous, supra, at 485-493. In this case, the legitimacy of the county’s interest in the enactment of Ordinance No. 11,855 is apparent from the face of the ordinance and has never been challenged.6 It was en- navigational servitude it does not “take” property because the damage sustained results “from the lawful exercise of a power to which the interests of riparian owners have always been subject.” Id., at 704. 6 See generally Plater, The Takings Issue in a Natural Setting: Floodlines and the Police Power, 52 Tex. L. Rev. 201 (1974); F. Bosselman, D. Callies, & J. Banta, The Taking Issue 147-155 (1973). 6 It is proper to take judicial notice of the ordinance. It provides, in relevant part: “ORDINANCE NO. 11,855. “An interim ordinance temporarily prohibiting the construction, reconstruction, placement or enlargement of any building or structure within any portion of the interim flood protection area delineated within Mill Creek, vicinity of Hidden Springs, declaring the urgency thereof and that this ordinance shall take immediate effect. “The Board of Supervisors of the County of Los Angeles does ordain as follows: “Section 4. Studies are now under way by the Department of Regional Planning in connection with the County Engineer and the Los Angeles County Flood Control District, to develop permanent flood protection areas for Mill Creek and other specific areas as part of a comprehensive flood plain management project. Mapping and evaluation of flood data has progressed to the point where an interim flood protection area in Mill FIRST LUTHERAN CHURCH v. LOS ANGELES COUNTY 327 304 Stevens, J., dissenting acted as an “interim” measure “temporarily prohibiting” certain construction in a specified area because the County Board believed the prohibition was “urgently required for the immediate preservation of the public health and safety.” Even if that were not true, the strong presumption of constitutionality that applies to legislative enactments certainly requires one challenging the constitutionality of an ordinance of this kind to allege some sort of improper purpose or insufficient justification in order to state a colorable federal claim for relief. A presumption of validity is particularly appropriate in this case because the complaint did not even allege that the ordinance is invalid, or pray for a declaration of invalidity or an injunction against its enforcement.7 Nor did it allege any facts indicating how the ordinance interfered with any future use of the property contemplated or planned by appellant. In light of the tragic flood and the loss of life that pre- Creek can be designated. Development is now occurring which will encroach within the limits of the permanent flood protection area and which will be incompatible with the anticipated uses to be permitted within the permanent flood protection area. If this ordinance does not take immediate effect, said uses will be established prior to the contemplated ordinance amendment, and once established may continue after such amendment has been made because of the provisions of Article 9 of Chapter 5 of Ordinance No. 1494. “By reason of the foregoing facts this ordinance is urgently required for the immediate preservation of the public health and safety, and the same shall take effect immediately upon passage thereof.” App. to Juris. Statement 31-32. 7 Because the complaint did not pray for an injunction against enforcement of the ordinance, or a declaration that it is invalid, but merely sought monetary relief, it is doubtful that we have appellate jurisdiction under 28 U. S. C. § 1257(2). Section 1257(2) provides: “(2) By appeal, where is drawn in question the validity of a statute of any state on the ground of its being repugnant to the Constitution, treaties or laws of the United States, and the decision is in favor of its validity.” Even if we do not have appellate jurisdiction, however, presumably the Court would exercise its certiorari jurisdiction pursuant to 28 U. S. C. § 1257(3). 328 OCTOBER TERM, 1986 Stevens, J., dissenting 482 U. S. cipitated the safety regulations here, it is hard to understand how appellant ever expected to rebuild on Lutherglen. Thus, although the Court uses the allegations of this complaint as a springboard for its discussion of a discrete legal issue, it does not, and could not under our precedents, hold that the allegations sufficiently alleged a taking or that the county’s effort to preserve life and property could ever constitute a taking. As far as the United States Constitution is concerned, the claim that the ordinance was a taking of Lutherglen should be summarily rejected on its merits. II There is no dispute about the proposition that a regulation which goes “too far” must be deemed a taking. See Pennsylvania Coal Co. v. Mahon, 260 U. S. 393, 415 (1922). When that happens, the government has a choice: it may abandon the regulation or it may continue to regulate and compensate those whose property it takes. In the usual case, either of these options is wholly satisfactory. Paying compensation for the property is, of course, a constitutional prerogative of the sovereign. Alternatively, if the sovereign chooses not to retain the regulation, repeal will, in virtually all cases, mitigate the overall effect of the regulation so substantially that the slight diminution in value that the regulation caused while in effect cannot be classified as a taking of property. We may assume, however, that this may not always be the case. There may be some situations in which even the temporary existence of a regulation has such severe consequences that invalidation or repeal will not mitigate the damage enough to remove the “taking” label. This hypothetical situation is what the Court calls a “temporary taking.” But, contrary to the Court’s implications, the fact that a regulation would constitute a taking if allowed to remain in effect permanently is by no means dispositive of the question whether the effect that the regulation has already had on the FIRST LUTHERAN CHURCH v. LOS ANGELES COUNTY 329 304 Stevens, J., dissenting property is so severe that a taking occurred during the period before the regulation was invalidated. A temporary interference with an owner’s use of his property may constitute a taking for which the Constitution requires that compensation be paid. At least with respect to physical takings, the Court has so held. See ante, at 318 (citing cases). Thus, if the government appropriates a leasehold interest and uses it for a public purpose, the return of the premises at the expiration of the lease would obviously not erase the fact of the government’s temporary occupation. Or if the government destroys a chicken farm by building a road through it or flying planes over it, removing the road or terminating the flights would not palliate the physical damage that had already occurred. These examples are consistent with the rule that even minimal physical occupations constitute takings which give rise to a duty to compensate. See Loretto v. Teleprompter Manhattan CATV Corp., 458 U. S. 419 (1982). But our cases also make it clear that regulatory takings and physical takings are very different in this, as well as other, respects. While virtually all physical invasions are deemed takings, see, e. g., Loretto, supra; United States v. Causby, 328 U. S. 256 (1946), a regulatory program that adversely affects property values does not constitute a taking unless it destroys a major portion of the property’s value. See Keystone Bituminous, 480 U. S., at 493-502; Hodel n. Virginia Surface Mining & Reclamation Assn., Inc., 452 U. S. 264, 296 (1981); Agins v. Tiburon, 447 U. S. 255, 260 (1980). This diminution of value inquiry is unique to regulatory takings. Unlike physical invasions, which are relatively rare and easily identifiable without making any economic analysis, regulatory programs constantly affect property values in countless ways, and only the most extreme regulations can constitute takings. Some dividing line must be established between everyday regulatory inconveniences and those so severe that they constitute takings. The diminution of value 330 OCTOBER TERM, 1986 Stevens, J., dissenting 482 U. S. inquiry has long been used in identifying that line. As Justice Holmes put it: “Government hardly could go on if to some extent values incident to property could not be diminished without paying for every such change in the general law.” Pennsylvania Coal, supra, at 413. It is this basic distinction between regulatory and physical takings that the Court ignores today. Regulations are three dimensional; they have depth, width, and length. As for depth, regulations define the extent to which the owner may not use the property in question. With respect to width, regulations define the amount of property encompassed by the restrictions. Finally, and for purposes of this case, essentially, regulations set forth the duration of the restrictions. It is obvious that no one of these elements can be analyzed alone to evaluate the impact of a regulation, and hence to determine whether a taking has occurred. For example, in Keystone Bituminous we declined to focus in on any discrete segment of the coal in the petitioners’ mines, but rather looked to the effect that the restriction had on their entire mining project. See 480 U. S., at 493-502; see also Penn Central Transportation Co. n. New York City, 438 U. S. 104, 137 (1978) (looking at owner’s other buildings). Similarly, in Penn Central, the Court concluded that it was error to focus on the nature of the uses which were prohibited without also examining the many profitable uses to which the property could still be put. Id., at 130-131; see also Agins, supra, at 262-263; Andrus v. Allard, 444 U. S. 51, 64-67 (1979). Both of these factors are essential to a meaningful analysis of the economic effect that regulations have on the value of property and on an owner’s reasonable investment-based expectations with respect to the property. Just as it would be senseless to ignore these first two factors in assessing the economic effect of a regulation, one cannot conduct the inquiry without considering the duration of the restriction. See generally Williams, Smith, Siemon, FIRST LUTHERAN CHURCH v. LOS ANGELES COUNTY 331 304 Stevens, J., dissenting Mandelker, & Babcock, The White River Junction Manifesto, 9 Vt. L. Rev. 193, 215-218 (1984). For example, while I agreed with the Chief Justice’s view that the permanent restriction on building involved in Penn Central constituted a taking, I assume that no one would have suggested that a temporary freeze on building would have also constituted a taking. Similarly, I am confident that even the dissenters in Keystone Bituminous would not have concluded that the restriction on bituminous coal mining would have constituted a taking had it simply required the mining companies to delay their operations until an appropriate safety inspection could be made. On the other hand, I am willing to assume that some cases may arise in which a property owner can show that prospective invalidation of the regulation cannot cure the taking— that the temporary operation of a regulation has caused such a significant diminution in the property’s value that compensation must be afforded for the taking that has already occurred. For this ever to happen, the restriction on the use of the property would not only have to be a substantial one, but it would also have to remain in effect for a significant percentage of the property’s useful life. In such a case an application of our test for regulatory takings would obviously require an inquiry into the duration of the restriction, as well as its scope and severity. See Williamson County Regional Planning Comm’n v. Hamilton Bank, 473 U. S. 172, 190-191 (1985) (refusing to evaluate taking claim when the longterm economic effects were uncertain because it was not clear that restrictions would remain in effect permanently). The cases that the Court relies upon for the proposition that there is no distinction between temporary and permanent takings, see ante, at 318, are inapposite, for they all deal with physical takings—where the diminution of value test is inapplicable.8 None of those cases is controversial; the state 8 In United States v. Dow, 357 U. S. 17 (1958), the United States had “entered into physical possession and began laying the pipe line through 332 OCTOBER TERM, 1986 Stevens, J., dissenting 482 U. S. certainly may not occupy an individual’s home for a month and then escape compensation by leaving and declaring the occupation “temporary.” But what does that have to do with the proper inquiry for regulatory takings? Why should there be a constitutional distinction between a permanent restriction that only reduces the economic value of the property by a fraction—perhaps one-third—and a restriction that merely postpones the development of a property for a fraction of its useful life—presumably far less than a third? In the former instance, no taking has occurred; in the latter case, the Court now proclaims that compensation for a taking must be provided. The Court makes no effort to explain these irreconcilable results. Instead, without any attempt to fit its proclamation into our regulatory takings cases, the Court boldly announces that once a property owner makes out a claim that a regulation would constitute a taking if allowed to stand, then he or she is entitled to damages for the period of time between its enactment and its invalidation. Until today, we have repeatedly rejected the notion that all temporary diminutions in the value of property automatically activate the compensation requirement of the Takings Clause. In Agins, we held: “The State Supreme Court correctly rejected the contention that the municipality’s good-faith planning activities, which did not result in successful prosecution of an eminent domain claim, so burdened the appellants’ enjoyment of their property as to constitute a taking. . . . Even if the appellants’ ability to sell their property was the tract.” Id., at 19. In Kimball Laundry Co. v. United States, 338 U. S. 1 (1949), the United States Army had taken possession of the laundry plant including all “the facilities of the company, except delivery equipment.” Id., at 3. In United States v. Petty Motor Co., 327 U. S. 372 (1946), the United States acquired by condemnation a building occupied by tenants and ordered the tenants to vacate. In United States v. General Motors Corp., 323 U. S. 373 (1945), the Government occupied a portion of a leased building. FIRST LUTHERAN CHURCH v. LOS ANGELES COUNTY 333 304 Stevens, J., dissenting limited during the pendency of the condemnation proceeding, the appellants were free to sell or develop their property when the proceedings ended. Mere fluctuations in value during the process of governmental decisionmaking, absent extraordinary delay, are ‘incidents of ownership. They cannot be considered as a “taking” in the constitutional sense.’” 447 U. S., at 263, n. 9, quoting Danforth v. United States, 308 U. S. 271, 285 (1939).9 Our more recent takings cases also cut against the approach the Court now takes. In Williamson, supra, and MacDonald, Sommer & Frates v. Yolo County, 477 U. S. 340 (1986), we held that we could not review a taking claim as long as the property owner had an opportunity to obtain a variance or some other form of relief from the zoning authorities that would permit the development of the property to go forward. See Williamson, supra, at 190-191; Yolo County, supra, at 348-353. Implicit in those holdings was the assumption that the temporary deprivation of all use of the property would not constitute a taking if it would be adequately remedied by a belated grant of approval of the developer’s plans. See Sallet, Regulatory “Takings” and Just Compensation: The Supreme Court’s Search for a Solution Continues, 18 Urb. Law. 635, 653 (1986). 9 The Court makes only a feeble attempt to explain why the holdings in Agins and Danforth are not controlling here. It is tautological to claim that the cases stand for the “unexceptional proposition that the valuation of property which has been taken must be calculated as of the time of the tak-W«” Ante, at 320 (emphasis added). The question in Danforth was when the taking occurred. The question addressed in the relevant portion of Agins was whether the temporary fluctuations in value themselves constituted a taking. In rejecting the claims in those cases, the Court necessarily held that the temporary effects did not constitute takings of their own right. The cases are therefore directly on point here. If even the temporary effects of a decision to condemn, the ultimate taking, do not ordinarily constitute a taking in and of themselves, then, a fortiori, the temporary effects of a regulation should not. 334 OCTOBER TERM, 1986 Stevens, J., dissenting 482 U. S. The Court’s reasoning also suffers from severe internal inconsistency. Although it purports to put to one side “normal delays in obtaining building permits, changes in zoning ordinances, variances and the like,” ante, at 321, the Court does not explain why there is a constitutional distinction between a total denial of all use of property during such “normal delays” and an equally total denial for the same length of time in order to determine whether a regulation has “gone too far” to be sustained unless the government is prepared to condemn the property. Precisely the same interference with a real estate developer’s plans may be occasioned by protracted proceedings which terminate with a zoning board’s decision that the public interest would be served by modification of its regulation and equally protracted litigation which ends with a judicial determination that the existing zoning restraint has “gone too far,” and that the board must therefore grant the developer a variance. The Court’s analysis takes no cognizance of these realities. Instead, it appears to erect an artificial distinction between “normal delays” and the delays involved in obtaining a court declaration that the regulation constitutes a taking.10 In my opinion, the question whether a “temporary taking” has occurred should not be answered by simply looking at the reason a temporary interference with an owner’s use of his property is terminated.11 Litigation challenging the validity of a land-use restriction gives rise to a delay that is just as “normal” as an administrative procedure seeking a variance 10 Whether delays associated with a judicial proceeding that terminates with a holding that a regulation was not authorized by state law would be a “normal delay” or a temporary taking depends, I suppose, on the unexplained rationale for the Court’s artificial distinction. 11 “[T]he Constitution measures a taking of property not by what a State says, or what it intends, but by what it does.” Hughes v. Washington, 389 U. S. 290, 298 (1967) (Stewart, J., concurring). The fact that the effects of the regulation are stopped by judicial, as opposed to administrative decree, should not affect the question whether compensation is required. FIRST LUTHERAN CHURCH v. LOS ANGELES COUNTY 335 304 Stevens, J., dissenting or an approval of a controversial plan.12 Just because a plaintiff can prove that a land-use restriction would constitute a taking if allowed to remain in effect permanently does not mean that he or she can also prove that its temporary application rose to the level of a constitutional taking. Ill The Court recognizes that the California courts have the right to adopt invalidation of an excessive regulation as the appropriate remedy for the permanent effects of overburden-some regulations, rather than allowing the regulation to stand and ordering the government to afford compensation for the permanent taking. See ante, at 319; see also Yolo County, supra, at 362-363, and n. 4 (White, J., dissenting); San Diego Gas & Electric Co. v. San Diego, 450 U. S. 621, 657 (1981) (Brennan, J., dissenting). The difference between these two remedies is less substantial than one might assume. When a court invalidates a regulation, the Legislative or Executive Branch must then decide whether to condemn the property in order to proceed with the regulatory scheme. On the other hand, if the court requires compensation for a permanent taking, the Executive or Legislative Branch may still repeal the regulation and thus prevent the permanent taking. The difference, therefore, is only in what will happen in the case of Legislative or Executive inertia. Many scholars have debated the respective merits of the alternative approaches in light of separation-of-powers concerns,13 but our only concern is with a state court’s decision on 12 States may surely provide a forum in their courts for review of general challenges to zoning ordinances and other regulations. Such a procedure then becomes part of the “normal” process. Indeed, when States have set up such procedures in their courts, we have required resort to those processes before considering takings claims. See Williamson County Regional Planning Comm’n v. Hamilton Bank, 473 U. S. 172 (1985). 13 See, e. g., Mandelker, Land Use Takings: The Compensation Issue, 8 Hastings Const. L. Q. 491 (1981); Williams, Smith, Siemon, Mandelker, & Babcock, The White River Junction Manifesto, 9 Vt. L. Rev. 193, 233-234 336 OCTOBER TERM, 1986 Stevens, J., dissenting 482 U. S. which procedure it considers more appropriate. California is fully competent to decide how it wishes to deal with the separation-of-powers implications of the remedy it routinely uses.14 Once it is recognized that California may deal with the permanent taking problem by invalidating objectionable regulations, it becomes clear that the California Court of Appeal’s decision in this case should be affirmed. Even if this Court is correct in stating that one who makes out a claim for a permanent taking is automatically entitled to some compensation for the temporary aspect of the taking as well, the States still have the right to deal with the permanent aspect of a taking by invalidating the regulation. That is all that the California courts have done in this case. They have refused to proceed upon a complaint which sought only damages, and which did not contain a request for a declaratory invalidation of the regulation, as clearly required by California precedent. The Court seriously errs, therefore, when it claims that the California court held that “a landowner who claims that his property has been ‘taken’ by a land-use regulation may not recover damages for the time before it is finally determined that the regulation constitutes a ‘taking’ of his property.” Ante, at 306-307. Perhaps the Court discerns such a practice from some of the California Supreme Court’s earlier decisions, but that is surely no reason for reversing a procedural judgment in a case in which the dismissal of the complaint was entirely consistent with an approach that the (1984); Berger & Kanner, Thoughts on the White River Junction Manifesto: A Reply to the “Gang of Five’s” Views on Just Compensation for Regulatory Taking of Property, 19 Loyola (LA) L. Rev. 685, 704-712 (1986); Comment, Just Compensation or Just Invalidation: The Availability of a Damages Remedy in Challenging Land Use Regulations, 29 UCLA L. Rev. 711, 725-726 (1982). 14 For this same reason, the parties’ and amici’s conflicting claims about whether this Court’s cases, such as Hurley v. Kincaid, 285 U. S. 95 (1932), provide that compensation is a less intrusive remedy than invalidation, are not relevant here. FIRST LUTHERAN CHURCH v. LOS ANGELES COUNTY 337 304 Stevens, J., dissenting Court endorses. Indeed, I am not all that sure how the California courts would deal with a landowner who seeks both invalidation of the regulation and damages for the temporary taking that occurred prior to the requested invalidation. As a matter of regulating the procedure in its own state courts, the California Supreme Court has decided that mandamus or declaratory relief rather than inverse condemnation provides “the appropriate relief” for one who challenges a regulation as a taking. Agins v. Tiburon, 24 Cal. 3d, at 277, 598 P. 2d, at 31. This statement in Agins can be interpreted in two quite different ways. First, it may merely require the property owner to exhaust his equitable remedies before asserting any claim for damages. Under that reading, a postponement of any consideration of monetary relief, or even a requirement that a “temporary regulatory taking” claim be asserted in a separate proceeding after the temporary interference has ended, would not violate the Federal Constitution. Second, the Agins opinion may be read to indicate that California courts will never award damages for a temporary regulatory taking.16 Even if we assume that such a rigid rule would bar recovery in the California courts in a few meritorious cases, we should not allow a litigant to challenge the rule unless his complaint contains allegations explaining why declaratory relief would not provide him with an adequate remedy, and unless his complaint at least complies with the California rule of procedure to the extent that the rule is clearly legitimate. Since the First Amendment is not implicated, the fact that California’s rule may be somewhat “overbroad” is no reason for permitting a party to complain about the impact of the rule on other property owners 16 The California Supreme Court’s discussion of the policy implications in Agins is entirely consistent with the view that the court was choosing between remedies (invalidation or compensation) with respect to the permanent effect of a regulation, and was not dealing with the temporary taking question at all. Subsequent California Supreme Court cases applying the Agins rule do not shed light on this question. 338 OCTOBER TERM, 1986 Stevens, J., dissenting 482 U. S. who actually file complaints that call California’s rule into question. In any event, the Court has no business speculating on how the California courts will deal with this problem when it is presented to them. Despite the many cases in which the California courts have applied the Agins rule, the Court can point to no case in which application of the rule has deprived a property owner of his rightful compensation. In criminal litigation we have steadfastly adhered to the practice of requiring the defendant to exhaust his or her state remedies before collaterally attacking a conviction based on a claimed violation of the Federal Constitution. That requirement is supported by our respect for the sovereignty of the several States and by our interest in having federal judges decide federal constitutional issues only on the basis of fully developed records. See generally Rose v. Lundy, 455 U. S. 509 (1982). The States’ interest in controlling land-use development and in exploring all the ramifications of a challenge to a zoning restriction should command the same deference from the federal judiciary. See Williamson, 473 U. S., at 194-197. And our interest in avoiding the decision of federal constitutional questions on anything less than a fully informed basis counsels against trying to decide whether equitable relief has forestalled a temporary taking until after we know what the relief is. In short, even if the California courts adhere to a rule of never granting monetary relief for a temporary regulatory taking, I believe we should require the property owner to exhaust his state remedies before confronting the question whether the net result of the state proceedings has amounted to a temporary taking of property without just compensation. In this case, the Church should be required to pursue an action demanding invalidation of the ordinance prior to seeking this Court’s review of California’s procedures.16 16 In the habeas corpus context, we have held that a prisoner has not exhausted his state remedies when the state court refuses to consider his FIRST LUTHERAN CHURCH v. LOS ANGELES COUNTY 339 304 Stevens, J., dissenting The appellant should not be permitted to circumvent that requirement by omitting any prayer for equitable relief from its complaint. I believe the California Supreme Court is justified in insisting that the owner recover as much of its property as possible before foisting any of it on an unwilling governmental purchaser. The Court apparently agrees with this proposition. Thus, even on the Court’s own radical view of temporary regulatory takings announced today, the California courts had the right to strike this complaint. IV There is, of course, a possibility that land-use planning, like other forms of regulation, will unfairly deprive a citizen of the right to develop his property at the time and in the manner that will best serve his economic interests. The “regulatory taking” doctrine announced in Pennsylvania Coal places a limit on the permissible scope of land-use restrictions. In my opinion, however, it is the Due Process Clause rather than that doctrine that protects the property owner from improperly motivated, unfairly conducted, or unnecessarily protracted governmental decisionmaking. Violation of the procedural safeguards mandated by the Due Process Clause will give rise to actions for damages under 42 U. S. C. § 1983, but I am not persuaded that delays in the development of property that are occasioned by fairly conducted administrative or judicial proceedings are compensable, except perhaps in the most unusual circumstances. On the contrary, I am convinced that the public interest in having important governmental decisions made in an orderly, fully informed way amply justifies the temporary burden on the citizen that is the inevitable by-product of democratic government. claim because he has not sought the appropriate state remedy. See Woods v. Nierstheimer, 328 U. S. 211, 216 (1946); Ex parte Hawk, 321 U. S. 114, 116-117 (1944). This rule should be applied with equal force here. 340 OCTOBER TERM, 1986 Stevens, J., dissenting 482 U. S. As I recently wrote: “The Due Process Clause of the Fourteenth Amendment requires a State to employ fair procedures in the administration and enforcement of all kinds of regulations. It does not, however, impose the utopian requirement that enforcement action may not impose any cost upon the citizen unless the government’s position is completely vindicated. We must presume that regulatory bodies such as zoning boards, school boards, and health boards, generally make a good-faith effort to advance the public interest when they are performing their official duties, but we must also recognize that they will often become involved in controversies that they will ultimately lose. Even though these controversies are costly and temporarily harmful to the private citizen, as long as fair procedures are followed, I do not believe there is any basis in the Constitution for characterizing the inevitable by-product of every such dispute as a ‘taking’ of private property.” Williamson, supra, at 205 (opinion concurring in judgment). The policy implications of today’s decision are obvious and, I fear, far reaching. Cautious local officials and land-use planners may avoid taking any action that might later be challenged and thus give rise to a damages action. Much important regulation will never be enacted,17 even perhaps in 17 It is no answer to say that “[a]fter all, if a policeman must know the Constitution, then why not a planner?” San Diego Gas & Electric Co. v. San Diego, 450 U. S. 621, 661, n. 26 (1981) (Brennan, J., dissenting). To begin with, the Court has repeatedly recognized that it itself cannot establish any objective rules to assess when a regulation becomes a taking. See Hodel v. Irving, 481 U. S. 704, 713-714 (1987); Andrus v. Allard, 444 U. S. 51, 65 (1979); Penn Central, 438 U. S., at 123-124. How then can it demand that land planners do any better? However confusing some of our criminal procedure cases may be, I do not believe they have been as open-ended and standardless as our regulatory takings cases are. As one commentator concluded: “The chaotic state of taking law makes it especially likely that availability of the damages remedy will induce land-use planning FIRST LUTHERAN CHURCH v. LOS ANGELES COUNTY 341 304 Stevens, J., dissenting the health and safety area. Were this result mandated by the Constitution, these serious implications would have to be ignored. But the loose cannon the Court fires today is not only unattached to the Constitution, but it also takes aim at a long line of precedents in the regulatory takings area. It would be the better part of valor simply to decide the case at hand instead of igniting the kind of litigation explosion that this decision will undoubtedly touch off. I respectfully dissent. officials to stay well back of the invisible line that they dare not cross.” Johnson, Compensation for Invalid Land-Use Regulations, 15 Ga. L. Rev. 559, 594 (1981); see also Sallet, The Problem of Municipal Liability for Zoning and Land-Use Regulation, 31 Cath. U. L. Rev. 465, 478 (1982); Charles v. Diamond, 41 N. Y. 2d 318, 331-332, 360 N. E. 2d 1295, 1305 (1977); Allen v. City and County of Honolulu, 58 Haw. 432, 439, 571 P. 2d 328, 331 (1977). Another critical distinction between police activity and land-use planning is that not every missed call by a policeman gives rise to civil liability; police officers enjoy individual immunity for actions taken in good faith. See Harlow v. Fitzgerald, 457 U. S. 800 (1982); Davis v. Scherer, 468 U. S. 183 (1984). Moreover, municipalities are not subject to civil liability for police officers’ routine judgment errors. See Monell v. New York City Dept, of Social Services, 436 U. S. 658 (1978). In the land regulation context, however, I am afraid that any decision by a competent regulatory body may establish a “policy or custom” and give rise to liability after today. 342 OCTOBER TERM, 1986 Syllabus 482 U. S. O’LONE, ADMINISTRATOR, LEESBURG PRISON COMPLEX, ET AL. v. ESTATE OF SHABAZZ et al. CERTIORARI TO THE UNITED STATES COURT OF APPEALS FOR THE THIRD CIRCUIT No. 85-1722. Argued March 24, 1987—Decided June 9, 1987 Respondents, prison inmates and members of the Islamic faith, brought suit under 42 U. S. C. § 1983 contending that two policies adopted by New Jersey prison officials prevented them from attending Jumu’ah, a Muslim congregational service held on Friday afternoons, and thereby violated their rights under the Free Exercise Clause of the First Amendment. The first such policy, Standard 853, required inmates in respondents’ custody classifications to work outside the buildings in which they were housed and in which Jumu’ah was held, while the second, a policy memorandum, prohibited inmates assigned to outside work from returning to those buildings during the day. The Federal District Court concluded that no constitutional violation had occurred, but the Court of Appeals vacated and remanded, ruling that the prison policies could be sustained only if the State showed that the challenged regulations were intended to and did serve the penological goal of security, and that no reasonable method existed by which prisoners’ religious rights could be accommodated without creating bona fide security problems. The court also held that the expert testimony of prison officials should be given due weight on, but is not dispositive of, the accommodation issue. Held: 1. The Court of Appeals erred in placing the burden on prison officials to disprove the availability of alternative methods of accommodating prisoners’ religious rights. That approach fails to reflect the respect and deference the Constitution allows for the judgment of prison administrators. P. 350. 2. The District Court’s findings establish that the policies challenged here are reasonably related to legitimate penological interests, and therefore do not offend the Free Exercise Clause. Both policies have a rational connection to the legitimate governmental interests in institutional order and security invoked to justify them, as is demonstrated by findings that Standard 853 was a response to critical overcrowding and was designed to ease tension and drain on the facilities during that part of the day when the inmates were outside, and that the policy memorandum was necessary since returns from outside work details generated congestion and delays at the main gate, a high risk area, and since the O’LONE v. ESTATE OF SHABAZZ 343 342 Syllabus need to decide return requests placed pressure on guards supervising outside work details. Rehabilitative concerns also support the policy memorandum, in light of testimony indicating that corrections officials sought thereby to simulate working conditions and responsibilities in society. Although the policies at issue may prevent some Muslim prisoners from attending Jumu’ah, their reasonableness is supported by the fact that they do not deprive respondents of all forms of religious exercise but instead allow participation in a number of Muslim religious ceremonies. Furthermore, there are no obvious, easy alternatives to the policies since both of respondents’ suggested accommodations would, in the judgment of prison officials, have adverse effects on the prison institution. Placing all Muslim inmates in inside work details would be inconsistent with the legitimate concerns underlying Standard 853, while providing weekend labor for Muslims would require extra supervision that would be a drain on scarce human resources. Both proposed accommodations would also threaten prison security by fostering “affinity groups” likely to challenge institutional authority, while any special arrangements for one group would create a perception of favoritism on the part of other inmates. Pp. 350-353. 3. Even where claims are made under the First Amendment, this Court will not substitute its judgment on difficult and sensitive matters of institutional administration for the determinations of those charged with the formidable task of running a prison. P. 353. 782 F. 2d 416, reversed. Rehnquist, C. J., delivered the opinion of the Court, in which White, Powell, O’Connor, and Scalia, JJ., joined. Brennan, J., filed a dissenting opinion, in which Marshall, Blackmun, and Stevens, JJ., joined, post, p. 354. Laurie M. Hodian, Deputy Attorney General of New Jersey, argued the cause for petitioners. With her on the briefs were W. Cary Edwards, Attorney General, and James J. Ciancia, Assistant Attorney General. Roger Clegg argued the cause for the United States as amicus curiae urging reversal. With him on the brief were Solicitor General Fried, Assistant Attorney General Weld, and Deputy Solicitor General Bryson. James Katz argued the cause and filed a brief for respondents. * *A brief of amici curiae urging reversal was filed for the Commonwealth of Pennsylvania et al. by LeRoy S. Zimmerman, Attorney General 344 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. Chief Justice Rehnquist delivered the opinion of the Court. This case requires us to consider once again the standard of review for prison regulations claimed to inhibit the exercise of constitutional rights. Respondents, members of the Is- of Pennsylvania, Amy Zapp, Deputy Attorney General, John G. Knorr III, Senior Deputy Attorney General, and Andrew S. Gordon, Chief Deputy Attorney General, Charles A. Graddick, Attorney General of Alabama, Ronald W. Lorensen, Acting Attorney General of Alaska, Robert K. Corbin, Attorney General of Arizona, Steven Clark, Attorney General of Arkansas, John K. Van de Kamp, Attorney General of California, Duane Woodard, Attorney General of Colorado, Charles M. Oberly III, Attorney General of Delaware, Jim Smith, Attorney General of Florida, Corinne K. A. Watanabe, Attorney General of Hawaii, James T. Jones, Attorney General of Idaho, Linley E. Pearson, Attorney General of Indiana, Robert P. Stephan, Attorney General of Kansas, William J. Guste, Jr., Attorney General of Louisiana, Stephen H. Sachs, Attorney General of Maryland, Francis X. Bellotti, Attorney General of Massachusetts, Frank J. Kelley, Attorney General of Michigan, Hubert H. Humphrey III, Attorney General of Minnesota, Edwin Lloyd Pittman, Attorney General of Mississippi, William L. Webster, Attorney General of Missouri, Robert M. Spire, Attorney General of Nebraska, Brian McKay, Attorney General of Nevada, Lacy H. Thornburg, Attorney General of North Carolina, Nicholas J. Spaeth, Attorney General of North Dakota, Anthony J. Celebrezze, Jr., Attorney General of Ohio, Dave Frohnmayer, Attorney General of Oregon, T. Travis Medlock, Attorney General of South Carolina, Mark V. Meierhenry, Attorney General of South Dakota, W. J. Michael Cody, Attorney General of Tennessee, Mary Sue Terry, Attorney General of Virginia, Kenneth 0. Eikenberry, Attorney General of Washington, A. G. McClintock, Attorney General of Wyoming, and James R. Murphy, Acting Corporate Counsel of the District of Columbia. Briefs of amici curiae urging affirmance were filed for the American Civil Liberties Union et al. by Eric Neisser, Alvin J. Bronstein, David B. Goldstein, Edward I. Koren, and Elizabeth Alexander; for the American Jewish Congress et al. by Marc D. Stem and Amy Adelson; for the Catholic League for Religious and Civil Rights et al. by Steven Frederick McDowell; for the Christian Legal Society et al. by Michael J. Woodruff and Samuel E. Ericsson; for the Prisoners’ Rights Project of the Legal Aid Society of the city of New York et al. by Philip L. Weinstein, David A. Lewis, and Stephen M. Latimer; for Imam Jamil Abdullah Al-Amin et al. by Ellen J. Winner, James G. Abourezk, and Albert P. Mokhiber; and for Len Marek et al. by Steven C. Moore and Walter R. Echo-Hawk. O’LONE v. ESTATE OF SHABAZZ 345 342 Opinion of the Court lamic faith, were prisoners in New Jersey’s Leesburg State Prison.1 They challenged policies adopted by prison officials which resulted in their inability to attend Jumu’ah, a weekly Muslim congregational service regularly held in the main prison building and in a separate facility known as “the Farm.” Jumu’ah is commanded by the Koran and must be held every Friday after the sun reaches its zenith and before the Asr, or afternoon prayer. See Koran 62: 9-10; Brief for Imam Jamil Abdullah Al-Amin et al. as Amici Curiae 18-31. There is no question that respondents’ sincerely held religious beliefs compelled attendance at Jumu’ah. We hold that the prison regulations here challenged did not violate respondents’ rights under the Free Exercise Clause of the First Amendment to the United States Constitution. Inmates at Leesburg are placed in one of three custody classifications. Maximum security and “gang minimum” security inmates are housed in the main prison building, and those with the lowest classification—full minimum—live in “the Farm.” Both respondents were classified as gang minimum security prisoners when this suit was filed, and respondent Mateen was later classified as full minimum. Several changes in prison policy prompted this litigation. In April 1983, the New Jersey Department of Corrections issued Standard 853, which provided that inmates could no longer move directly from maximum security to full minimum status, but were instead required to first spend a period of time in the intermediate gang minimum status. App. 147. This change was designed to redress problems that had arisen when inmates were transferred directly from the restrictive maximum security status to full minimum status, with its markedly higher level of freedom. Because of serious overcrowding in the main building, Standard 853 further mandated that gang minimum inmates ordinarily be assigned jobs outside the main building. Ibid. These inmates work in details of 8 to 15 persons, supervised by one guard. 'Respondent Shabazz died on January 15, 1986. 346 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. Standard 853 also required that full minimum inmates work outside the main institution, whether on or off prison grounds, or in a satellite building such as the Farm. Ibid. Corrections officials at Leesburg implemented these policies gradually and, as the District Court noted, with some difficulty. Shabazz v. O'Lone, 595 F. Supp. 928, 929 (NJ 1984). In the initial stages of outside work details for gang minimum prisoners, officials apparently allowed some Muslim inmates to work inside the main building on Fridays so that they could attend Jumu’ah. This alternative was eventually eliminated in March 1984, in light of the directive of Standard 853 that all gang minimum inmates work outside the main building. Significant problems arose with those inmates assigned to outside work details. Some avoided reporting for their assignments, while others found reasons for returning to the main building during the course of the workday (including their desire to attend religious services). Evidence showed that the return of prisoners during the day resulted in security risks and administrative burdens that prison officials found unacceptable. Because details of inmates were supervised by only one guard, the whole detail was forced to return to the main gate when one prisoner desired to return to the facility. The gate was the site of all incoming foot and vehicle traffic during the day, and prison officials viewed it as a high security risk area. When an inmate returned, vehicle traffic was delayed while the inmate was logged in and searched. In response to these burdens, Leesburg officials took steps to ensure that those assigned to outside details remained there for the whole day. Thus, arrangements were made to have lunch and required medications brought out to the prisoners, and appointments with doctors and social workers were scheduled for the late afternoon. These changes proved insufficient, however, and prison officials began to study alternatives. After consulting with the director of social services, the director of professional services, and the O’LONE v. ESTATE OF SHABAZZ 347 342 Opinion of the Court prison’s imam and chaplain, prison officials in March 1984 issued a policy memorandum which prohibited inmates assigned to outside work details from returning to the prison during the day except in the case of emergency. The prohibition of returns prevented Muslims assigned to outside work details from attending Jumu’ah. Respondents filed suit under 42 U. S. C. § 1983, alleging that the prison policies unconstitutionally denied them their Free Exercise rights under the First Amendment, as applied to the States through the Fourteenth Amendment. The District Court, applying the standards announced in an earlier decision of the Court of Appeals for the Third Circuit, concluded that no constitutional violation had occurred. The District Court decided that Standard 853 and the March 1984 prohibition on returns “plausibly advance” the goals of security, order, and rehabilitation. 595 F. Supp., at 934. It rejected alternative arrangements suggested by respondents, finding that “no less restrictive alternative could be adopted without potentially compromising a legitimate institutional objective.” Ibid. The Court of Appeals, sua sponte hearing the case en banc, decided that its earlier decision relied upon by the District Court was not sufficiently protective of prisoners’ free exercise rights, and went on to state that prison policies could be sustained only if: “the state . . . show[s] that the challenged regulations were intended to serve, and do serve, the important penological goal of security, and that no reasonable method exists by which [prisoners’] religious rights can be accommodated without creating bona fide security problems. The expert testimony of prison officials should be given due weight, but such testimony is not dispositive of the issue whether no reasonable adjustment is possible. . . . Where it is found that reasonable methods of accommodation can be adopted without sacrificing either the state’s interest in security or the prisoners’ interest 348 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. in freely exercising their religious rights, the state’s refusal to allow the observance of a central religious practice cannot be justified and violates the prisoner’s first amendment rights.” Shabazz v. O’Lone, 782 F. 2d 416, 420 (CA3 1986) (footnotes omitted). In considering whether a potential method of accommodation is reasonable, the court added, relevant factors include cost, the effects of overcrowding, understaffing, and inmates’ demonstrated proclivity to unruly conduct. See id., at 420, n. 3. The case was remanded to the District Court for reconsideration under the standards enumerated in the opinion. We granted certiorari to consider the important federal constitutional issues presented by the Court of Appeals’ decision, and to resolve apparent confusion among the Courts of Appeals on the proper standards to be applied in considering prisoners’ free exercise claims. 479 U. S. 881 (1986). Several general principles guide our consideration of the issues presented here. First, “convicted prisoners do not forfeit all constitutional protections by reason of their conviction and confinement in prison.” Bell v. Wolfish, 441 U. S. 520, 545 (1979). See Turner v. Safley, ante, at 84; Jones n. North Carolina Prisoners’ Labor Union, Inc., 433 U. S. 119, 129 (1977). Inmates clearly retain protections afforded by the First Amendment, Pell v. Procunier, 417 U. S. 817, 822 (1974), including its directive that no law shall prohibit the free exercise of religion. See Cruz n. Beto, 405 U. S. 319 (1972) (per curiam). Second, “[l]awful incarceration brings about the necessary withdrawal or limitation of many privileges and rights, a retraction justified by the considerations underlying our penal system.” Price v. Johnston, 334 U. S. 266, 285 (1948). The limitations on the exercise of constitutional rights arise both from the fact of incarceration and from valid penological objectives—including deterrence of crime, rehabilitation of prisoners, and institutional security. Pell n. Procunier, supra, at 822-823; Procunier v. Martinez, 416 U. S. 396, 412 (1974). O’LONE v. ESTATE OF SHABAZZ 349 342 Opinion of the Court In considering the appropriate balance of these factors, we have often said that evaluation of penological objectives is committed to the considered judgment of prison administrators, “who are actually charged with and trained in the running of the particular institution under examination.” Bell v. Wolfish, supra, at 562. See Turner n. Safiey, ante, at 86-87. To ensure that courts afford appropriate deference to prison officials, we have determined that prison regulations alleged to infringe constitutional rights are judged under a “reasonableness” test less restrictive than that ordinarily applied to alleged infringements of fundamental constitutional rights. See, e. g., Jones n. North Carolina Prisoners’ Labor Union, Inc., supra, at 128. We recently restated the proper standard: “[W]hen a prison regulation impinges on inmates’ constitutional rights, the regulation is valid if it is reasonably related to legitimate penological interests.” Turner v. Safiey, ante, at 89.2 This approach ensures the ability of corrections officials “to anticipate security problems and to adopt innovative solutions to the intractable problems of prison administration,” ibid., and avoids unnecessary intrusion of the judiciary into problems particu 2 Our decision in Turner v. Safiey rejected respondents’ principal argument in this case—that more rigorous scrutiny is appropriate unless a court can conclude that the activity for which prisoners seek protection is “presumptively dangerous.” See Brief for Respondents 30. See also Abdul Wali v. Coughlin, 754 F. 2d 1015, 1033 (CA2 1985). As we noted in Turner, ante, at 89, “[t]he determination that an activity is ‘presumptively dangerous’ appears simply to be a conclusion about the reasonableness of the prison restriction in light of the articulated security concerns. It therefore provides a tenuous basis for creating a hierarchy of standards of review.” Nor are we convinced that heightened scrutiny is appropriate whenever regulations effectively prohibit, rather than simply limit, a particular exercise of constitutional rights. See Brief for Respondents 30. As Turner makes clear, the presence or absence of alternative accommodations of prisoners’ rights is properly considered a factor in the reasonableness analysis rather than a basis for heightened scrutiny. See Turner, ante, at 88, 90-91. 350 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. larly ill suited to “resolution by decree.” Procunier n. Martinez, supra, at 405. See also Turner n. Safley, ante, at 89; Bell v. Wolfish, supra, at 548. We think the Court of Appeals decision in this case was wrong when it established a separate burden on prison officials to prove “that no reasonable method exists by which [prisoners’] religious rights can be accommodated without creating bona fide security problems.” 782 F. 2d, at 420. See also id., at 419 (Prison officials should be required “to produce convincing evidence that they are unable to satisfy their institutional goals in any way that does not infringe inmates’ free exercise rights”). Though the availability of accommodations is relevant to the reasonableness inquiry, we have rejected the notion that “prison officials . . . have to set up and then shoot down every conceivable alternative method of accommodating the claimant’s constitutional complaint.” Turner v. Safley, ante, at 90-91. By placing the burden on prison officials to disprove the availability of alternatives, the approach articulated by the Court of Appeals fails to reflect the respect and deference that the United States Constitution allows for the judgment of prison administrators. Turning to consideration of the policies challenged in this case, we think the findings of the District Court establish clearly that prison officials have acted in a reasonable manner. Turner n. Safley drew upon our previous decisions to identify several factors relevant to this reasonableness determination. First, a regulation must have a logical connection to legitimate governmental interests invoked to justify it. Ante, at 89-90. The policies at issue here clearly meet that standard. The requirement that full minimum and gang minimum prisoners work outside the main facility was justified by concerns of institutional order and security, for the District Court found that it was “at least in part a response to a critical overcrowding in the state’s prisons, and ... at least in part designed to ease tension and drain on the facilities O’LONE v. ESTATE OF SHABAZZ 351 342 Opinion of the Court during that part of the day when the inmates were outside the confines of the main buildings.” 595 F. Supp., at 929. We think it beyond doubt that the standard is related to this legitimate concern. The subsequent policy prohibiting returns to the institution during the day also passes muster under this standard. Prison officials testified that the returns from outside work details generated congestion and delays at the main gate, a high risk area in any event. Return requests also placed pressure on guards supervising outside details, who previously were required to “evaluate each reason possibly justifying a return to the facilities and either accept or reject that reason.” Id., at 931. Rehabilitative concerns further supported the policy; corrections officials sought a simulation of working conditions and responsibilities in society. Chief Deputy Ucci testified: “One of the things that society demands or expects is that when you have a job, you show up on time, you put in your eight hours, or whatever hours you are supposed to put in, and you don’t get off ... . If we can show inmates that they’re supposed to show up for work and work a full day, then when they get out at least we’ve done something.” Tr. 89. These legitimate goals were advanced by the prohibition on returns; it cannot seriously be maintained that “the logical connection between the regulation and the asserted goal is so remote as to render the policy arbitrary or irrational.” Turner v. Safley, ante, at 89-90. Our decision in Turner also found it relevant that “alternative means of exercising the right. . . remain open to prison inmates.” Ante, at 90. There are, of course, no alternative means of attending Jumu’ah; respondents’ religious beliefs insist that it occur at a particular time. But the very stringent requirements as to the time at which Jumu’ah may be held may make it extraordinarily difficult for prison officials to assure that every Muslim prisoner is able to attend that serv-ice. While we in no way minimize the central importance of Jumu’ah to respondents, we are unwilling to hold that prison 352 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. officials are required by the Constitution to sacrifice legitimate penological objectives to that end. In Turner, we did not look to see whether prisoners had other means of communicating with fellow inmates, but instead examined whether the inmates were deprived of “all means of expression.” Ante, at 92. Here, similarly, we think it appropriate to see whether under these regulations respondents retain the ability to participate in other Muslim religious ceremonies. The record establishes that respondents are not deprived of all forms of religious exercise, but instead freely observe a number of their religious obligations. The right to congregate for prayer or discussion is “virtually unlimited except during working hours,” Tr. 182 (testimony of O’Lone), and the state-provided imam has free access to the prison. Muslim prisoners are given different meals whenever pork is served in the prison cafeteria. Special arrangements are also made during the month-long observance of Ramadan, a period of fasting and prayer. During Ramadan, Muslim prisoners are awakened at 4 a.m. for an early breakfast, and receive dinner at 8:30 each evening. We think this ability on the part of respondents to participate in other religious observances of their faith supports the conclusion that the restrictions at issue here were reasonable. Finally, the case for the validity of these regulations is strengthened by examination of the impact that accommodation of respondents’ asserted right would have on other inmates, on prison personnel, and on allocation of prison resources generally. See Turner n. Safley, ante, at 90. Respondents suggest several accommodations of their practices, including placing all Muslim inmates in one or two inside work details or providing weekend labor for Muslim inmates. See Brief for Respondents 52-53. As noted by the District Court, however, each of respondents’ suggested accommodations would, in the judgment of prison officials, have adverse effects on the institution. Inside work details for gang minimum inmates would be inconsistent with the legitimate con- O’LONE v. ESTATE OF SHABAZZ 353 342 Opinion of the Court cems underlying Standard 853, and the District Court found that the extra supervision necessary to establish weekend details for Muslim prisoners “would be a drain on scarce human resources” at the prison. 595 F. Supp., at 932. Prison officials determined that the alternatives would also threaten prison security by allowing “affinity groups” in the prison to flourish. Administrator O’Lone testified that “we have found out and think almost every prison administrator knows that any time you put a group of individuals together with one particular affinity interest . . . you wind up with ... a leadership role and an organizational structure that will almost invariably challenge the institutional authority.” Tr. 179-180. Finally, the officials determined that special arrangements for one group would create problems as “other inmates [see] that a certain segment is escaping a rigorous work detail” and perceive favoritism. Id., at 178-179. These concerns of prison administrators provide adequate support for the conclusion that accommodations of respondents’ request to attend Jumu’ah would have undesirable results in the institution. These difficulties also make clear that there are no “obvious, easy alternatives to the policy adopted by petitioners.” Turner v. Safley, ante, at 93. We take this opportunity to reaffirm our refusal, even where claims are made under the First Amendment, to “substitute our judgment on . . . difficult and sensitive matters of institutional administration,” Block v. Rutherford, 468 U. S. 576, 588 (1984), for the determinations of those charged with the formidable task of running a prison. Here the District Court decided that the regulations alleged to infringe constitutional rights were reasonably related to legitimate penological objectives. We agree with the District Court, and it necessarily follows that the regulations in question do not offend the Free Exercise Clause of the First Amendment to the United States Constitution. The judgment of the Court °f Appeals is therefore Reversed. 354 OCTOBER TERM, 1986 Brennan, J., dissenting 482 U. S. Justice Brennan, with whom Justice Marshall, Justice Blackmun, and Justice Stevens join, dissenting. The religious ceremony that these respondents seek to attend is not presumptively dangerous, and the prison has completely foreclosed respondents’ participation in it. I therefore would require prison officials to demonstrate that the restrictions they have imposed are necessary to further an important government interest, and that these restrictions are no greater than necessary to achieve prison objectives. See Turner n. Safley, ante, at 101, n. 1 (Stevens, J., concurring in part and dissenting in part) (citing Abdul Wali v. Coughlin, 754 F. 2d 1015 (CA2 1985)). As a result, I would affirm the Court of Appeals’ order to remand the case to the District Court, and would require prison officials to make this showing. Even were I to accept the Court’s standard of review, however, I would remand the case to the District Court, since that court has not had the opportunity to review respondents’ claim under the new standard established by this Court in Turner. As the record now stands, the reasonableness of foreclosing respondents’ participation in Jumu’ah has not been established. I Prisoners are persons whom most of us would rather not think about. Banished from everyday sight, they exist in a shadow world that only dimly enters our awareness. They are members of a “total institution”1 that controls their daily existence in a way that few of us can imagine: “[P]rison is a complex of physical arrangements and of measures, all wholly governmental, all wholly performed by agents of government, which determine the total existence of certain human beings (except perhaps in the realm of the spirit, and inevitably there as well) from sundown to sundown, sleeping, waking, speaking, silent, 1 See E. Goffman, Asylums: Essays on the Social Situation of Mental Patients and Other Inmates 1-125 (1961) O’LONE v. ESTATE OF SHABAZZ 355 342 Brennan, J., dissenting working, playing, viewing, eating, voiding, reading, alone, with others. It is not so, with members of the general adult population. State governments have not undertaken to require members of the general adult population to rise at a certain hour, retire at a certain hour, eat at certain hours, live for periods with no companionship whatever, wear certain clothing, or submit to oral and anal searches after visiting hours, nor have state governments undertaken to prohibit members of the general adult population from speaking to one another, wearing beards, embracing their spouses, or corresponding with their lovers.” Morales v. Schmidt, 340 F. Supp. 544, 550 (WD Wis. 1972). It is thus easy to think of prisoners as members of a separate netherworld, driven by its own demands, ordered by its own customs, ruled by those whose claim to power rests on raw necessity. Nothing can change the fact, however, that the society that these prisoners inhabit is our own. Prisons may exist on the margins of that society, but no act of will can sever them from the body politic. When prisoners emerge from the shadows to press a constitutional claim, they invoke no alien set of principles drawn from a distant culture. Rather, they speak the language of the charter upon which all of us rely to hold official power accountable. They ask us to acknowledge that power exercised in the shadows must be restrained at least as diligently as power that acts in the sunlight. In reviewing a prisoner’s claim of the infringement of a constitutional right, we must therefore begin from the premise that, as members of this society, prisoners retain constitutional rights that limit the exercise of official authority against them. See Bell v. Wolfish, 441 U. S. 520, 545 (1979). At the same time, we must acknowledge that incarceration by its nature changes an individual’s status in society. Prison officials have the difficult and often thankless job of preserving security in a potentially explosive setting, 356 OCTOBER TERM, 1986 Brennan, J., dissenting 482 U. S. as well as of attempting to provide rehabilitation that prepares some inmates for re-entry into the social mainstream. Both these demands require the curtailment and elimination of certain rights. The challenge for this Court is to determine how best to protect those prisoners’ rights that remain. Our objective in selecting a standard of review is therefore not, as the Court declares, “[t]o ensure that courts afford appropriate deference to prison officials.” Ante, at 349. The Constitution was not adopted as a means of enhancing the efficiency with which government officials conduct their affairs, nor as a blueprint for ensuring sufficient reliance on administrative expertise. Rather, it was meant to provide a bulwark against infringements that might otherwise be justified as necessary expedients of governing. The practice of Europe, wrote James Madison, was “charters of liberty . . . granted by power”; of America, “charters of power granted by liberty.” 6 Writings of James Madison 83 (G. Hunt ed. 1906). While we must give due consideration to the needs of those in power, this Court’s role is to ensure that fundamental restraints on that power are enforced. In my view, adoption of “reasonableness” as a standard of review for all constitutional challenges by inmates is inadequate to this task. Such a standard is categorically deferential, and does not discriminate among degrees of deprivation. From this perspective, restricting use of the prison library to certain hours warrants the same level of scrutiny as preventing inmates from reading at all. Various “factors” may be weighed differently in each situation, but the message to prison officials is clear: merely act “reasonably” and your actions will be upheld. If a directive that officials act “reasonably” were deemed sufficient to check all exercises of power, the Constitution would hardly be necessary. Yet the Court deems this single standard adequate to restrain any type of conduct in which prison officials might engage. O’LONE v. ESTATE OF SHABAZZ 357 342 Brennan, J., dissenting It is true that the degree of deprivation is one of the factors in the Court’s reasonableness determination. This by itself does not make the standard of review appropriate, however. If it did, we would need but a single standard for evaluating all constitutional claims, as long as every relevant factor were considered under its rubric. Clearly, we have never followed such an approach. A standard of review frames the terms in which justification may be offered, and thus delineates the boundaries within which argument may take place.2 The use of differing levels of scrutiny proclaims that on some occasions official power must justify itself in a way that otherwise it need not. A relatively strict standard of review is a signal that a decree prohibiting a political demonstration on the basis of the participants’ political beliefs is of more serious concern, and therefore will be scrutinized more closely, than a rule limiting the number of demonstrations that may take place downtown at noon. Thus, even if the absolute nature of the deprivation may be taken into account in the Court’s formulation, it makes a difference that this is merely one factor in determining if official conduct is “reasonable.” Once we provide such an elastic and deferential principle of justification, “[t]he principle . . . lies about like a loaded weapon ready for the hand of any authority that can bring forth a plausible claim of an urgent need. Every repetition imbeds that principle more deeply in our law and thinking and expands it to new purposes.” Korematsu v. United States, 323 U. S. 214, 246 (1944) (Jackson, J., 2 As one scholar has commented: ‘The language that the lawyer uses and remakes is a language of meaning in the fullest sense. It is a language in which our perceptions of the natural universe are constructed and related, in which our values and motives are defined, in which our methods of reasoning are elaborated and enacted; and it gives us our terms for constructing a social universe by defining roles and actors and by establishing expectations as to the propriety of speech and conduct.” J. B. White, Rhetoric and Law: The Arts of Cultural and Communal Life, in Heracles’ Bow: Essays on the Rhetoric and Poetics of the Law 36 (1985). 358 OCTOBER TERM, 1986 Brennan, J., dissenting 482 U. S. dissenting). Mere assertions of exigency have a way of providing a colorable defense for governmental deprivation, and we should be especially wary of expansive delegations of power to those who wield it on the margins of society. Prisons are too often shielded from public view; there is no need to make them virtually invisible. An approach better suited to the sensitive task of protecting the constitutional rights of inmates is laid out by Judge Kaufman in Abdul Wali v. Coughlin, 754 F. 2d 1015 (CA2 1985). That approach maintains that the degree of scrutiny of prison regulations should depend on “the nature of the right being asserted by prisoners, the type of activity in which they seek to engage, and whether the challenged restriction works a total deprivation (as opposed to a mere limitation) on the exercise of that right.” Id., at 1033. Essentially, if the activity in which inmates seek to engage is presumptively dangerous, or if a regulation merely restricts the time, place, or manner in which prisoners may exercise a right, a prison regulation will be invalidated only if there is no reasonable justification for official action. Ibid. Where exercise of the asserted right is not presumptively dangerous, however, and where the prison has completely deprived an inmate of that right, then prison officials must show that “a particular restriction is necessary to further an important governmental interest, and that the limitations on freedoms occasioned by the restrictions are no greater than necessary to effectuate the governmental objective involved.” Ibid. The court’s analytical framework in Abdul Wali recognizes that in many instances it is inappropriate for courts “to substitute our judgments for those of trained professionals with years of firsthand experience.” Ibid. It would thus apply a standard of review identical to the Court’s “reasonableness” standard in a significant percentage of cases. At the same time, the Abdul Wali approach takes seriously the Constitution’s function of requiring that official power be called to account when it completely deprives a person of a right that O’LONE v. ESTATE OF SHABAZZ 359 342 Brennan, J., dissenting society regards as basic. In this limited number of cases, it would require more than a demonstration of “reasonableness” to justify such infringement. To the extent that prison is meant to inculcate a respect for social and legal norms, a requirement that prison officials persuasively demonstrate the need for the absolute deprivation of inmate rights is consistent with that end. Furthermore, prison officials are in control of the evidence that is essential to establish the superiority of such deprivation over other alternatives. It is thus only fair for these officials to be held to a stringent standard of review in such extreme cases. The prison in this case has completely prevented respondent inmates from attending the central religious service of their Muslim faith. I would therefore hold prison officials to the standard articulated in Abdul Wali, and would find their proffered justifications wanting. The State has neither demonstrated that the restriction is necessary to further an important objective nor proved that less extreme measures may not serve its purpose. Even if I accepted the Court’s standard of review, however, I could not conclude on this record that prison officials have proved that it is reasonable to preclude respondents from attending Jumu’ah. Petitioners have provided mere unsubstantiated assertions that the plausible alternatives proposed by respondents are infeasible. II In Turner, the Court set forth a framework for reviewing allegations that a constitutional right has been infringed by prison officials. The Court found relevant to that review “whether there are alternative means of exercising the right that remain open to prison inmates.” Ante, at 90. The Court in this case acknowledges that “respondents’ sincerely held religious beliefs compe[l] attendance at Jumu’ah,” ante, at 345, and concedes that there are “no alternative means of attending Jumu’ah.” Ante, at 351. Nonetheless, the Court finds that prison policy does not work a complete 360 OCTOBER TERM, 1986 Brennan, J., dissenting 482 U. S. deprivation of respondents’ asserted religious right, because respondents have the opportunity to participate in other religious activities. Ante, at 352. This analysis ignores the fact that, as the District Court found, Jumu’ah is the central religious ceremony of Muslims, “comparable to the Saturday service of the Jewish faith and the Sunday service of the various Christian sects.” Shabazz v. O’Lone, 595 F. Supp. 928, 930 (NJ 1984). As with other faiths, this ceremony provides a special time in which Muslims “assert their identity as a community covenanted to God.” Brief for Imam Jamil Abdullah Al-Amin et al. as Amici Curiae 32. As a result: “unlike other Muslim prayers which are performed individually and can be made up if missed, the Jumu’ah is obligatory, cannot be made up, and must be performed in congregation. The Jumu’ah is therefore regarded as the central service of the Muslim religion, and the obligation to attend is commanded by the Qur’an, the central book of the Muslim religion.” 595 F. Supp., at 930. Jumu’ah therefore cannot be regarded as one of several essentially fungible religious practices. The ability to engage in other religious activities cannot obscure the fact that the denial at issue in this case is absolute: respondents are completely foreclosed from participating in the core ceremony that reflects their membership in a particular religious community. If a Catholic prisoner were prevented from attending Mass on Sunday, few would regard that deprivation as anything but absolute, even if the prisoner were afforded other opportunities to pray, to discuss the Catholic faith with others, and even to avoid eating meat on Friday if that were a preference. Prison officials in this case therefore cannot show that “ ‘other avenues’ remain available for the exercise of the asserted right.” Turner, ante, at 90 (quoting Jones v. North Carolina Prisoners' Union, 433 U. S. 119, 131 (1977)). Under the Court’s approach, as enunciated in Turner, the availability of other means of exercising the right in question O’LONE v. ESTATE OF SHABAZZ 361 342 Brennan, J., dissenting counsels considerable deference to prison officials. Ante, at 90. By the same token, the infliction of an absolute deprivation should require more than mere assertion that such a deprivation is necessary. In particular, “the existence of obvious, easy alternatives may be evidence that the regulation is not reasonable, but is an ‘exaggerated response’ to prison concerns.” Ibid. In this case, petitioners have not established the reasonableness of their policy, because they have provided only bare assertions that the proposals for accommodation offered by respondents are infeasible. As discussed below, the federal policy of permitting inmates in federal prisons to participate in Jumu’ah, as well as Leesburg’s own policy of permitting participation for several years, lends plausibility to respondents’ suggestion that their religious practice can be accommodated. In Turner, the Court found that the practices of the Federal Bureau of Prisons were relevant to the availability of reasonable alternatives to the policy under challenge.3 In upholding a ban on inmate-to-inmate mail, the Court noted that the Bureau had adopted “substantially similar restrictions.” Ante, at 93 (citing 28 CFR §540.17 (1986)). In finding that there were alternatives to a stringent restriction on the ability to marry, the Court observed that marriages by inmates in federal prisons were generally permitted absent a threat to security or public safety. See ante, at 97 (citing 28 CFR §551.10 (1986)). In the present case, it is therefore worth noting that Federal Bureau of Prisons regulations require the adjustment of work assignments to permit inmate participation in religious ceremonies, absent a threat to “security, safety, and good order.” 28 CFR §548.14 (1986). The Bureau’s Directive implementing the regulations on Religious Beliefs and Practices of Committed Offend 3See also Procunier v. Martinez, 416 U. S. 396, 414, n. 14 (1974) ( “While not necessarily controlling, the policies followed at other well-run institutions would be relevant to a determination of the need for a particular type of restriction”). 362 OCTOBER TERM, 1986 Brennan, J., dissenting 482 U. S. ers, 28 CFR §§548.10-548.15 (1986), states that, with respect to scheduling religious observances, “[t]he more central the religious activity is to the tenets of the inmate’s religious faith, the greater the presumption is for relieving the inmate from the institution program or assignment.” App. to Brief for Respondents 8a. Furthermore, the Chaplain Director of the Bureau has spoken directly to the issue of participation of Muslim inmates in Jumu’ah: “Provision is made, by policy, in all Bureau facilities for the observance of Jumu-ah by all inmates in general population who wish to keep this faith practice. The service is held each Friday afternoon in the general time frame that corresponds to the requirements of Islamic jurisprudence. . . . “Subject only to restraints of security and good order in the institution all routine and normal work assignments are suspended for the Islamic inmates to ensure freedom to attend such services. . . . “In those institutions where the outside work details contain Islamic inmates, they are permitted access to the inside of the institution to attend the Jumu-ah.” Id., at la. That Muslim inmates are able to participate in Jumu’ah throughout the entire federal prison system suggests that the practice is, under normal circumstances, compatible with the demands of prison administration.4 Indeed, the Leesburg State Prison permitted participation in this ceremony for five years, and experienced no threats to security or safety as a result. In light of both standard federal prison practice and Leesburg’s own past practice, a reasonableness test in this 4 See also American Correctional Association, Manual of Correctional Standards xxi (3d ed. 1966) (“Religion represents a rich resource in the moral and spiritual regeneration of mankind. Especially trained chaplains, religious instruction and counseling, together with adequate facilities for group worship of the inmate’s own choice, are essential elements in the program of a correctional institution”). O’LONE v. ESTATE OF SHABAZZ 363 342 Brennan, J., dissenting case demands at least minimal substantiation by prison officials that alternatives that would permit participation in Jumu’ah are infeasible.5 Under the standard articulated by the Court in Turner, this does not mean that petitioners are responsible for identifying and discrediting these alternatives; “prison officials do not have to set up and then shoot down every conceivable alternative method of accommodating the claimant’s constitutional complaint.” Ante, at 90-91. When prisoners themselves present alternatives, however, and when they fairly call into question official claims that these alternatives are infeasible, we must demand at least some evidence beyond mere assertion that the religious practice at issue cannot be accommodated. Examination of the alternatives proposed in this case indicates that prison officials have not provided such substantiation. Ill Respondents’ first proposal is that gang minimum prisoners be assigned to an alternative inside work detail on Friday, as they had been before the recent change in policy. Prison officials testified that the alternative work detail is now restricted to maximum security prisoners, and that they did not wish maximum and minimum security prisoners to 6 This is particularly true in light of the fact that Black Muslims in prisons in this country have not always been provided the same opportunities to practice their religion as members of other denominations. As the American Bar Association Section of Criminal Justice has observed: “The real problem comes not with facilities for religious service, but with attempts of prison officials to prevent or restrict certain religious movements within the prison. Chief among these movements has been the Black Muslims, whose lawsuits to compel recognition of their religion were the opening volley in prison litigation. See, e. g., Sostre v. McGinnis, 334 F. 2d 906 (2d Cir. 1964); Pierce v. LaVallee, 293 F. 2d 233 (2d Cir. 1961), on remand, 212 F. Supp. 865 (N. D. N. Y. 1962), aff’dper curiam, 319 F. 2d 844 (2d Cir. 1963); Bryant v. Wilkins, 258 N. Y. S. 2d 455, 45 Mise. 2d 923 (Sup. Ct. 1965).” ABA Committee on the Legal Status of Prisoners, Legal Status of Prisoners (Tent. Draft 1977), 14 Am. Crim. L. Rev. 377, 508 (1977). 364 OCTOBER TERM, 1986 Brennan, J., dissenting 482 U. S. mingle. Even the District Court had difficulty with this assertion, as it commented that “[t]he defendants did not explain why inmates of different security levels are not mixed on work assignments when otherwise they are mixed.” 595 F. Supp., at 932. The court found, nonetheless, that this alternative would be inconsistent with Standard 853’s mandate to move gang minimum inmates to outside work details. Ibid. This conclusion, however, neglects the fact that the very issue is whether the prison’s policy, of which Standard 853 is a part, should be administered so as to accommodate Muslim inmates. The policy itself cannot serve as a justification for its failure to provide reasonable accommodation. The record as it now stands thus does not establish that the Friday alternative work detail would create a problem for the institution. Respondents’ second proposal is that gang minimum inmates be assigned to work details inside the main building on a regular basis. While admitting that the prison used inside details in the kitchen, bakery, and tailor shop, officials stated that these jobs are reserved for the riskiest gang minimum inmates, for whom an outside job might be unwise. Ibid. Thus, concluded officials, it would be a bad idea to move these inmates outside to make room for Muslim gang minimum inmates. Respondents contend, however, that the prison’s own records indicate that there are a significant number of jobs inside the institution that could be performed by inmates posing a lesser security risk. This suggests that it might not be necessary for the riskier gang minimum inmates to be moved outside to make room for the less risky inmates. Officials provided no data on the number of inside jobs available, the number of high-risk gang minimum inmates performing them, the number of Muslim inmates that might seek inside positions, or the number of staff that would be necessary to monitor such an arrangement. Given the plausibility of respondents’ claim, prison officials should present at least O’LONE v. ESTATE OF SHABAZZ 365 342 Brennan, J., dissenting this information in substantiating their contention that inside assignments are infeasible. Third, respondents suggested that gang minimum inmates be assigned to Saturday or Sunday work details, which would allow them to make up any time lost by attending Jumu’ah on Friday. While prison officials admitted the existence of weekend work details, they stated that “[s]ince prison personnel are needed for other programs on weekends, the creation of additional weekend details would be a drain on scarce human resources.” Ibid. The record provides no indication, however, of the number of Muslims that would seek such a work detail, the current number of weekend details, or why it would be infeasible simply to reassign current Saturday or Sunday workers to Friday, rather than create additional details. The prison is able to arrange work schedules so that Jewish inmates may attend services on Saturday and Christian inmates may attend services on Sunday. Id., at 935. Despite the fact that virtually all inmates are housed in the main building over the weekend, so that the demand on the facility is greater than at any other time, the prison is able to provide sufficient staff coverage to permit Jewish and Christian inmates to participate in their central religious ceremonies. Given the prison’s duty to provide Muslims a “reasonable opportunity of pursuing [their] faith comparable to the opportunity afforded fellow prisoners who adhere to conventional religious precepts,” Cruz v. Beto, 405 U. S. 319, 322 (1972), prison officials should be required to provide more than mere assertions of the infeasibility of weekend details for Muslim inmates. Finally, respondents proposed that minimum security inmates living at the Farm be assigned to jobs either in the Farm building or in its immediate vicinity. Since Standard 853 permits such assignments for full minimum inmates, and since such inmates need not return to prison facilities through the main entrance, this would interfere neither with Standard 853 nor the concern underlying the no-retum pol 366 OCTOBER TERM, 1986 Brennan, J., dissenting 482 U. S. icy.6 Nonetheless, prison officials stated that such an arrangement might create an “affinity group” of Muslims representing a threat to prison authority. Officials pointed to no such problem in the five years in which Muslim inmates were permitted to assemble for Jumu’ah, and in which the alternative Friday work detail was in existence. Nor could they identify any threat resulting from the fact that during the month of Ramadan all Muslim prisoners participate in both breakfast and dinner at special times.7 Furthermore, there was no testimony that the concentration of Jewish or Christian inmates on work details or in religious services posed any type of “affinity group” threat. As the record now stands, prison officials have declared that a security risk is created by a grouping of Muslim inmates in the least dangerous security classification, but not by a grouping of maximum security inmates who are concentrated in a work detail inside the main building, and who are the only Muslims assured of participating in Jumu’ah. Surely, prison officials should be required to provide at least some substantiation for this facially implausible contention. Petitioners also maintained that the assignment of full minimum Muslim inmates to the Farm or its near vicinity might provoke resentment because of other inmates’ perception that Muslims were receiving special treatment. Officials pointed to no such perception during the period in which the alternative Friday detail was in existence, nor to any resentment of the fact that Muslims’ dietary preferences are accommodated and that Muslims are permitted to operate on a special schedule during the month of Ramadan. Nor do they identify any such problems created by the accommodation of 6 The Chief Deputy testified that there was no congestion problem with respect to the entrance to the full minimum security Farm building. Tr. 119. 7 Indeed, the Chief Deputy testified that full minimum Muslim inmates presented no greater threat to security or discipline than non-Muslim inmates. Id., at 138-139. O’LONE v. ESTATE OF SHABAZZ 367 342 Brennan, J., dissenting the religious preferences of inmates of other faiths. Once again, prison officials should be required at a minimum to identify the basis for their assertions. Despite the plausibility of the alternatives proposed by respondents in light of federal practice and the prison’s own past practice, officials have essentially provided mere pronouncements that such alternatives are not workable. If this Court is to take seriously its commitment to the principle that “[p]rison walls do not form a barrier separating prison inmates from the protections of the Constitution,” Turner, ante, at 84, it must demand more than this record provides to justify a Muslim inmate’s complete foreclosure from participation in the central religious service of the Muslim faith. IV That the record in this case contains little more than assertions is not surprising in light of the fact that the District Court proceeded on the basis of the approach set forth in St. Claire v. Cuyler, 634 F. 2d 109 (CA3 1980). That case held that mere “sincer[e]” and “arguably correct” testimony by prison officials is sufficient to demonstrate the need to limit prisoners’ exercise of constitutional rights. Id., at 114 (quoting Jones, 433 U. S., at 127). This Court in Turner, ante, p. 78, however, set forth a more systematic framework for analyzing challenges to prison regulations. Turner directed attention to two factors of particular relevance to this case: the degree of constitutional deprivation and the availability of reasonable alternatives. The respondents in this case have been absolutely foreclosed from participating in the central religious ceremony of their Muslim faith. At least a colorable claim that such a drastic policy is not necessary can be made in light of the ability of federal prisons to accommodate Muslim inmates, Leesburg’s own past practice of doing so, and the plausibility of the alternatives proposed by respondents. If the Court’s standard of review is to represent anything more than reflexive deference to prison officials, any 368 OCTOBER TERM, 1986 Brennan, J., dissenting 482 U. S. finding of reasonableness must rest on firmer ground than the record now presents. Incarceration by its nature denies a prisoner participation in the larger human community. To deny the opportunity to affirm membership in a spiritual community, however, may extinguish an inmate’s last source of hope for dignity and redemption.8 Such a denial requires more justification than mere assertion that any other course of action is infeasible. While I would prefer that this case be analyzed under the approach set out in Part I, supra, I would at a minimum remand to the District Court for an analysis of respondents’ claims in accordance with the standard enunciated by the Court in Turner and in this case. I therefore dissent. 8 As one federal court has stated: “Treatment that degrades the inmate, invades his privacy, and frustrates the ability to choose pursuits through which he can manifest himself and gain self-respect erodes the very foundations upon which he can prepare for a socially useful life. Religion in prison subserves the rehabilitative function by providing an area within which the inmate may reclaim his dignity and reassert his individuality.” Barnett v. Rodgers, 133 U. S. App. D. C. 296, 303, 410 F. 2d 995, 1002 (1969) (footnotes omitted). See also Comment, Religious Rights of the Incarcerated, 125 U. Pa. L. Rev. 812, 853-854 (1977) (“An inmate’s conscience is no less inviolable than that of an unconfined citizen, and a violation could well work an even greater harm upon the inmate, whose means of spiritual recovery are limited by the prison environment”). On the important role of religious commitment in penological rehabilitation, see generally Batson, Sociobiology and the Role of Religion in Promoting Prosocial Behavior: An Alternative View, 45 J. of Personality and Social Psychology 1380 (1983); Heintzelman & Fehr, Relationship Between Religious Orthodoxy and Three Personality Variables, 38 Psych. Reports 756 (1976). BOARD OF PARDONS v. ALLEN 369 Syllabus BOARD OF PARDONS et al. v. ALLEN et al. CERTIORARI TO THE UNITED STATES COURT OF APPEALS FOR THE NINTH CIRCUIT No. 86-461. Argued April 1, 1987—Decided June 9, 1987 In Greenholtz v. Nebraska Penal Inmates, 442 U. S. 1, the Court held that the mandatory language and structure of a Nebraska parole-release statute created an “expectancy of release,” a liberty interest entitled to protection under the Due Process Clause. The Montana statute at issue in this case provides that a prisoner eligible for parole “shall” be released when there is a reasonable probability that no detriment will result to him or the community, and specifies that parole shall be ordered for the best interests of society, and when the State Board of Pardons (Board) believes that the prisoner is able and willing to assume the obligations of a law-abiding citizen. After being denied parole, respondent prisoners filed a civil rights action against petitioners, the Board and its Chair, alleging that the Board denied them due process by failing to apply the statutorily mandated criteria in determining parole eligibility, and failing adequately to explain its reasons for parole denials. Although acknowledging that the case was controlled by principles established in Greenholtz, the District Court ruled that respondents were not entitled to due process protections in connection with their parole denials, concluding that, because the Board is required to make determinations with respect to the best interests of the community and the prisoner, its discretion is too broad to provide a prisoner with a liberty interest in parole release. The Court of Appeals reversed and remanded, finding the Montana statute virtually indistinguishable in structure and language from the statute considered in Greenholtz. Held: When scrutinized under the Greenholtz standards, the Montana statute clearly creates a liberty interest in parole release that is protected by the Due Process Clause of the Fourteenth Amendment. Although, as in Greenholtz, the release decision here is “necessarily . . . subjective and predictive” and the Board’s discretion “very broad,” nevertheless, the Montana statute, like the Nebraska statute, uses mandatory language (“shall”) to create a presumption that parole release will be granted when the designated findings are made. This presumption exists whether, as in Greenholtz, the statute mandates release “unless” the required findings are made, or whether, as here, release is necessary “when” or “if” the findings are made or is mandated “subject to” them. Moreover, the “substantive predicates” of release in Montana are similar 370 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. to those in Nebraska, since each statute requires consideration of the impact of release on both the prisoner and the community, of the prisoner’s ability to lead a law-abiding life, and of whether release will cause a “detriment to . . . the community,” and each statute vests the State’s parole board with equivalent discretion. That the Montana statute places significant limits on the Board’s discretion is further demonstrated by its replacement of an earlier statute which allowed absolute discretion, its specifying as its purpose the creation of restrictions on that discretion, and its addition of a provision authorizing judicial review of parole-release decisions. Pp. 373-381. 792 F. 2d 1404, affirmed. Brennan, J., delivered the opinion of the Court, in which White, Marshall, Blackmun, Powell, and Stevens, JJ., joined. O’Connor, J., filed a dissenting opinion, in which Rehnquist, C. J., and Scalia, J., joined, post, p. 381. Clay R. Smith, Assistant Attorney General of Montana, argued the cause for petitioners. With him on the briefs was Michael T. Greely, Attorney General. Stephen L. Pevar argued the cause for respondents. With him on the brief were Edward I. Koren, Elizabeth Alexander, and Alvin J. Bronstein.* Justice Brennan delivered the opinion of the Court. The question presented is whether respondents have a liberty interest in parole release that is protected under the Due Process Clause of the Fourteenth Amendment. I Respondents are George Allen and Dale Jacobsen, inmates of the Montana State Prison.1 In 1984, after their applica- * Randall D. Schmidt filed a brief for Eugene Newbury as amicus curiae urging affirmance. Dennis E. Curtis, Judith Resnik, William J. Genego, John L. Potten-ger, Jr., and Stephen Wizner filed a brief for the Yale Law School Legal Services Organization et al. ^oth respondents were released on parole after this suit was filed. 792 F. 2d 1404, 1408, n. 2 (1986). The action is not moot, however. In addition to requesting injunctive and declaratory relief, the complaint sought damages from Henry Burgess, Chair of the Board of Pardons, in BOARD OF PARDONS v. ALLEN 371 369 Opinion of the Court tions for parole were denied, they filed this action pursuant to 42 U. S. C. § 1983 on behalf of a class of all present and future inmates of the Montana State Prison who were or might become eligible for parole. Seeking declaratory and injunctive relief, as well as compensatory damages, the complaint charged the State Board of Pardons (Board) and its Chair with violations of the inmates’ civil rights. Specifically, respondents alleged that the Board does not apply the statutorily mandated criteria in determining inmates’ eligibility for parole, Complaint 51516—9, App. 5a-6a, and that the Board does not adequately explain its reasons for denial of parole, id., 559, 10, App. 6a.2 The District Court first acknowledged that the case was controlled by the principles established in this Court’s decision in Greenholtz v. Nebraska Penal Inmates, 442 U. S. 1 (1979). In Greenholtz the Court held that, despite the necessarily subjective and predictive nature of the parole-release decision, see id., at 12, state statutes may create liberty interests in parole release that are entitled to protection under the Due Process Clause. The Court concluded that the mandatory language and the structure of the Nebraska statute at issue in Greenholtz created an “expectancy of release,” which is a liberty interest entitled to such protection. Ibid. both his official and personal capacities. Because “this Court has not decided whether state parole officials enjoy absolute immunity as a matter of federal law,” Cleavinger v. Saxner, 474 U. S. 193, 200 (1985), “the validity of respondents’ claim for damages ... is not so insubstantial or so clearly foreclosed by prior decisions that this case may not proceed.” Memphis Light, Gas & Water Division v. Craft, 436 U. S. 1, 8-9 (1978). 2 Of the 350 individuals released from prison in Montana in 1985, 276 were conditionally released, the vast majority of them on parole; only 74 persons released had served their full sentences. See U. S. Dept, of Justice, Bureau of Justice Statistics, Prisoners in State and Federal Institutions on December 31, 1985, Table 43 (1985). Only 69 of 363 released in 1984 had discharged their full sentences. See U. S. Dept, of Justice, Bureau of Justice Standards, Prisoners in State and Federal Institutions on December 31, 1984, Table 13 (1984). 372 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. Although the District Court recognized that the Montana statute, like the Nebraska statute in Greenholtz, contained language mandating release under certain circumstances, it decided that respondents “were not entitled to due process protections in connection with the board’s denial of parole.” App. 17a. The court concluded that, because the Board is required to make determinations with respect to the best interest of the community and the prisoner, its discretion is too broad to provide a prisoner with a liberty interest in parole release. The Court of Appeals reversed. It compared the provisions of the Montana statute to those of the Nebraska statute in Greenholtz and found their structure and language virtually indistinguishable: “The Montana statute, like the Nebraska statute at issue in Greenholtz, uses mandatory language. It states that the Board ‘shall’ release a prisoner on parole when it determines release would not be harmful, unless specified conditions exist that would preclude parole. There is no doubt that it, like the Nebraska provision in Greenholtz, vests great discretion in the Board. Under both statutes the Board must make difficult and highly subjective decisions about risks of releasing inmates. However, the Board may not deny parole under either statute once it determines that harm is not probable.” 792 F. 2d 1404, 1406 (CA9 1986). The court thus held that respondents had stated a claim upon which relief could be granted, and remanded the case to the District Court for consideration of “the nature of the process which is due [respondents]” and “whether Montana’s present procedures accord that due process.” Id., at 1408. We granted certiorari, 479 U. S. 947 (1986), and now affirm. BOARD OF PARDONS v. ALLEN 373 369 Opinion of the Court II Greenholtz set forth two major holdings. The Court first held that the presence of a parole system by itself does not give rise to a constitutionally protected liberty interest in parole release.3 The Court also held, however, that the Nebraska statute did create an “expectation of parole” protected by the Due Process Clause. 442 U. S., at 11. To decide whether the Montana statute also gives rise to a constitutionally protected liberty interest, we scrutinize it under the standards set forth in Greenholtz. The Nebraska statute involved in Greenholtz provides as follows: ’There is far more to liberty than interests conferred by language in state statutes. See Hewitt v. Helms, 459 U. S. 460, 466 (1983); Connecticut Board of Pardons v. Dumschat, 452 U. S. 458, 468 (1981) (White, J., concurring). Four Members of this Court are of the view that the existence of a liberty interest in parole release is not solely a function of the wording of the governing statute. See Greenholtz v. Nebraska Penal Inmates, 442 U. S., at 18 (Powell, J., concurring in part and dissenting in part) (“I do not believe, however, that the application of the Due Process Clause to parole-release determinations depends upon the particular wording of the statute governing the deliberations of the parole board”); id., at 22 (Marshall, J., with Brennan and Stevens, JJ., dissenting in part) (“[A] ll prisoners potentially eligible for parole have a liberty interest of which they may not be deprived without due process, regardless of the particular statutory language that implements the parole system”). At stake in the parole-release decision is a return to freedom, albeit conditional freedom; liberty from bodily restraint is at the heart of the liberty protected by the Due Process Clause. Thus, inmates may have a liberty interest in parole release “derived solely from the existence of a system that permit[s] criminal offenders to serve their sentences on probation or parole.” Id., at 24-25 (Marshall, J., dissenting in part); see also id., at 19 (Powell, J., concurring in part and dissenting in part) (“[W]hen a state adopts a parole system that applies general standards of eligibility, prisoners justifiably expect that parole will be granted fairly and according to law whenever those standards are met”). We proceed, however, to apply the Court’s analysis in Greenholtz, because it too necessitates the conclusion that Montana inmates have a liberty interest in parole release. 374 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. “Whenever the Board of Parole considers the release of a committed offender who is eligible for release on parole, it shall order his release unless it is of the opinion that his release should be deferred because: “(a) There is a substantial risk that he will not conform to the conditions of parole; “(b) His release would depreciate the seriousness of his crime or promote disrespect for law; “(c) His release would have a substantially adverse effect on institutional discipline; or “(d) His continued correctional treatment, medical care, or vocational or other training in the facility will substantially enhance his capacity to lead a law-abiding life when released at a later date.” Neb. Rev. Stat. §83-1,114(1) (1981) (emphasis added). The statute also sets forth a list of 14 factors (including one catchall factor permitting the Nebraska Board to consider other information it deems relevant) that the Board must consider in reaching a decision. §§83-l,114(2)(a)-(n). In deciding that this statute created a constitutionally protected liberty interest, the Court found significant its mandatory language—the use of the word “shall”—and the presumption created—that parole release must be granted unless one of four designated justifications for deferral is found. See Greenholtz, 442 U. S., at 11-12.4 The Court recognized—indeed highlighted—that parolerelease decisions are inherently subjective and predictive, see id., at 12, but nonetheless found that Nebraska inmates 4 Cf. Hewitt v. Helms, supra, at 471-472. In that case the Court held that Pennsylvania’s administrative segregation statutes and regulations created a protected liberty interest in remaining in the general prison population. The Court relied on the State’s use of “language of an unmistakably mandatory character” and its specification of “substantive predicates” to confinement—“the need for control,” or “the threat of a serious disturbance.” BOARD OF PARDONS v. ALLEN 375 369 Opinion of the Court possessed a liberty interest in release. The Court observed that parole release is an equity-type judgment involving “a synthesis of record facts and personal observation filtered through the experience of the decisionmaker and leading to a predictive judgment as to what is best both for the individual inmate and for the community,” id., at 8,5 and acknowledged that the Nebraska statute, like most parole statutes, “vest[ed] very broad discretion in the Board,” id., at 13. Nevertheless, the Court rejected the Board’s argument “that a presumption [of release] would be created only if the statutory conditions for deferral were essentially factual, . . . rather than predictive.” Id., at 12. The Court thus held in Greenholtz that the presence of general or broad release criteria—delegating significant discretion to the decisionmaker—did not deprive the prisoner of the liberty interest in parole release created by the Nebraska statute. In essence, the Court made a distinction between two entirely distinct uses of the term discretion. In one sense of the word, an official has discretion when he or she “is simply not bound by standards set by the authority in question.” R. Dworkin, Taking Rights Seriously 32 (1977). In this sense, officials who have been told to parole whomever they wish have discretion. In Greenholtz, the Court determined that a scheme awarding officials this type of discretion does not create a liberty interest in parole release. But the term discretion may instead signify that “an official must use judgment in applying the standards set him [or her] by authority”; in other words, an official has discretion when the standards set by a statutory or regulatory scheme “cannot be applied mechanically.” Dworkin, supra, at 31, 32; see also ^d., at 69 (“[W]e say that a man has discretion if his duty is 5 See also Greenholtz, supra, at 10 (quoting Kadish, The Advocate and the Expert—Counsel in the Peno-Correctional Process, 45 Minn. L. Rev. 803, 813 (1961)) (“The decision turns on a ‘discretionary assessment of a multiplicity of imponderables, entailing primarily what a man is and what he may become rather than simply what he has done’ ”). 376 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. defined by standards that reasonable [people] can interpret in different ways”). The Court determined in Greenholtz that the presence of official discretion in this sense is not incompatible with the existence of a liberty interest in parole release when release is required after the Board determines (in its broad discretion) that the necessary prerequisites exist. Throughout this litigation, the Board’s arguments have had a single theme: that the holding of the Court of Appeals is inconsistent with our decision in Greenholtz.6 The Board is mistaken. The Montana statute, like the Nebraska statute, creates a liberty interest in parole release. It provides in pertinent part: “Prisoners eligible for parole. (1) Subject to the following restrictions, the board shall release on parole. . . any person confined in the Montana state prison or the women’s correction center . . . when in its opinion there is reasonable probability that the prisoner can be released without detriment to the prisoner or to the community[.] “(2) A parole shall be ordered only for the best interests of society and not as an award of clemency or a reduction of sentence or pardon. A prisoner shall be placed on parole only when the board believes that he is able and willing to fulfill the obligations of a law-abiding 6 See Pet. for Cert. 8 (“Reasons for Granting the Writ[:] The Court of Appeals’ Opinion Clearly Misconstrues Greenholtz”); Brief for Petitioners 10 (The conclusion that respondents had no protected liberty interest under the Montana statute “is consistent with, and required by, Greenholtz”); id., at 11 (“The Court of Appeals’ opinion deviates from Greenholtz, as well as from related decisions, and must therefore be reversed”); Reply Brief for Petitioners 3, n. 1 (“The parties . . . have not urged abandonment of Greenholtz, but rather have contended that it is consonant with their respective positions”). BOARD OF PARDONS v. ALLEN 377 369 Opinion of the Court citizen.” Mont. Code Ann. § 46-23-201 (1985) (emphasis added).7 Significantly, the Montana statute, like the Nebraska statute, uses mandatory language (“shall”)8 to “creat[e] a presumption that parole release will be granted” when the desig 7 This section also provides that “(a) No convict. . . may be paroled until he has served at least one-half of his full term, . . . except that a convict designated as a nondangerous offender . . . may be paroled after he has served one-quarter of his full term .... Any offender serving a time sentence may be paroled after he has served . . . 1772 years. “(b) No convict serving a life sentence may be paroled until he has served 30 years . . . .” Mont. Code Ann. §46-23-201 (1985). 8Cf. Grifaldo v. State, 182 Mont. 287, 596 P. 2d 847 (1979) (Section 46-18-404(1) provides that the sentencing court “shall” designate a defendant a nondangerous offender if either of two conditions are met; this mandatory language entitled the defendants to the designation and the paroleeligibility status that accompanies it). The Board argues that this Court is bound by statements of the Montana Supreme Court that parole is a privilege, a matter of grace, not of right. It is true that a State has no duty to establish a parole system or to provide for parole for all categories of convicted persons, see Greenholtz, 442 U. S., at 7, and that a State may place conditions on parole release; only in this sense is parole a privilege, not a right. None of the Montana cases cited by the Board decide whether parole release is mandatory for an eligible inmate upon a finding that the statutory prerequisites have been met. See Cavanaugh v. Crist, 189 Mont. 274, 615 P. 2d 890 (1980) (upholding the constitutionality of a statute authorizing a sentencing judge to forbid parole release of certain offenders); Lopez v. Crist, 176 Mont. 352, 578 P. 2d 312 (1978) (allowing the Board to keep a defendant whose parole had been wrongfully revoked in custody for up to 30 days to devise an acceptable new parole plan, because the Board has a statutory duty to impose and supervise conditions of parole); In re Frost, 146 Mont. 18, 403 P. 2d 612 (1965) (finding no blanket entitlement to parole after serving statutory minimum period); In re Hart, 145 Mont. 203, 399 P. 2d 984 (1965) (permitting the reincarceration of a defendant who ignored the conditions of his parole); State ex rel. Herman v. Powell, 139 Mont. 583, 367 P. 2d 553 (1961) (finding that the Board has no right to extinguish a sentence by paroling an individual on a subsequent sentence); Goffv. State, 139 Mont. 641, 367 P. 2d 557 (1961) (finding that the inmate was not denied equal protection because his codefendant was paroled before he was). 378 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. nated findings are made. Greenholtz, 442 U. S., at 12.9 See Statement of Assistant Attorney General of Montana, Tr. of Oral Arg. 6 (“under our statute once the Board of Pardons determines that the facts underlying a particular parole application are such that the release can occur consistently with the three criteria the statute specifies, then under our law the Board is required to order release”). We reject the argument that a statute that mandates release “unless” certain findings are made is different from a statute that mandates release “if,” “when,” or “subject to” such findings being made. Any such statute “creates a presumption that parole release will be granted.” Greenholtz, supra, at 12.10 9 The District Court found significant that, while the statute at issue in Greenholtz lists 14 factors that the Nebraska Board is obligated to consider in making the designated findings, the Montana statute “lists no factors required to be considered by the parole board.” App. 17a. In Montana, however, the Board considers these same 14 factors, which are set forth in the Board’s regulations. See Administrative Rules of Montana § 20.25.505 (1980). This Court, and the Courts of Appeals, see n. 10, infra, have recognized the relevance of regulations to a determination of whether a certain scheme gives rise to a liberty interest. See Hewitt v. Helms, 459 U. S., at 470-471; see also Connecticut Board of Pardons v. Dumschat, 452 U. S., at 467 (Brennan, J., concurring in judgment). In addition, the Montana statute does obligate the Board to consider certain information in making its parole-release decision. See Mont. Code Ann. § 46-23-202(1) (1985) (“[T]he board shall consider. . . the circumstances of his offense, his previous social history and criminal record, his conduct, employment, and attitude in prison, and the reports of any physical and mental examinations which have been made”). 10 As Justice White has pointed out, the Circuits have split on the question whether the absence of mandatory language creating a presumption of release precludes a finding that a statute or regulation creates a liberty interest. See Anderson v. Winsett, 449 U. S. 1093 (1981) (White, J., dissenting from denial of certiorari). But, as the following analysis of the decisions of the Courts of Appeals demonstrates, even under the most “restrictive interpretation of Greenholtz, ” Baumann v. Arizona Department of Corrections, 754 F. 2d 841, 844 (CA9 1985), courts have held that the presence of mandatory language in the statute gives rise to a liberty interest in parole release. The Montana statute, by its use of the word “shall BOARD OF PARDONS v. ALLEN 379 369 Opinion of the Court Moreover, the “substantive predicates,” see Hewitt v. Helms, 459 U. S. 460, 472 (1983), of parole release in Montana are similar to those in Nebraska. In both States, the and the phrase “[s]ubject to the following restrictions,” creates a liberty interest under this most restrictive interpretation. Courts of Appeals’ decisions since Greenholtz fall into four categories. When statutes or regulatory provisions are phrased in mandatory terms or explicitly create a presumption of release, courts find a liberty interest. See Parker v. Corrothers, 750 F. 2d 653, 661 (CA8 1984) (Arkansas regulation); Mayes v. Trammell, 751 F. 2d 175,178 (CA6 1984) (Tennessee Board of Parole Rule); 'Williams v. Missouri Board of Probation and Parole, 661 F. 2d 697, 698 (CA8 1981) (Missouri statute), cert, denied, 455 U. S. 993 (1982). Conversely, statutes or regulations that provide that a parole board “may” release an inmate on parole do not give rise to a protected liberty interest. See Dace v. Mickelson, 797 F. 2d 574, 576 (CA8 1986) (South Dakota statute); Parker v. Corrothers, supra, at 657 (Arkansas statute); Gale v. Moore, 763 F. 2d 341, 343 (CA8 1985) (amended Missouri statute); Dock v. Latimer, 729 F. 2d 1287,1288 (CAIO 1984) (Utah statute); Irving v. Thigpen, 732 F. 2d 1215, 1216 (CA5 1984) (Mississippi statute); Candelaria v. Griffin, 641 F. 2d 868, 869 (CAIO 1981) (New Mexico statute); Williams v. Briscoe, 641 F. 2d 274, 276 (CA5) (Texas statute), cert, denied, 454 U. S. 854 (1981); Schuemann v. Colorado State Board of Adult Parole, 624 F. 2d 172, 174 (CAIO 1980) (Colorado statute); Shirley v. Chestnut, 603 F. 2d 805, 806-807 (CAIO 1979) (Oklahoma statute); Wagner v. Gilligan, 609 F. 2d 866, 867 (CA6 1979) (Ohio statute). A third type of statute provides that an individual shall not be released unless or shall be released only when certain conditions are met; courts have divided on whether such statutes create a liberty interest. Most courts have found that such statutes set forth criteria that must be met before release, but that they do not require release if those findings are made. See Patten v. North Dakota Parole Board, 783 F. 2d 140, 142 (CA8 1986) (North Dakota statute); Huggins v. Isenbarger, 798 F. 2d 203, 204-205 (CA7 1986) (Indiana statute); Berard v. State of Vermont Parole Board, 730 F. 2d 71, 75 (CA21984) (Vermont statute); Thomas v. Sellers, 691F. 2d 487, 488 (CA11 1982) (Alabama statute); Staton v. Wainwright, 665 F. 2d 686, 688 (CA5 1982) (Florida statute); Jackson v. Reese, 608 F. 2d 159, 160 (CA5 1979) (Georgia statute); Boothe v. Hammock, 605 F. 2d 661, 664 (CA2 1979) (New York statute); but see United States ex rel. Scott v. Illinois Parole and Pardon Board, 669 F. 2d 1185,1188 (CA7 1982) (Illinois statute). Yet a fourth type of analysis finds a liberty interest when a statute or a regulatory parole-release scheme uses elaborate and explicit guidelines to struc- 380 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. Parole Board must assess the impact of release on both the prisoner and the community. A central concern of each is the prisoner’s ability “to lead a law-abiding life.” Neb. Rev. Stat. §83-l,114(l)(d) (1981); see §83-l,114(l)(a) (prisoner may not be released if there is “a substantial risk that he will not conform to the conditions of parole”); Mont. Code Ann. § 46-23-201(2) (1985) (prisoner must be released when, inter alia, it will cause no detriment to him or her and must not be released unless the prisoner is “able and willing to fulfill the obligations of a law-abiding citizen”). An interrelated concern of both statutes is whether the release can be achieved without “detriment to . . . the community.” Mont. Code Ann. § 46-23-201(1) (1985); see § 46-23-201(2) (prisoner must be released only “for the best interests of society”); see Neb. Rev. Stat. §83-l,114(l)(b) (1981) (prisoner must not be released if it “would depreciate the seriousness of his crime or promote disrespect for law”). The discretion left with the parole boards is equivalent in Montana and Nebraska. The legislative history further supports the conclusion that this statute places significant limits on the discretion of the Board. The statute was enacted in 1955, replacing a 1907 statute which had granted absolute discretion to the Board: “ Parole of prisoners in State Prison.—The Governor may recommend and the State Board of Prison Commissioners may parole any inmate of the State Prison, under such reasonable conditions and regulations as may be deemed expedient, and adopted by such state board.” Mont. Rev. Code §9573 (1907). The new statute made release mandatory upon certain findings and specified its purpose in its title: “An Act Creating a Board of Pardons and Prescribing the Appointment and Composition Thereof, With Power and Duty to Grant Pa- ture the exercise of discretion. See Dace v. Mickelson, supra, at 577-578 (South Dakota regulations); Green v. Black, 755 F. 2d 687, 688 (CA8 1985) (Missouri policy statement); Winsett v. McGinnes, 617 F. 2d 996, 1007 (CA3 1980) (Delaware regulations), cert, denied 449 U. S. 1093 (1981). BOARD OF PARDONS v. ALLEN 381 369 O’Connor, J., dissenting roles, Within Restrictions . . . .” Act of Mar. 3, 1955, 1955 Mont. Laws, ch. 153 (emphasis added). The new statute also added a provision for judicial review of the Board’s parole-release decisions, see Mont. Code Ann. §46-23-107 (1985), thus providing a further indication of a legislative intent to cabin the discretion of the Board. Here, as in Greenholtz, the release decision is “necessarily subjective . . . and predictive,” see 442 U. S., at 13; here, as in Greenholtz, the discretion of the Board is “very broad,” see ibid.; and here, as in Greenholtz, the Board shall release the inmate when the findings prerequisite to release are made. See supra, at 377-378 and 379-380. Thus, we find in the Montana statute, as in the Nebraska statute, a liberty interest protected by the Due Process Clause. The judgment of the Court of Appeals is Affirmed. Justice O’Connor, with whom The Chief Justice and Justice Scalia join, dissenting. Relying on semantics and ignoring altogether the sweeping discretion granted to the Board of Pardons by Montana law, the Court today concludes that respondents had a legitimate expectation of parole sufficient to give rise to an interest protected by procedural due process. Because I conclude that the discretion accorded the Board of Pardons belies any reasonable claim of entitlement to parole, I respectfully dissent. In Board of Regents v. Roth, 408 U. S. 564 (1972), this Court observed that to have a protected interest, one “clearly must have more than an abstract need or desire for it. He must have more than a unilateral expectation of it. He must, instead, have a legitimate claim of entitlement to it.” Id., at 577. Applying these principles, the Roth Court found that a teacher had no property interest in a renewal of his 1-year contract despite the fact that most teachers hired °n a year-to-year basis by the university were rehired. Id., at 578, n. 16. The Court concluded that the teacher had no legitimate entitlement to continued employment because the 382 OCTOBER TERM, 1986 O’Connor, J., dissenting 482 U. S. discretion of the university officials to renew or not renew such a contract was subject to no “cause” limitations. The Roth decision teaches that a mere expectation of a benefit—even if that expectation is supported by consistent government practice—is not sufficient to create an interest protected by procedural due process. Instead, the statute at issue must create an entitlement to the benefit before procedural due process rights are triggered. In my view, the distinction between an “entitlement” and a mere “expectancy” must necessarily depend on the degree to which the decisionmakers’ discretion is constrained by law. An individual simply has nothing more than a mere hope of receiving a benefit unless the decision to confer that benefit is in a real sense channeled by law. Because the crucial inquiry in determining the creation of a protected interest is whether a statutory entitlement is created, it cannot be sufficient merely to point to the existence of some “standard.” Instead, to give rise to a protected liberty interest, the statute must act to limit meaningfully the discretion of the decisionmakers. In the administrative law context we have long recognized that some purported standards “‘are drawn in such broad terms that in a given case there is no law to apply.’” Citizens to Preserve Overton Park v. Volpe, 401 U. S. 402, 410 (1971) (quoting S. Rep. No. 752, 79th Cong., 1st Sess., 26 (1945)). Accordingly, we have held that some agency action is committed to agency discretion within the meaning of the Administrative Procedure Act; as a result, agency action is not subject to judicial review if “no judicially manageable standards are available for judging how and when an agency should exercise its discretion.” Heckler v. Chaney, 470 U. S. 821, 830 (1985). It is no less critical in determining whether a statute creates a protected liberty interest to consider whether the statute includes standards that place real limits on decisionmaker discretion. Under our precedents, an entitlement is created by statute only if “particularized standards or criteria” constrain the BOARD OF PARDONS v. ALLEN 383 369 O’Connor, J., dissenting relevant decisionmakers. Connecticut Board of Pardons v. Dumschat, 452 U. S. 458, 467 (1981) (Brennan, J., concurring). In Meachum v. Fano, 427 U. S. 215 (1976), for example, we concluded that a state statute did not create a liberty interest in remaining in a particular penal facility because that statute “conferred no right on the prisoner to remain in the prison to which he was initially assigned, defeasible only upon proof of specific acts of misconduct.” Id., at 226 (emphasis added). The broad discretion granted prison officials to make transfer decisions negated any claim to the creation of a liberty interest: “A prisoner’s behavior may precipitate a transfer; and absent such behavior, perhaps transfer would not take place at all. But, as we have said, Massachusetts prison officials have the discretion to transfer prisoners for any number of reasons. Their discretion is not limited to instances of serious misconduct. As we understand it no legal interest or right of these respondents under Massachusetts law would have been violated by their transfer whether or not their misconduct had been proved in accordance with procedures that might be required by the Due Process Clause in other circumstances. Whatever expectation the prisoner may have in remaining at a particular prison so long as he behaves himself, it is too ephemeral and insubstantial to trigger procedural due process protections as long as prison officials have discretion to transfer him for whatever reason or for no reason at all.” Id., at 228. See also Olim v. Wakinekona, 461 U. S. 238, 249 (1983) (“[A] State creates a protected liberty interest by placing substantive limitations on official discretion”); Hewitt n. Helms, 459 U. S. 460, 472 (1983) (observing that the statute in question provided “explicitly mandatory language in connection with requiring specific substantive predicates”) (emphasis added); Montanye v. Haymes, 427 U. S. 236, 243 (1976) (no liberty interest in remaining in particular facility created by state 384 OCTOBER TERM, 1986 O’Connor, J., dissenting 482 U. S. law because state law did not limit transfer only on the occurrence of misconduct); Wolff n. McDonnell, 418 U. S. 539, 557 (1974) (“[T]he State itself has not only provided a statutory right to good time but also specifies that it is to be forfeited only for serious misbehavior”). Although paying lipservice to the principle that a statute creates an entitlement sufficient to trigger due process protections only when the decisionmakers’ discretion is limited by standards, the Court today utterly fails to consider whether the purported “standards” meaningfully constrain the discretion of state officials. Even a cursory examination of the Montana statute reveals that the Board of Pardons is subject to no real restraint, and that the standards are anything but “particularized.” In sharp contrast to the statute at issue in Wolff v. McDonnell, supra, and like the statutes at issue in Meachum v. Fano, supra, and Montanye n. Haymes, supra, the Montana statute does not require specific acts of misconduct before the Board may deny parole. Instead, the Board may deny parole when it determines: that there is not a “reasonable probability that the prisoner can be released without detriment to the prisoner or to the community,” Mont. Code Ann. §46-23-201(1) (1985); that parole is not in “the best interests of society,” §46-23-201(2); or that the Board believes that the prisoner is not “able and willing to fulfill the obligations of a law-abiding citizen.” Ibid. An appellate court reviewing the decision of the Board that the release of a prisoner would not be “in the best interests of society” or would be “detriment[al] ... to the community” would have little or no basis for taking issue with the judgment of the Board. These broadly framed standards essentially leave the decision whether or not to grant release on parole to the discretion of the Board, and therefore the statute simply fails to create a legitimate entitlement to release. See Herman, The New Liberty: The Procedural Due Process Rights of Prisoners and Others Under the Burger Court, 59 N. Y. U. L. Rev. 482, 550 (1984) (“A parole statute providing BOARD OF PARDONS v. ALLEN 385 369 O’Connor, J., dissenting that parole shall be granted unless the prospective parolee ‘poses a danger to society’ is not significantly different from one under which the parole board’s decisions are nonreview-able, since a court would be unlikely to reverse a parole board decision made under such a discretionary standard”). Admittedly, the statute at issue in Greenholtz v. Nebraska Penal Inmates, 442 U. S. 1 (1979), did not offer “particularized” standards, and did not significantly restrain the parole decision. Greenholtz is thus an aberration and should be reexamined and limited strictly to its facts. Nonetheless, in marked contrast to the Montana statute, at least the Nebraska statute limited to some degree the scope of the factors that parole officials could consider dispositive in granting or denying release on parole. While the Montana statute permits denial of parole when the prisoner’s release is not in “the best interests of society” or is “detriment[al] to the prisoner or to the community,” the Nebraska statute permits consideration only of four more focused factors: (1) whether “there is a substantial risk that [the prisoner] will not conform to the conditions of parole,” (2) whether the “release would depreciate the seriousness of [the] crime or promote disrespect for law,” (3) whether the “release would have a substantially adverse effect on institutional discipline,” and (4) whether the prisoner’s “continued correctional treatment, medical care, or vocational or other training in the facility will substantially enhance [the prisoner’s] capacity to lead a law-abiding life.” Neb. Rev. Stat. §83-1,114(1) (1981). Therefore, the result m this case is not compelled by Greenholtz, even assuming that case was correctly decided. In sum, the Court has abandoned the essential inquiry in determining whether a statute creates a liberty interest. Instead of requiring particularized standards that actually constrain the discretion of the relevant decisionmakers, the Court is satisfied simply by the presence of a purported standard.” Because I find the Court’s approach at odds with our liberty interest jurisprudence, I dissent. 386 OCTOBER TERM, 1986 Syllabus 482 U. S. CATERPILLAR INC. et al. v. WILLIAMS ET AL. CERTIORARI TO THE UNITED STATES COURT OF APPEALS FOR THE NINTH CIRCUIT No. 86-526. Argued April 21, 1987—Decided June 9, 1987 Caterpillar Tractor Co. (Caterpillar) hired respondents to work at its San Leandro, California, facility in positions covered by its collectivebargaining agreement with a union. Respondents eventually assumed management and other positions outside the bargaining unit, and allegedly were repeatedly assured by Caterpillar that, if the San Leandro facility ever closed, Caterpillar would employ them at other facilities. Subsequently, they were downgraded to unionized positions, but allegedly assured that the downgrades were temporary. However, Caterpillar later notified them that its San Leandro plant would close and that they would be laid off. Respondents then filed this action, based solely on state law, in a California state court, alleging that Caterpillar thereby breached their individual employment contracts. Caterpillar removed the action to Federal District Court, arguing that removal was proper because any individual employment contracts made with respondents were, as a matter of federal substantive labor law, merged into and superseded by the collective-bargaining agreement. Respondents denied that they alleged any federal claim and sought remand of the action to the state court. The Federal District Court held that removal was proper, and dismissed the case when respondents refused to amend the complaint to attempt to state a claim under § 301 of the Labor Management Relations Act, 1947, which confers federal jurisdiction as to suits for violations of collective-bargaining agreements. The Court of Appeals reversed, holding that the case was improperly removed. Held: Respondents’ state-law complaint for breach of the individual employment contracts is not removable to Federal District Court. Pp-391-399. (a) The presence or absence of federal-question jurisdiction that will support removal is governed by the “well-pleaded complaint rule,” under which federal jurisdiction exists only when a federal question is presented on the face of the properly pleaded complaint. Ordinarily, a case may not be removed on the basis of a federal defense, including the defense of pre-emption, even if the defense is anticipated in the complaint, and even if both parties concede that the federal defense is the only question truly at issue. However, under the “complete pre-emption doctrine,” which is a corollary to the well-pleaded complaint rule, once an area of state law has been completely pre-empted, any claim purportedly CATERPILLAR INC. v. WILLIAMS 387 386 Syllabus based on that pre-empted state law is considered, from its inception, a federal claim, and therefore arises under federal law. Pp. 391-394. (b) Respondents’ state-law contract claims are not “completely preempted” § 301 claims. Section 301 governs claims founded directly on rights created by collective-bargaining agreements and claims substantially dependent on analysis of such agreements. However, respondents alleged that Caterpillar breached individual employment contracts with them, and § 301 says nothing about the content or validity of such contracts. Although respondents, as bargaining unit employees at the time of the plant closing, could have brought suit under the collective agreement, they, as masters of the complaint, chose not to do so. Moreover, their complaint is not substantially dependent upon interpretation of the collective-bargaining agreement. Pp. 394-395. (c) J. I. Case Co. v. NLRB, 321 U. S. 332, does not support Caterpillar’s contention that when respondents returned to the collectivebargaining unit, their individual employment contracts were subsumed into, or eliminated by, the collective-bargaining agreement so as to be pre-empted by § 301. That decision does not stand for the general proposition that all individual employment contracts are inevitably superseded by a subsequent collective agreement. The fact that an employer may raise such a question in state court and might ultimately prove that the employee’s claims are pre-empted does not establish that they are removable. Pp. 395-398. (d) There is no merit to Caterpillar’s argument that § 301 pre-empts a state-law claim when the employer raises only a defense that requires a court to interpret or apply a collective-bargaining agreement, such as Caterpillar’s defense claiming that, in its collective-bargaining agreement, its unionized employees waived any pre-existing individual employment contract rights. The presence of a federal question, even a § 301 question, in a defensive argument does not overcome the paramount policies embodied in the well-pleaded complaint rule. Pp. 398-399. 786 F. 2d 928, affirmed. Brennan, J., delivered the opinion for a unanimous Court. Gerald D. Skoning argued the cause for petitioners. With him on the briefs were Charles C. Jackson, J. Stephen Poor, Theodore R. Johnson, and Nancy L. Snowden. Fritz Wollett argued the cause and filed a brief for respondents. * Briefs of amici curiae urging reversal were filed for the Chamber of Commerce of the United States by James C. Paras, Kingsley R. Browne, and Stephen A. Bokat; for the Equal Employment Advisory Council by 388 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. Justice Brennan delivered the opinion of the Court. The question for decision is whether respondents’ state-law complaint for breach of individual employment contracts is completely pre-empted by §301 of the Labor Management Relations Act, 1947 (LMRA), 61 Stat. 156, 29 U. S. C. § 185, and therefore removable to Federal District Court. I At various times between 1956 and 1968, Caterpillar Tractor Company (Caterpillar) hired respondents to work at its San Leandro, California, facility. Complaint IfH 10-26, App. to Pet. for Cert. (App.) A-40—A-42. Initially, each respondent filled a position covered by the collective-bargaining agreement between Caterpillar and Local Lodge No. 284, International Association of Machinists (Union). Each eventually became either a managerial or a weekly salaried employee, positions outside the coverage of the collective-bargaining agreement. Respondents held the latter positions for periods ranging from 3 to 15 years; all but two respondents served 8 years or more. App. A-97—A-98. Respondents allege that, “[d]uring the course of [their] employment, as management or weekly salaried employees,” Caterpillar made oral and written representations that “they could look forward to indefinite and lasting employment with the corporation and that they could count on the corporation to take care of them.” Complaint H1Ì27A, 27D, App. A-43. Robert E. Williams, Douglas S. McDowell, and Garen E. Dodge; for the Merchants and Manufacturers Association et al. by Charles G. Bakaly, Jr.; and for the Union Pacific Railroad Co. by I. Michael Greenberger and Mark B. Goodwin. Briefs of amici curiae urging affirmance were filed for the American Federation of Labor and Congress of Industrial Organizations by Marsha Berzon, David M. Silberman, and Laurence Gold; for the Employment Law Center of the San Francisco Legal Aid Society by Joan M. Graff, Robert Barnes, and John M. True; and for the Plaintiff Employment Lawyers Association by Paul H. Tobias. CATERPILLAR INC. v. WILLIAMS 389 386 Opinion of the Court More specifically, respondents claim that, “while serving Caterpillar as managers or weekly salaried employees, [they] were assured that if the San Leandro facility of Caterpillar ever closed, Caterpillar would provide employment opportunities for [them] at other facilities of Caterpillar, its subsidiaries, divisions, or related companies.” Id. 1I27F, App. A-48.1 Respondents maintain that these “promises were continually and repeatedly made,” and that they created “a total employment agreement wholly independent of the collective-bargaining agreement pertaining to hourly employees.” Id. 1T29, App. A-49.2 In reliance on these promises, respondents assert, they “continued to remain in Caterpillar’s employ rather than seeking other employment.” Id. If 31, App. A-49. Between May 1980 and January 1984, Caterpillar downgraded respondents from managerial and weekly salaried positions to hourly positions covered by the collectivebargaining agreement. Respondents allege that, at the time they were downgraded to unionized positions, Caterpillar supervisors orally assured them that the downgrades were temporary. Id. H27F, App. A-48. On December 15, 1983, Caterpillar notified respondents that its San Leandro plant would close and that they would be laid off. 1 The complaint also avers that Caterpillar “made clear ... its intention to employ [respondents] indefinitely by promoting them from entry level hourly positions to mid-level technical or weekly positions and to manage-nient positions,” and by giving respondents “favorable performance evaluations,” “payment increases and bonuses,” and “training ... to provide additional job security.” Complaint Uli 27A, 27B, 27C, App. A-43. Written representations with respect to job security were allegedly contained in employment memoranda, manuals, brochures, handbooks, and in Caterpillar’s “Code of Worldwide Business Conduct and Operating Principles.” Id. U27E, App. A-43—A-48. Under California law, an implied contract of employment may arise from a combination of factors, including longevity of service, commendations and promotions, oral and written assurances of stable and continuous employment, and an employer’s personnel practices. See Pugh v. See’s Candies, Inc., 116 Cal. App. 3d 311, 327-329 (1981); Cleary n. American Airlines, Inc., Ill Cal. App. 3d 443, 455-456 (1980). 390 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. On December 17, 1984, respondents filed an action based solely on state law in California state court, contending that Caterpillar “breached [its] employment agreement by notifying [respondents] that the San Leandro plant would be closed and subsequently advising [respondents] that they would be terminated” without regard to the individual employment contracts. Id. H32, App. A-49.3 Caterpillar then removed the action to federal court, arguing that removal was proper because any individual employment contracts made with respondents “were, as a matter of federal substantive labor law, merged into and superseded by the . . . collective bargaining agreements.” Petition for Removal, App. A-36. Respondents denied that they alleged any federal claim and immediately sought remand of the action to the state court. In an oral opinion, the District Court held that removal to federal court was proper, and dismissed the case when respondents refused to amend their complaint to attempt to state a claim under § 301 of the LMRA. App. A-4. The Court of Appeals for the Ninth Circuit reversed, holding that the case was improperly removed. 786 F. 2d 928 (1986). The court determined that respondents’ state-law claims were not grounded, either directly or indirectly, upon rights or liabilities created by the collective-bargaining agreement. Caterpillar’s claim that its collective-bargaining agreement with the Union superseded and extinguished all previous individual employment contracts alleged by respondents was deemed irrelevant. The court labeled this argument a “defensive allegation,” “raised to defeat the [respondents’] claims grounded in those independent contracts.” Id., at 936. Since respondents’ cause of action did not require interpretation or application of the collective-bargaining agree- 8 Respondents also brought state-law causes of action for breach of a covenant of good faith and fair dealing, intentional infliction of mental distress, and fraud. See Complaint HU 36-55, App. A-51—A-55. Petitioners principally rely on the breach-of-contract claim to support removal to federal court. CATERPILLAR INC. v. WILLIAMS 391 386 Opinion of the Court ment, the court concluded that the complaint did not arise under § 301 and was not removable to federal court.4 We granted certiorari, 479 U. S. 960 (1986), and now affirm. II A The Court recently set forth in some detail “[t]he century-old jurisdictional framework governing removal of federal 4 The Court of Appeals also appears to have held that a case may not be removed to federal court on the ground that it is completely pre-empted unless the federal cause of action relied upon provides the plaintiff with a remedy. For example, the court stated: “[A] state law cause of action has been ‘completely pre-empted’ when federal law both displaces and supplements the state law—that is, when federal law provides both a superseding remedy replacing the state cause of action and preempts that state law cause of action. These are two distinct inquiries, both of which must be satisfied to permit removal of an action to federal court.” 786 F. 2d, at 932 (emphasis in original; citations omitted). This analysis is squarely contradicted by our decision in Avco Corp. v. Machinists, 390 U. S. 557 (1968). We there held that a § 301 claim was properly removed to federal court although, at the time, the relief sought by the plaintiff could be obtained only in state court. We reasoned as follows: “The nature of the relief available after jurisdiction attaches is, of course, different from the question whether there is jurisdiction to adjudicate the controversy. . . . [T]he breadth or narrowness of the relief which may be granted under federal law in § 301 cases is a distinct question from whether the court has jurisdiction over the parties and the subject matter.” Id., at 561. Thus, although we affirm the Court of Appeals’ judgment, we reject its reasoning insofar as it is inconsistent with Avco. See also Franchise Tax Bd. of Cal. v. Construction Laborers Vacation Trust for Southern Cal., 463 U. S. 1, 23 (1983) (“The Court of Appeals held, [in Avco,] and we affirmed,. . . that the petitioner’s action ‘arose under’ § 301, and thus could be removed to federal court, although the petitioner had undoubtedly pleaded an adequate claim for relief under the state law of contracts and had sought a remedy available only under state law”) (emphasis in original; citation omitted). 392 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. question cases from state into federal courts,” Metropolitan Life Insurance Co. n. Taylor, 481 U. S. 58, 63 (1987) (citing Franchise Tax Board of Cal. v. Construction Laborers Vacation Trust for Southern Cal., 463 U. S. 1 (1983)), and we sketch only its outline here. Only state-court actions that originally could have been filed in federal court may be removed to federal court by the defendant.5 Absent diversity of citizenship, federal-question jurisdiction is required.6 The presence or absence of federal-question jurisdiction is governed by the “well-pleaded complaint rule,” which provides that federal jurisdiction exists only when a federal question is presented on the face of the plaintiff’s properly pleaded complaint. See Gully v. First National Bank, 299 U. S. 109, 112-113 (1936). The rule makes the plaintiff the master of the claim; he or she may avoid federal jurisdiction by exclusive reliance on state law.7 Ordinarily federal pre-emption is raised as a defense to the allegations in a plaintiff’s complaint. Before 1887, a federal defense such as pre-emption could provide a basis for removal, but, in that year, Congress amended the removal 6 Title 28 U. S. C. § 1441 provides: “(a) Except as otherwise expressly provided by Act of Congress, any civil action brought in a State court of which the district courts of the United States have original jurisdiction, may be removed by the defendant or the defendants, to the district court of the United States for the district and division embracing the place where such action is pending.” 6 Federal district courts have original jurisdiction over “all civil actions arising under the Constitution, laws, or treaties of the United States.” 28 U. S. C. § 1331. 7See The Fair v. Kohler Die & Specialty Co., 228 U. S. 22, 25 (1913) (“Of course, the party who brings a suit is master to decide what law he will rely upon”) (Holmes, J.); see also Merrell Dow Pharmaceuticals, Inc. v. Thompson, 478 U. S. 804, 809, n. 6 (1986) (“Jurisdiction may not be sustained on a theory that the plaintiff has not advanced”); Great North R. Co. v. Alexander, 246 U. S. 276, 282 (1918) (“[T]he plaintiff may by the allegations of his complaint determine the status with respect to removability of a case”). CATERPILLAR INC. v. WILLIAMS 393 386 Opinion of the Court statute. We interpret that amendment to authorize removal only where original federal jurisdiction exists. See Act of Mar. 3, 1887, ch. 373, 24 Stat. 552, as amended by Act of Aug. 13, 1888, ch. 866, 25 Stat. 433. Thus, it is now settled law that a case may not be removed to federal court on the basis of a federal defense, including the defense of preemption, even if the defense is anticipated in the plaintiff’s complaint, and even if both parties concede that the federal defense is the only question truly at issue. See Franchise Tax Board, 463 U. S., at 12. There does exist, however, an “independent corollary” to the well-pleaded complaint rule, id., at 22, known as the “complete pre-emption” doctrine. On occasion, the Court has concluded that the pre-emptive force of a statute is so “extraordinary” that it “converts an ordinary state commonlaw complaint into one stating a federal claim for purposes of the well-pleaded complaint rule.” Metropolitan Life Insurance Co., supra, at 65.8 Once an area of state law has been completely pre-empted, any claim purportedly based on that pre-empted state law is considered, from its inception, a federal claim, and therefore arises under federal law. See Franchise Tax Board, supra, at 24 (“[I]f a federal cause of action completely pre-empts a state cause of action any complaint that comes within the scope of the federal cause of action necessarily ‘arises under’ federal law”). The complete pre-emption corollary to the well-pleaded complaint rule is applied primarily in cases raising claims preempted by § 301 of the LMRA. Section 301 provides: 8 See, e. g., Metropolitan Life Insurance Co. v. Taylor (state contract and tort claims completely pre-empted by §§ 502(a)(1)(B) and 502(f) of the Employee Retirement Income Security Act of 1974, 88 Stat. 891, 892); Oneida Indian Nation v. County of Oneida, 414 U. S. 661, 675 (1974) (state-law complaint that alleges a present right to possession of Indian tribal lands necessarily “asserts a present right to possession under federal law,” and is thus completely pre-empted and arises under federal law); Avco, supra (discussed infra). 394 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. “Suits for violation of contracts between an employer and a labor organization representing employees in an industry affecting commerce as defined in this chapter, or between any such labor organizations, may be brought in any district court of the United States having jurisdiction of the parties, without respect of the amount in controversy or without regard to the citizenship of the parties.” 29 U. S. C. § 185(a). In Avco Corp. v. Machinists, the Court of Appeals decided that “[s]tate law does not exist as an independent source of private rights to enforce collective bargaining contracts.” 376 F. 2d 337, 340 (CA6 1967), aff’d, 390 U. S. 557 (1968). In affirming, we held that, when “[t]he heart of the [state-law] complaint [is] a . . . clause in the collective bargaining agreement,” id., at 558, that complaint arises under federal law: “[T]he pre-emptive force of §301 is so powerful as to displace entirely any state cause of action ‘for violation of contracts between an employer and a labor organization.’ Any such suit is purely a creature of federal law, notwithstanding the fact that state law would provide a cause of action in the absence of §301.” Franchise Tax Board, supra, at 23. B Caterpillar asserts that respondents’ state-law contract claims are in reality completely pre-empted §301 claims, which therefore arise under federal law. We disagree. Section 301 governs claims founded directly on rights created by collective-bargaining agreements, and also claims “substantially dependent on analysis of a collective-bargaining agreement.” Electrical Workers n. Hechler, 481 U. S. 851, 859, n. 3 (1987); see also Allis-Chalmers Corp. v. Lueck, 471 U. S. 202, 220 (1985). Respondents allege that Caterpillar has entered into and breached individual employment contracts with them. Section 301 says nothing about the content or validity of individual employment contracts. It is CATERPILLAR INC. v. WILLIAMS 395 386 Opinion of the Court true that respondents, bargaining unit members at the time of the plant closing, possessed substantial rights under the collective agreement, and could have brought suit under § 301. As masters of the complaint, however, they chose not to do so. Moreover, contrary to Caterpillar’s assertion, see Reply Brief for Petitioner 10, respondents’ complaint is not substantially dependent upon interpretation of the collectivebargaining agreement. It does not rely upon the collective agreement indirectly, nor does it address the relationship between the individual contracts and the collective agreement.9 As the Court has stated, “it would be inconsistent with congressional intent under [§301] to pre-empt state rules that proscribe conduct, or establish rights and obligations, independent of a labor contract.” Allis-Chalmers Corp., supra, at 212. Caterpillar next relies on this Court’s decision in J. I. Case Co. v. NLRB, 321 U. S. 332 (1944), arguing that when respondents returned to the collective-bargaining unit, their 9 Caterpillar contends for example, that, under California law governing implied contracts of employment, the state court will have to examine the collective-bargaining agreement as part of its evaluation of the “totality of the parties’ relationship.” Brief for Petitioners 35, n. 24. But respondents rely on contractual agreements made while they were in managerial or weekly salaried positions—agreements in which the collective-bargaining agreement played no part. The irrelevance of the collective-bargaining agreement to these individual employment contracts is illustrated by the District Court’s disposition of the claim of Mr. Chambers, who was not in the bargaining unit at the time he was laid off. His claim was deemed solely a matter of state law (and, by implication, not intertwined with the collective-bargaining agreement), and thus was remanded to state court. See App. A-22. Moreover, it is unclear whether an examination of the collective-bargaining agreement is truly required by California law. See Youngman v. Nevada Irrigation Dist., 70 Cal. 2d 240, 246-247, 449 P. 2d 462, 466 (1969) (“In pleading a cause of action on an agreement implied from conduct, only the facts from which the promise is implied must be alleged”); 2 J. Chadbourn, H. Grossman, & A. Van Alstyne, California Pleading § 1011, p. 159 (1961) (same). 396 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. individual employment agreements were subsumed into, or eliminated by, the collective-bargaining agreement. Thus, Caterpillar contends, respondents’ claims under their individual contracts actually are claims under the collective agreement and pre-empted by §301. Caterpillar is mistaken. First, J. I. Case does not stand for the proposition that all individual employment contracts are subsumed into, or eliminated by, the collective-bargaining agreement. In fact, the Court there held: “Individual contracts cannot subtract from collective ones, and whether under some circumstances they may add to them in matters covered by the collective bargain, we leave to be determined by appropriate forums under the law of contracts applicable, and to the Labor Board if they constitute unfair labor practices.” 321 U. S., at 339. Thus, individual employment contracts are not inevitably superseded by any subsequent collective agreement covering an individual employee, and claims based upon them may arise under state law. Caterpillar’s basic error is its failure to recognize that a plaintiff covered by a collective-bargaining agreement is permitted to assert legal rights independent of that agreement, including state-law contract rights, so long as the contract relied upon is not a collective-bargaining agreement. See Allis-Chalmers Corp., supra.10 Caterpillar im- 10 Section 301 does not, as Caterpillar suggests, require that all “employment-related matters involving unionized employees” be resolved through collective bargaining and thus be governed by a federal common law created by § 301. Brief for Petitioners 26. The Court has stated that “not every dispute concerning employment, or tangentially involving a provision of a collective-bargaining agreement, is pre-empted by § 301 or other provisions of the federal labor law.” Allis-Chalmers Corp. v. Lueck, 471 U. S. 202, 211 (1985). Claims bearing no relationship to a collectivebargaining agreement beyond the fact that they are asserted by an individual covered by such an agreement are simply not pre-empted by §301. See also Franchise Tax Board, 463 U. S., at 25, n. 28 (“[E]ven under § 301 we have never intimated that any action merely relating to a contract CATERPILLAR INC. v. WILLIAMS 397 386 Opinion of the Court permissibly attempts to create the prerequisites to removal by ignoring the set of facts (i. e., the individual employment contracts) presented by respondents, along with their legal characterization of those facts, and arguing that there are different facts respondents might have alleged that would have constituted a federal claim. In sum, Caterpillar does not seek to point out that the contract relied upon by respondents is in fact a collective agreement; rather it attempts to justify removal on the basis of facts not alleged in the complaint. The “artful pleading” doctrine cannot be invoked in such circumstances.11 Second, if an employer wishes to dispute the continued legality or viability of a pre-existing individual employment contract because an employee has taken a position covered by a collective agreement, it may raise this question in state court. The employer may argue that the individual employment contract has been pre-empted due to the principle of exclusive representation in §9(a) of the National Labor Relations Act (NLRA), 29 U. S. C. § 159(a). See Machinists v. Wisconsin Employment Relations Common, 427 U. S. 132, 146 (1976) (quoting Teamsters v. Morton, 377 U. S. 252, 260 (1964)) (NLRA pre-empts state law that “‘upset[s] the balance of power between labor and management expressed in our national labor policy’”). Or the employer may contend that enforcement of the individual employment contract arguably would constitute an unfair labor practice under the NLRA, and is therefore pre-empted. See San Diego Build- within the coverage of § 301 arises exclusively under that section. For instance, a state battery suit growing out of a violent strike would not arise under §301 simply because the strike may have been a violation of an employer-union contract”). 11 Cf. Federated Department Stores, Inc. v. Moitie, 452 U. S. 394, 410, n. 6 (1981) (Brennan, J., dissenting) (Although “occasionally the removal court will seek to determine whether the real nature of the claim is federal, regardless of plaintiff’s characterization, . . . most of them correctly confine this practice to areas of the law pre-empted by federal substantive law”) (internal quotations omitted). 398 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. ing Trades Council v. Garmon, 359 U. S. 236 (1959) (state law that infringes upon the National Labor Relations Board’s primary jurisdiction over unfair labor practice charges is preempted). The fact that a defendant might ultimately prove that a plaintiff’s claims are pre-empted under the NLRA does not establish that they are removable to federal court.12 Finally, Caterpillar argues that § 301 pre-empts a state-law claim even when the employer raises only a defense that requires a court to interpret or apply a collective-bargaining agreement. Caterpillar asserts such a defense claiming that, in its collective-bargaining agreement, its unionized employees waived any pre-existing individual employment contract rights.13 It is true that when a defense to a state claim is based on the terms of a collective-bargaining agreement, the state court will have to interpret that agreement to decide whether the state claim survives. But the presence of a federal question, even a § 301 question, in a defensive argument does not overcome the paramount policies embodied in the well-pleaded complaint rule—that the plaintiff is the master of the 12 Caterpillar also contends that enforcement of individual employment contracts negotiated with employees covered by the collective-bargaining agreement would violate §9(a) of the NLRA, 49 Stat. 453, 29 U. S. C. § 159(a), because, with exceptions not here relevant, it is an unfair labor practice “for the employer to disregard the bargaining representative by negotiating with individual employees, whether a majority or a minority, with respect to wages, hours, and working conditions.” Medo Photo Supply Corp. v. NLRB, 321 U. S. 678, 684 (1944). Even if these individual employment contracts were negotiated with respondents while the latter were covered by a collective agreement (which is disputed), this fact is irrelevant to the removal question. For reasons similar to those stated in text, see supra, at 394-397, and this page, respondents’ state-law claims might be pre-empted by the NLRA, but they would not be transformed into claims arising under federal law. 13 We intimate no view on the merits of this or any of the pre-emption arguments discussed above. These are questions that must be addressed in the first instance by the state court in which respondents filed their claims. CATERPILLAR INC. v. WILLIAMS 399 386 Opinion of the Court complaint, that a federal question must appear on the face of the complaint, and that the plaintiff may, by eschewing claims based on federal law, choose to have the cause heard in state court. When a plaintiff invokes a right created by a collective-bargaining agreement, the plaintiff has chosen to plead what we have held must be regarded as a federal claim, and removal is at the defendant’s option. But a defendant cannot, merely by injecting a federal question into an action that asserts what is plainly a state-law claim, transform the action into one arising under federal law, thereby selecting the forum in which the claim shall be litigated.14 If a defendant could do so, the plaintiff would be master of nothing. Congress has long since decided that federal defenses do not provide a basis for removal. See supra, at 392, and n. 5, 392-393.15 III Respondents’ claims do not arise under federal law and therefore may not be removed to federal court. The judgment of the Court of Appeals is Affirmed. 14See, e. g., Cook v. Georgetown Steel Corp., 770 F. 2d 1272 (CA4 1985); Medlin v. Boeing Vertol Co., 620 F. 2d 957 (CA3 1980). 15 Caterpillar contends that the Court of Appeals’ decision offends the paramount national labor policy of referring disputes to arbitration, since its collective-bargaining agreement with the Union contains an arbitration cause. Brief for Petitioners 36. This argument presumes that respondents’ claims are arbitrable, when, in fact, they are alleged to grow out of individual employment contracts to which the grievance-arbitration procedures in the collective-bargaining agreement have no application. 400 OCTOBER TERM, 1986 Syllabus 482 U. S. CALIFORNIA v. SUPERIOR COURT OF CALIFORNIA, SAN BERNARDINO COUNTY (SMOLIN et al., REAL PARTIES IN INTEREST) CERTIORARI TO THE SUPREME COURT OF CALIFORNIA No. 86-381. Argued April 22, 1987—Decided June 9, 1987 The Extradition Act, which implements the Extradition Clause of Article IV, requires an asylum State to give up to a demanding State a fugitive against whom a properly certified indictment has been lodged. After a California custody decree was modified to give Richard Smolin sole custody of his minor children and he secured a California warrant to obtain custody, he and his father picked up the children in Louisiana, where they were living with their mother. The mother then swore out an affidavit charging the Smolins with kidnaping, on the basis of which an information was filed charging them with violating a Louisiana statute prohibiting a parent’s intentional taking of his own child from any person to whom custody has been awarded by any state court of competent jurisdiction. After the Governor of Louisiana formally notified the Governor of California of the charges and demanded that the Smolins be delivered up for trial, the California Superior Court granted them a writ of habeas corpus to block the extradition warrants against them. Taking judicial notice of the California custody orders, the court concluded that the Smolins were not substantially charged with crime under Louisiana law. Although the California Court of Appeal then issued a writ of mandate on the ground that the Superior Court had abused its discretion, the State Supreme Court reversed, finding that the California custody decrees were properly considered by the Superior Court, and that, under the full faith and credit provisions of the federal Parental Kidnaping Prevention Act of 1980, those decrees conclusively established that Richard Smolin was the childrens’ lawful custodian at the time he took them. The court ruled that the Smolins had not been substantially charged with a crime, since, under Louisiana law, the lawful custodian of children cannot be guilty of kidnaping them. Held: The Extradition Act prohibits the California Supreme Court from refusing to permit extradition. The language, history, and subsequent construction of the Act establish that extradition is meant to be a summary procedure, and that the asylum State’s courts may do no more than ascertain whether (a) the extradition documents on their face are in order; (b) the petitioner has been charged with a crime in the demanding CALIFORNIA v. SUPERIOR COURT OF CALIFORNIA 401 400 Syllabus State; (c) the petitioner is the person named in the request for extradition; and (d) the petitioner is a fugitive. Here, the only such inquiry in doubt is whether the Smolins have been charged with a crime in Louisiana, which question must be answered in the affirmative since the information charging them is in proper form, and they do not dispute that the wife’s affidavit, and documents incorporated by reference therein, set forth facts that clearly satisfy each element of the crime defined in the state parental kidnaping statute. Their contention that the requirement of Roberts v. Reilly, 116 U. S. 80, 95, that the person demanded be “substantially charged” permits an inquiry by the asylum State into whether the charging instrument is sufficient to withstand a generalized motion to dismiss or common-law demurrer is without merit. To the contrary, the asylum State may do no more than ascertain whether the requisites of the Extradition Act have been met, and may not entertain defenses or determine the guilt or innocence of the charged party. Thus, it is for the Louisiana courts to determine whether the wife’s affidavit is fraudulent, whether the California custody decrees establish Richard Smolin as the children’s lawful custodian under the full faith and credit provision of the federal Parental Kidnaping Prevention Act, and whether the Smolins were, accordingly, not guilty of violating the Louisiana statute. Pp. 405-412. 41 Cal. 3d 758, 716 P. 2d 991, reversed. O’Connor, J., delivered the opinion of the Court, in which Rehnquist, C. J., and White, Marshall, Blackmun, Powell, and Scalia, JJ., joined. Stevens, J., filed a dissenting opinion, in which Brennan J., joined, post, p. 412. J. Robert Jibson, Supervising Deputy Attorney General of California, argued the cause for petitioner. With him on the brief were John K. Van de Kamp, Attorney General, and Steve White, Chief Assistant Attorney General. Dennis P. Riordan argued the cause for respondent. With him on the brief was Karen L. Snell* *A brief of amici curiae urging reversal was filed for the State of Alaska et al. by James E. Tierney, Attorney General of Maine, William R. Stokes, Assistant Attorney General, and by the Attorneys General for their respective States as follows: Harold M. Brown of Alaska, Corinne K. A. Watanabe of Hawaii, Jim Jones of Idaho, Linley E. Pearson of Indiana, Robert T. Stephan of Kansas, David L. Armstrong of Kentucky, William J- Guste, Jr., of Louisiana, Mike Greely of Montana, Brian McKay of Nevada, Stephen E. Merrill of New Hampshire, Michael Turpen of Okla 402 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. Justice O’Connor delivered the opinion of the Court. At issue in this case are the limits imposed by federal law upon state court habeas corpus proceedings challenging an extradition warrant. I Richard and Judith Smolin were divorced in California in 1978. Sole custody of their two children, Jennifer and Jamie, was awarded to Judith Smolin, subject to reasonable visitation rights for Richard. Until November 1979, all the parties remained in San Bernardino County, California, and Richard apparently paid his child support and exercised his visitation rights without serious incident. In August 1979, however, Judith married James Pope, and in November, Mr. Pope’s work required that the family relocate to Oregon. When the Popes moved without informing Richard, the battle over the custody of the minor children began in earnest. It is unnecessary to recite in detail all that ensued. Richard alleged, and the California courts later found, that the Popes deliberately attempted to defeat Richard’s visitation rights and to preclude him from forming a meaningful relationship with his children in the course of their succeeding relocations from Oregon to Texas to Louisiana. On February 13, 1981, the Popes obtained a decree from a Texas court granting full faith and credit to the original California order awarding sole custody to Judith. Richard was served but did not appear in the Texas proceeding. Before the Texas decree was issued, however, Richard sought and obtained in California Superior Court modification of the underlying California decree, awarding joint custody to Richard and Judith. Though properly served, the Popes did not appear in these homa, LeRoy S. Zimmerman of Pennsylvania, Hector Rivera Cruz of Puerto Rico, Mark V. Meierhenry of South Dakota, J. Michael Cody of Tennessee, Jim Mattox of Texas, David L. Wilkinson of Utah, Leroy A. Mercer of The Virgin Islands, and Archie G. McClintock of Wyoming. Ephraim Margolin filed a brief for California Attorneys for Criminal Justice et al. as amici curiae urging affirmance. CALIFORNIA v. SUPERIOR COURT OF CALIFORNIA 403 400 Opinion of the Court California proceedings; and, though served with the modification order, the Popes neither complied with its terms, nor notified the Texas court of its existence. On January 9, 1981, Richard instituted an action in California Superior Court to find Judith in contempt and to again modify the custody decree to give him sole custody. In February 1981, sole custody was granted to Richard by the California court, subject to reasonable visitation rights for Judith. This order also was ignored by the Popes, apparently acting on the advice of counsel that the California courts no longer had jurisdiction over the matter. Richard did not in fact obtain physical custody for over two years. When he finally located the Popes in Louisiana, they began an adoption proceeding, later described by the California courts as “verging on the fraudulent,” to sever Richard’s legal tie to Jennifer and Jamie. App. 51. After securing a California warrant to obtain custody of the children on February 27, 1984, Richard and his father, Gerard Smolin, resorted to self-help. On March 9, 1984, they picked up Jennifer and Jamie as they were waiting for their school bus in Slidell, Louisiana, and brought them back to California. On April 11, 1984, the Popes submitted to the jurisdiction of the California Superior Court and instituted an action to modify the 1981 order granting Richard sole custody. 41 Cal. 3d 758, 764, n. 4, 716 P. 2d 991, 994, n. 4 (1986). Those proceedings are apparently still pending before the California courts. Meanwhile, the Popes raised the stakes by instituting a criminal action against Richard and Gerard Smolin in Louisiana. On April 30, 1984, after the Popes instituted modification proceedings in California, Judith Pope swore out an affidavit charging Richard and Gerard Smolin with kidnaping Jennifer and Jamie from her custody and asserting that they had acted “without authority to remove children from [her] custody.” App. B to Pet. for Cert. 6. On the basis of this affidavit, the Assistant District Attorney for the 22d Judicial District of Louisiana, William Alford, Jr., filed an informa 404 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. tion charging Richard and Gerard Smolin each with two counts of violating La. Rev. Stat. Ann. § 14:45 (West 1986), the Louisiana kidnaping statute. On June 14, 1984, the Governor of Louisiana formally notified the Governor of California that Richard and Gerard Smolin were charged with “simple kidnaping” in Louisiana and demanded that they be delivered up for trial. 41 Cal. 3d, at 763, 716 P. 2d, at 993-994. In early August 1984, the Smolins petitioned in the California Superior Court for a writ of habeas corpus to block the anticipated extradition warrants. On August 17, 1984, the anticipated warrants issued and on August 24, 1984, the Superior Court orally granted a writ of habeas corpus after taking judicial notice of the various custody orders that had been issued. The court concluded “that the findings in the family law case adequately demonstrate that, in fact, the process initiated by Mrs. Pope in Louisiana and her declarations and affidavits were totally insufficient to establish any basis for rights of either herself personally or for the State ... of Louisiana.” App. C to Pet. for Cert. 5. California then sought a writ of mandate in the California Court of Appeal on the ground that the Superior Court had abused its discretion in blocking extradition. The Court of Appeal reluctantly issued the writ: “Although we abhor Judy’s apparent willingness to take advantage of our federal system to further this custody battle, and are sympathetic to [the Smolins’] position, we must conclude that their arguments are irrelevant to the only issue a court in the asylum state may properly address: are the documents on their face in order.” App. B to Pet. for Cert. 16. A divided California Supreme Court reversed. The majority interpreted the Superior Court’s finding to be that the Smolins were not substantially charged with a crime. It found that the California custody decrees were properly con- CALIFORNIA v. SUPERIOR COURT OF CALIFORNIA 405 400 Opinion of the Court sidered by the Superior Court, and that its conclusion that the Smolins were not substantially charged was correct. Under the full faith and credit provisions of the federal Parental Kidnaping Prevention Act of 1980, 28 U. S. C. § 1738A, the majority determined that those decrees conclusively established that Richard Smolin was the lawful custodian of the children at the time that they were taken from Louisiana to California.* Finally, the court found that, under Louisiana law, the lawful custodian cannot be guilty of kidnaping children in his custody. State v. Elliott, 171 La. 306, 311, 131 So. 28, 30 (1930). We granted certiorari, 479 U. S. 982 (1986), to consider whether the Extradition Clause, Art. IV, §2, cl. 2, and the Extradition Act, 18 U. S. C. §3182, prevent the California Supreme Court from refusing to permit extradition on these grounds. II The Federal Constitution places certain limits on the sovereign powers of the States, limits that are an essential part of the Framers’ conception of national identity and Union. One such limit is found in Art. IV, § 2, cl. 2, the Extradition Clause: “A person charged in any State with Treason, Felony, or other Crime, who shall flee from Justice, and be found in another State, shall on Demand of the executive Author *The California Supreme Court found that under the Parental Kidnaping Prevention Act, California had exclusive modification jurisdiction over the original custody decree. 41 Cal. 3d 758, 770, 716 P. 2d 991, 999 (1986). See 28 U. S. C. § 1738A(d) (“The jurisdiction of a court of a State which has made a child custody determination consistently with the provisions of this section continues as long as [such court has jurisdiction under the law of such State] and such State remains the residence of the child or any contestant”); 28 U. S. C. § 1738A(f) (“A court of a State may modify a determination of the custody of the same child made by a court of another State, if— • • • (2) the court of the other State no longer has jurisdiction, or it has declined to exercise such jurisdiction to modify such determination”). 406 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. ity of the State from which he fled, be delivered up, to be removed to the State having Jurisdiction of the Crime.” The obvious objective of the Extradition Clause is that no State should become a safe haven for the fugitives from a sister State’s criminal justice system. As this Court noted in its first opportunity to construe the Extradition Clause: “[T]he statesmen who framed the Constitution were fully sensible, that from the complex character of the Government, it must fail unless the States mutually supported each other and the General Government; and that nothing would be more likely to disturb its peace, and end in discord, than permitting an offender against the laws of a State, by passing over a mathematical line which divides it from another, to defy its process, and stand ready, under the protection of the State, to repeat the offence as soon as another opportunity offered.” Kentucky v. Dennison, 24 How. 66, 100 (1861). The Extradition Clause, however, does not specifically establish a procedure by which interstate extradition is to take place, and, accordingly, has never been considered to be selfexecuting. See, e. g., Hyatt v. People ex rel. Corkran, 188 U. S. 691, 708-709 (1903); Kentucky v. Dennison, supra, at 104. Early in our history, the lack of an established procedure led to a bitter dispute between the States of Virginia and Pennsylvania. J. Scott, Law of Interstate Rendition 5-7 (1917). In 1791, Pennsylvania demanded the extradition of three men charged with kidnaping a free black man and selling him into slavery. Virginia refused to comply with Pennsylvania’s demand. The controversy was finally submitted to President Washington who, relying upon the advice of Attorney General Randolph, 9 National State Papers of the United States 1789-1817, pt. II, pp. 144-145 (E. Carzo ed. 1985), personally appeared before the Congress to obtain the enactment of a law to regulate the extradition process. Con- CALIFORNIA v. SUPERIOR COURT OF CALIFORNIA 407 400 Opinion of the Court gress responded by enacting the Extradition Act of 1793, which provides in its current form: “Whenever the executive authority of any State or Territory demands any person as a fugitive from justice, of the executive authority of any State, District or Territory to which such person has fled, and produces a copy of an indictment found or an affidavit made before a magistrate of any State or Territory, charging the person demanded with having committed treason, felony or other crime, certified as authentic by the governor or chief magistrate of the State or Territory from whence the person so charged has fled, the executive authority of the State, District or Territory to which such person has fled shall cause him to be arrested and secured, and notify the executive authority making such demand, or the agent of such authority appointed to receive the fugitive, and shall cause the fugitive to be delivered to such agent when he shall appear.” 18 U. S. C. §3182. This Court has held the Extradition Act of 1793 to be a proper exercise of Congress’ powers under the Extradition Clause and Art. IV, § 1, to “prescribe the manner in which acts, records and proceedings shall be proved, and the effect thereof.” Kentucky v. Dennison, supra, at 105; Prigg v. Pennsylvania, 16 Pet. 539, 618-622 (1842). By the express terms of federal law, therefore, the asylum State is bound to deliver up to the demanding State’s agent a fugitive against whom a properly certified indictment or affidavit charging a crime is lodged. The language, history, and subsequent construction of the Extradition Act make clear that Congress intended extradition to be a summary procedure. As we have repeatedly held, extradition proceedings are “to be kept within narrow bounds”; they are “emphatically” not the appropriate time or place for entertaining defenses or determining the guilt or innocence of the charged party. Biddinger n. Commissioner 408 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. of Police, 245 U. S. 128, 135 (1917); see also, e. g., Michigan n. Doran, 439 U. S. 282, 288 (1978); Drew n. Thaw, 235 U. S. 432, 440 (1914); Pierce v. Creecy, 210 U. S. 387, 405 (1908); In re Strauss, 197 U. S. 324, 332-333 (1905). Those inquiries are left to the prosecutorial authorities and courts of the demanding State, whose duty it is to justly enforce the demanding State’s criminal law—subject, of course, to the limitations imposed by the Constitution and laws of the United States. Biddinger n. Commissioner of Police, supra, at 135; Drew v. Thaw, supra, at 440. The courts of asylum States may do no more than ascertain vzhether the requisites of the Extradition Act have been met. As the Court held in Michigan v. Doran, supra, the Act leaves only four issues open for consideration before the fugitive is delivered up: “(a) whether the extradition documents on their face are in order; (b) whether the petitioner has been charged with a crime in the demanding state; (c) whether the petitioner is the person named in the request for extradition; and (d) whether .the petitioner is a fugitive.” 439 U. S., at 289. The parties argue at length about the propriety of the California courts taking judicial notice of their prior child custody decrees in this extradition proceeding. But even if taking judicial notice of the decrees is otherwise proper, the question remains whether the decrees noticed were relevant to one of these four inquiries. The Smolins do not dispute that the extradition documents are in order, that they are the persons named in the documents and that they meet the technical definition of a “fugitive.” Their sole contention is that, in light of the earlier California custody decrees and the federal Parental Kidnaping Prevention Act of 1980, 28 U. S. C. § 1738A, they have not been properly charged with a violation of Louisiana’s kidnaping statute, La. Rev. Stat. Ann. § 14:45 (West 1986). CALIFORNIA v. SUPERIOR COURT OF CALIFORNIA 409 400 Opinion of the Court Section 14:45A(4) prohibits the “intentional taking, enticing or decoying away and removing from the state, by any parent, of his or her child, from the custody of any person to whom custody has been awarded by any court of competent jurisdiction of any state, without the consent of the legal custodian, with intent to defeat the jurisdiction of the said court over the custody of the child.” A properly certified Louisiana information charges the Smolins with violating this statute by kidnaping Jennifer and Jamie Smolin. The information is based on the sworn affidavit of Judith Pope which asserts: “‘On March 9, 1984, at approximately 7:20 a. m., Richard Smolin and Gerard Smolin, kidnapped Jennifer Smolin, aged 10, and James C. Smolin, aged 9, from the affiant’s custody while said children were at a bus stop in St. Tammany Parish, Louisiana. “The affiant has custody of the said children by virtue of a Texas court order dated February 5, 1981, a copy of said order attached hereto and made part hereof. The information regarding the actual kidnapping was told to the affiant by witnesses Mason Galatas and Cheryl Galatas of 2028 Mallard Street, Slidell, Louisiana, and Jimmie Huessler of 2015 Dridle Street, Slidell, Louisiana. Richard Smolin and Gerard Smolin were without authority to remove children from affiant’s custody.’” App. B to Pet. for Cert. 5-6. The information is in proper form, and the Smolins do not dispute that the affidavit, and documents incorporated by reference therein, set forth facts that clearly satisfy each element of the crime of kidnaping as it is defined in La. Rev. Stat. Ann. § 14:45A(4) (West 1986). If we accept as true every fact alleged, the Smolins are properly charged with kidnaping under Louisiana law. In our view, this ends the inquiry into the issue whether or not a crime is charged for purposes of the Extradition Act. 410 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. The Smolins argue, however, that more than a formal charge is required, citing the following language from Roberts v. Reilly, 116 U. S. 80, 95 (1885): “It must appear, therefore, to the governor of the State to whom such a demand is presented, before he can lawfully comply with it, first, that the person demanded is substantially charged with a crime against the laws of the State from whose justice he is alleged to have fled, by an indictment or an affidavit, certified as authentic by the governor of the State making the demand. . . . “[This] is a question of law, and is always open upon the face of the papers to judicial inquiry, on an application for a discharge under a writ of habeas corpus.” The Smolins claim that this language in Roberts spawned a widespread practice of permitting the fugitive, upon a petition for writ of habeas corpus in the asylum State’s courts, to show that the demanding State’s charging instrument is so insufficent that it cannot withstand some generalized version of a motion to dismiss or common-law demurrer. Tr. of Oral Arg. 29-36. The cases the Smolins principally rely upon as support for this asserted practice are People ex rel. Lewis n. Commissioner of Correction of City of New York, 100 Mise. 2d 48, 417 N. Y. S. 2d 377 (1979), aff’d, 75 App. Div. 2d 526, 426 N. Y. S. 2d 969 (1980), and Application of Varona, 38 Wash. 2d 833, 232 P. 2d 923 (1951). See Brief for Respondent 15-17. In Lewis, however, the New York trial court actually granted extradition despite its apparent misgivings about the substantiality of the criminal charge. Lewis, supra, at 56, 417 N. Y. S. 2d, at 382. And, in Varona, the Washington Supreme Court relied on the fact that the indictment, on its face, did not charge a crime under California law. Application of Varona, supra, at 833-834, 232 P. 2d, at 923-924. Neither case, in our view, supports the broad proposition that the asylum State’s courts may entertain motions to dismiss or demurrers to the indictment or information from the demanding State. CALIFORNIA v. SUPERIOR COURT OF CALIFORNIA 411 400 Opinion of the Court To the contrary, our cases make clear that no such inquiry is permitted. For example, in Pierce v. Creecy, decided after Roberts, supra, this Court refused to grant relief from extradition over multiple objections to the sufficiency of the indictment. The Pierce Court concluded that it was enough that “the indictment, whether good or bad, as a pleading, unmistakably describes every element of the crime of false swearing, as it is defined in the Texas Penal Code ....” 210 U. S., at 404. It reasoned: “If more were required it would impose upon courts, in the trial of writs of habeas corpus, the duty of a critical examination of the laws of States with whose jurisprudence and criminal procedure they can have only a general acquaintance. Such a duty would be an intolerable burden, certain to lead to errors in decision, irritable to the just pride of the States and fruitful of miscarriages of justice. The duty ought not be assumed unless it is plainly required by the Constitution, and, in our opinion, there is nothing in the letter or the spirit of that instrument which requires or permits its performance.” Id., at 405. Similarly, in Biddinger v. Commissioner of Police, 245 U. S. 12t8 (1917), the appellant argued that he had a seemingly valid statute of limitations defense based on the fact that more than three years, the limitations period, had elapsed since the date of the crime recited in the indictment and that he had been publicly and openly resident in the demanding State for that entire period. The Court found that the question of limitations was properly considered only in the demanding State’s courts. Id., at 135; see also Drew v. Thaw, 235 U. S., at 439-440 (whether the escape of a person committed to a mental institution is a crime “is a question as to the law of New York which the New York courts must decide”). This proceeding is neither the time nor place for the Smolins’ arguments that Judith Pope’s affidavit is fraudulent 412 OCTOBER TERM, 1986 Stevens, J., dissenting 482 U. S. and that the California custody decrees establish Richard as the lawful custodian under the full faith and credit provision of the federal Parental Kidnaping Prevention Act of 1980. There is nothing in the record to suggest that the Smolins are not entirely correct in all of this: that California had exclusive modification jurisdiction over the custody of Jennifer and Jamie; that, under the California decrees, Richard Smolin had lawful custody of the children when he brought them to California; and, that, accordingly, the Smolins did not violate La. Rev. Stat. Ann. § 14:45A(4) (West 1986) as is charged. Of course, the Parental Kidnaping Prevention Act of 1980 creates a uniform federal rule governing custody determinations, a rule to which the courts of Louisiana must adhere when they consider the Smolins’ case on the merits. We are not informed by the record why it is that the States of California and Louisiana are so eager to force the Smolins halfway across the continent to face criminal charges that, at least to a majority of the California Supreme Court, appear meritless. If the Smolins are correct, they are not only innocent of the charges made against them, but also victims of a possible abuse of the criminal process. But, under the Extradition Act, it is for the Louisiana courts to do justice in this case, not the California courts: “surrender is not to be interfered with by the summary process of habeas corpus upon speculations as to what ought to be the result of a trial in the place where the Constitution provides for its taking place.” Drew n. Thaw, supra, at 440. The judgment of the California Supreme Court is Reversed. Justice Stevens, with whom Justice Brennan joins, dissenting. There is no constitutional or statutory reason why the scope of an asylum State’s judicial inquiry need be so narrow that it precludes the grant of habeas corpus in this case. It has been settled for over a century that before the Governor of an asylum State can lawfully comply with a requesting CALIFORNIA v. SUPERIOR COURT OF CALIFORNIA 413 400 Stevens, J., dissenting State’s demand for extradition, it must appear that the person sought is “substantially charged with a crime” and is also a fugitive from justice. Roberts v. Reilly, 116 U. S. 80, 95 (1885).1 “The first of these prerequisites is a question of law, and is always open upon the face of the papers to judicial inquiry, on an application for a discharge under a writ of habeas corpus. ” Ibid. Because there is no reasonable possibility that the charges of simple kidnaping filed against Richard and Gerard Smolin in Louisiana are valid, I agree with the California Supreme Court’s conclusion that they have not been substantially charged with a crime. In addition, the Parental Kidnaping Prevention Act of 1980, 28 U. S. C. § 1738A, makes clear that Richard had custody of his daughters and thus there is no reasonable possibility that his travel from Louisiana to California with them made him a fugitive from justice. I The scope of the legal inquiry preceding extradition is extremely restricted because the courts of the asylum State cannot be expected to make “a critical examination of the laws of States with whose jurisprudence and criminal procedure they can have only a general acquaintance.” Pierce n. Creecy, 210 U. S. 387, 405 (1908). Nevertheless, our precedents make clear that if a critical allegation of fact in the indictment is “impossible in law,” see Roberts, 116 U. S., at 96, the asylum State must refuse the extradition demand because the person has not been substantially charged with a crime. Munsey v. Clough, 196 U. S. 364, 373 (1905). In Drew v. Thaw, 235 U. S. 432 (1914), the habeas corpus petitioner was under a New York indictment for conspiracy to obstruct the due administration of laws; he was charged with plotting to effect his own escape from an insane asylum to which he had been committed. Justice Holmes’ opinion for ‘See also Hyatt v. Corkran, 188 U. S. 691, 709-710 (1903); Munsey v. Clough, 196 U. S. 364, 372-373 (1905); Pierce v. Creecy, 210 U. S. 387, 401, 405 (1908). 414 OCTOBER TERM, 1986 Stevens, J., dissenting 482 U. S. the Court held that the indictment charged a crime because New York courts could decide that the conspiracy charged “did tend to obstruct the due administration of the law.” Id., at 439. Even though the habeas court could not inquire “upon the facts or the law of the matter to be tried,” Justice Holmes made it clear that there nevertheless must be a “reasonable possibility” that the crime charged “may be such.” Id., at 439-440.2 In Pierce n. Creecy, the Court acknowledged that “an objection which, if well founded, would destroy the sufficiency of the indictment, as a criminal pleading, might conceivably go far enough to destroy also its sufficiency as a charge of crime.” 210 U. S., at 404. The Court concluded that the objections to the indictment in that case were not of that nature. Likewise, in In re Strauss, 197 U. S. 324 (1905), Ohio sought a fugitive who had been charged by affidavit before a justice of the peace for a felony which was subject to trial only upon an indictment. This Court found no constitutional barrier to extradition on those facts, but observed that the availability of extradition must be balanced against the duty of courts to avoid injustice: “It may be true, as counsel urge, that persons are sometimes wrongfully extradited, particularly in cases like the present; that a creditor may wantonly swear to an affidavit charging a debtor with obtaining goods under false pretences. . . . While courts will always endeavor to see that no such attempted wrong is successful, on the other hand care must be taken that the 2 “When, as here, the identity of the person, the fact that he is a fugitive from justice, the demand in due form, the indictment by a grand jury for what it and the Governor of New York allege to be a crime in that State and the reasonable possibility that it may be such, all appear, the constitutionally required surrender is not to be interfered with by the summary process of habeas corpus upon speculations as to what ought to be the result of a trial in the place where the Constitution provides for its taking place.” Drew v. Thaw, 235 U. S. 432, 440 (1914) (emphasis supplied). CALIFORNIA v. SUPERIOR COURT OF CALIFORNIA 415 400 Stevens, J., dissenting process of extradition be not so burdened as to make it practically valueless.” Id., at 332-333. The inquiry undertaken by the California courts in this case established the “impossibility in law” of convicting the Smolins and therefore the injustice of their extradition. The crime charged was two counts of simple kidnaping in violation of Louisiana law, which defines the crime, in relevant part, as: “The intentional taking, enticing or decoying away and removing from the state, by any parent of his or her child, from the custody of any person to whom custody has been awarded by any court of competent jurisdiction of any state, without the consent of the legal custodian, with intent to defeat the jurisdiction of the said court over the custody of the child.” La. Rev. Stat. Ann. § 14:45A(4) (West 1986). In my opinion the limited scope of the inquiry open to the California courts in this case did not preclude an examination of either federal law or California’s own judicial decrees. This summary examination was permissible because it had a direct bearing on whether the information “substantially charged” the Smolins with a crime or whether there was no reasonable possibility that the crime of simple kidnaping charged “may be such.” Drew v. Thaw, 235 U. S., at 440. The Smolins’ conviction for this crime was an impossibility for three reasons. First, a California court, the court of competent jurisdiction under the federal Parental Kidnaping Prevention Act,3 had awarded sole custody of Jennifer and Jamie to Richard Smolin more than three years before he took them to California; he plainly could not be convicted of removing the children from his own custody. Second, regardless of whether Richard or Judith Smolin had custody of the children, he clearly believed that custody had been 3 See ante, at 405, n. 416 OCTOBER TERM, 1986 Stevens, J., dissenting 482 U. S. awarded to him by a California court which retained jurisdiction. His act of taking the children to California therefore could not have been accomplished with the intent to defeat the jurisdiction of that court. Third, because he did not believe that a Louisiana court had jurisdiction over the custody determination, he could not logically be convicted under the kidnaping statute for departing from Louisiana with the intent to defeat the jurisdiction of the courts of that State. There is, in short, no possibility—and certainly no “reasonable possibility”—that his conduct violated the Louisiana statute cited in the extradition papers.4 A sensible application of the requirement that a fugitive must be “substantially charged” with a crime, informed by the twin necessities of avoiding a trial-like inquiry into the law of sister States and preventing the injustice of extradition to face a legally impossible charge, leads me to conclude that the judgment of the California Supreme Court should be affirmed. The Court’s heavy reliance on the dicta in Michigan v. Doran, 439 U. S. 282, 288 (1978), and Biddinger v. Commissioner of Police, 245 U. S. 128, 135 (1917), is misplaced. The issue in Doran was whether a court in the asylum State could review the demanding State’s judicial determination that there was probable cause for the fugitive’s arrest—an issue that is entirely unrelated to the substantiality of the criminal charge. The fact that the Court omitted the word “substan- 4 The Louisiana Assistant District Attorney who filed the information against the Smolins was aware of the California custody orders at the time he filed the information. He believed, however, that a crime had been committed because “‘he viewed the California judgment as being void, having been obtained by fraudulent misrepresentations, and the valid order having been that issued by Texas on February 13, 1981.”’ 41 Cal. 3d 758, 763, n. 1, 716 P. 2d 991, 993, n. 1 (1986). In my opinion that speculation on the part of the Assistant District Attorney is inadequate to overcome the fact that Richard Smolin, as the holder of a custody determination that was valid on its face, could not be substantially charged with a crime for his exercise of the parental rights conferred upon him by that custody determination. CALIFORNIA v. SUPERIOR COURT OF CALIFORNIA 417 400 Stevens, J., dissenting tial” in its summary description of the proper inquiry in the asylum State surely was not intended to modify or eliminate a requirement that this Court had recognized for decades. See, e. g., McNichols v. Pease, 207 U. S. 100, 108-109 (1907) (accused must be “substantially charged with crime against the laws of the demanding State”); Ex parte Reggel, 114 U. S. 642, 651 (1885) (indictment accompanying the requisition was valid because it substantially charged the crime). In recognition of this longstanding requirement, the Uniform Criminal Extradition Act, which both Louisiana and California have adopted, specifies that the “indictment, information, or affidavit made before the magistrate must substantially charge the person demanded with having committed a crime under the law of that state.” 11 U. L. A. 92 (1974); La. Code Crim. Proc. Ann., Art. 263 (West 1967); Cal. Penal Code Ann. § 1548.2 (West 1982). The Biddinger case relied upon by the Court is also inapposite because the validity of the fugitive’s statute of limitations defense in that case depended on the law of the demanding State; the fact that the limitations period had expired between the date of the offense and the charge did not foreclose the possibility that the statute had been tolled. 245 U. S., at 131-132, 135. The common thread in Doran and Biddinger, as in Drew v. Thaw, supra, is that an asylum state court’s inquiry may not reach the merits of issues that could be fully litigated in the charging State; such examinations entangle the asylum State’s judicial system in laws with which it is unfamiliar and endanger the summary nature of extradition proceedings. To obtain habeas relief, “[t]here must be objections which reach deeper into the indictment than those which would be good against it in the court where it is pending.” Pierce v. Creecy, 210 U. S., at 401; cf. Pacileo v. Walker, 449 U. S. 86, 87-88 (1980) (per curiam) (California Supreme Court erred in granting habeas relief to fugitive by directing its Superior Court to determine whether prison conditions in demanding State violated Eighth Amendment). 418 OCTOBER TERM, 1986 Stevens, J., dissenting 482 U. S. Neither of those dangers is posed by the respondent California Superior Court’s conclusion that the Smolins had legal custody and thus were not “substantially charged” with kidnaping.5 II Prima facie proof that the accused be “a fugitive from the justice of the demanding State” is a “condition precedent to the surrender of the accused.” Ex parte Reggel, 114 U. S., at 652-653. Deeming Richard Smolin a “fugitive from justice” would not serve the purpose of the Extradition Clause. The Framers’ provision for extradition was designed to prevent state boundaries from becoming impermeable walls within which “the fugitives from a sister State’s criminal justice system” may find “safe haven.” Ante, at 406 (quoting Kentucky n. Dennison, 24 How. 66, 100 (1861)); cf. Jones v. Helms, 452 U. S. 412, 419 (1981) (State’s right to obtain extradition of criminal necessarily qualifies that citizen’s right to interstate travel). The requirement that fugitivity be established nevertheless has some teeth to it;6 otherwise state boundaries would become mere markings in an atlas, and the demanding State could exercise criminal jurisdiction over a person anywhere in the Union regardless of the extent 5 An asylum State’s review of a determination by a magistrate in the requesting State that probable cause exists to arrest the fugitive may cause “friction and delay,” but nothing indicates that “routine and basic inquiry” into the existence of a charge “has led to frustration of the extradition process.” Michigan v. Doran, 439 U. S. 282, 296-297, n. 7 (1978) (Blackmun, J., concurring in result). 6 “Any other interpretation would lead to the conclusion that the mere requisition by the executive of the demanding State, accompanied by the copy of an indictment, or an affidavit before a magistrate, certified by him to be authentic, charging the accused with crime committed within her limits, imposes upon the executive of the State or Territory where the accused is found, the duty of surrendering him, although he may be satisfied, from incontestable proof, that the accused had, in fact, never been in the demanding State, and, therefore, could not be said to have fled from its justice.” Ex parte Reggel, 114 U. S. 642, 652 (1885). CALIFORNIA v. SUPERIOR COURT OF CALIFORNIA 419 400 Stevens, J., dissenting of that person’s culpable connection with the State.7 Thus, to be a fugitive from justice it is necessary “that having within a State committed that which by its laws constitutes a crime, when he is sought to be subjected to its criminal process to answer for his offence, he has left its jurisdiction and is found within the territory of another.” Roberts n. Reilly, 116 U. S., at 97 (emphasis added). “For all that is necessary to convert a criminal under the laws of a State into a fugitive from justice is that he should have left the State after having incurred guilt there.” Strassheim v. Daily, 221 U. S. 280, 285 (1911) (citing Roberts v. Reilly, supra). See also Appleyard v. Massachusetts, 203 U. S. 222, 227 (1906). Despite this seemingly sweeping language, we have previously rejected the claim that a person could be considered a fugitive if he could establish that he was outside of the demanding State at the time of the alleged offense, even if “constructive presence” would be a sufficient basis for criminal liability. In Munsey v. Clough, we wrote: “When it is conceded, or when it is so conclusively proved, that no question can be made that the person was not within the demanding State when the crime is said to have been committed, and his arrest is sought on the ground only of a constructive presence at that time, in the demanding State, then the court will discharge the defendant. Hyatt v. Corkran, 188 U. S. 691 [(1903)], affirming the judgment of the New York Court of Appeals, 172 N. Y. 176 [1902].” 196 U. S., at 374-375. See also South Carolina v. Bailey, 289 U. S. 412, 421-422 (1933); McNichols v. Pease, 207 U. S., at 109-110 (1907); Ex parte Reggel, 114 U. S., at 651. 7 In the context of extradition—a form of recognition of sister-state indictments—no less than in the context of recognition of judgments or of laws, “[s]tate boundaries are neither irrelevancies nor licenses to disengage.” Brilmayer, Credit Due Judgments and Credit Due Laws: The Respective Roles of Due Process and Full Faith and Credit in the Interstate Context, 70 Iowa L. Rev. 95, 112 (1984). 420 OCTOBER TERM, 1986 Stevens, J., dissenting 482 U. S. Similarly, I believe that we should today reject the notion that a parent who holds custody as determined by the Parental Kidnaping Prevention Act of 1980, 28 U. S. C. § 1738A, must be extradited as a charged kidnaper. Three reasons compel this conclusion. First, when the fleeing parent lacks child custody under federal law, it is proper to subject him or her to extradition in order to face criminal prosecution. But when the parent acts consistently with the federal law that governs interstate custody disputes, he should not be deemed to have fled from the judicial process of the demanding State. By allowing the custodial parent under federal law to be branded as a fugitive, the Court implicitly approves nonadherence to the uniform federal rule governing custody determinations. Second, requiring the extradition of Richard Smolin is at cross-purposes with Congress’ intent to “discourage continuing interstate controversies over child custody” and to “deter interstate abductions and other unilateral removals of children undertaken to obtain custody and visitation awards.” See 28 U. S. C. § 1738A note.8 Compelling extradition to face a criminal charge which cannot lead to a conviction, no less than “child snatching,” is the coerced transportation of a party to a custodial dispute to another forum in order to serve a private interest. It is anomalous that the Act, which 8 A uniform rule establishing which parent has custody deters “child snatching.” See Field, Sources of Law: The Scope of Federal Common Law, 99 Harv. L. Rev. 881, 959, n. 340 (1986). The Parental Kidnaping Prevention Act achieves a uniform rule in practice by establishing the circumstances under which a State may render or modify a child custody determination and requiring that other States give full faith and credit to judgments that conform to these standards. See 28 U. S. C. §§ 1738A(a), (c)-(g). If States were free not to give full faith and credit to the custody judgments of other States, a forum-shopping parent would have an incentive to remove the child to a State which was more likely to render a custodial decree in favor of that parent. See Brilmayer, supra, at 103. CALIFORNIA v. SUPERIOR COURT OF CALIFORNIA 421 400 Stevens, J., dissenting was clearly intended to deter the former type of coercion, should not also be interpreted to discourage the latter.9 Third, the Extradition Clause should be construed consistently with the Parental Kidnaping Prevention Act because both are expressions of the constitutional command of full faith and credit that governs relations among the several States. The Extradition Clause “articulated, in mandatory language, the concepts of comity and full faith and credit, found in the immediately preceding clause of Art. IV.” Michigan v. Doran, 439 U. S., at 287-288. The courts of every State best adhere to this principle, when considering an extradition request for alleged parental kidnaping, by giving full faith and credit to custody judgments rendered by other States as commanded by the Act. It is clear to a court performing this task that the Smolins are not fugitives within the meaning of the extradition request; as the custodial parent under the federal statute, Richard Smolin did not commit while in Louisiana “an act which by the law of the State constitutes a crime.” Hogan v. O'Neill, 255 U. S. 52, 56 (1921). 9 Of course, persons who remove a child from a State in violation of the Parental Kidnaping Prevention Act should be brought to justice. Indeed, Congress has explicitly pointed out that the Fugitive Felon Act, 18 U. S. C. § 1073, which makes it a federal crime for a person to move or travel “in interstate or foreign commerce with intent... to avoid prosecution . . . under the laws of the place from which he flees, for a crime . . . which is a felony under the laws of the place from which the fugitive flees” applies to parental kidnaping. 28 U. S. C. § 1738A note. The Act also makes available, in certain limited instances, the assistance of the Federal Bureau of Investigation in apprehending interstate abductors. See generally Donigan, Child Custody Jurisdiction: New Legislation Reflects Public Policy Against Parental Abduction, 19 Gonz. L. Rev. 1, 64-66 (1983-1984) (Department of Justice does not interpret Act to require routine federal involvement in parental abductions). Congress’ assertions of the federal interest in regulating parental abduction require habeas courts to exercise particular vigilance that a custodial parent not be extradited as a fugitive from justice. 422 OCTOBER TERM, 1986 Stevens, J., dissenting 482 U. S. Ill The Court is scrupulously fair in its recital of the facts and frank in its acknowledgment that the criminal process may have been abused in this case. The reasoning the Court follows nevertheless adopts an overly restrictive view of the questions that the habeas courts of a rendering State must pose. The law governing interstate rendition for criminal proceedings does not foreclose a summary inquiry into whether the crime charged is legally impossible. Moreover, in an area in which Congress has seen fit to enact nationwide legislation, I cannot agree that respect for the criminal laws of other States requires the State of California indiscriminately to render as fugitives those citizens who are con-clusorily charged with simple kidnaping for their exercise of a right conferred upon them by a valid custody decree issued by a California court. The Court’s contrary conclusion will, I fear, produce unnecessary inconvenience and injustice in this case and provide estranged parents with an inappropriate weapon to use against each other as they wage custody disputes throughout this land. I respectfully dissent. MILLER v. FLORIDA 423 Syllabus MILLER v. FLORIDA CERTIORARI TO THE SUPREME COURT OF FLORIDA No. 86-5344. Argued April 21, 1987—Decided June 9, 1987 Florida’s sentencing guidelines law assigns points for particular offenses and other factors and provides a presumptive sentence range for a defendant’s composite score, within which the sentencing judge has unreviewable discretion to fix a sentence without written explanation. If the judge wishes to depart from the range, however, he must give clear and convincing written reasons based on facts proved beyond a reasonable doubt, and the sentence he imposes is subject to appellate review. At the time petitioner committed the sexual battery and other crimes for which he was convicted, the sentencing guidelines would have resulted in a presumptive sentence of 372 to 472 years’ imprisonment. However, the guidelines were subsequently revised to increase the number of points assigned to sexual offenses, and, at the time petitioner was sentenced, called for a presumptive sentence of 572 to 7 years for his crimes. The sentencing judge, rejecting petitioner’s ex post facto argument, applied the revised guidelines to impose a 7-year sentence. The State District Court of Appeal vacated the sentence, but the State Supreme Court reversed. Held: Application of the revised guidelines law to petitioner, whose crimes occurred before the law’s effective date, violates the Ex Post Facto Clause of Article I of the Federal Constitution. The revised law evidences all of the elements necessary to bring it within the ex post facto prohibition. Pp. 429-435. (a) The revised guidelines law is retrospective in that it changes the legal consequences of acts committed before its effective date. The State’s argument that there was no ex post facto violation since the law provides for continuous review of the guidelines and thereby gave petitioner “fair warning” that he would be sentenced under the guidelines in effect on his sentencing date is not persuasive, since the law did not warn petitioner of the specific punishment prescribed for his crimes. The ex post facto prohibition cannot be avoided merely by adding to a law notice of the obvious fact that it might be changed. Pp. 430-431. (b) The revised guidelines law is more onerous than the law in effect at the time of petitioner’s crimes, in that it substantially disadvantages petitioner and similarly situated sexual offenders and has no ameliorative features. The State’s contention that the change in laws is not disadvantageous because the trial judge could have imposed a 7-year 424 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. sentence under the old guidelines by departing from the presumptive sentence range then in existence is without merit, since the revised law foreclosed petitioner’s ability to challenge the sentence on review because it is within the new presumptive range. Pp. 431-433. (c) The revised guidelines law is not merely a procedural change, since it increases the quantum of punishment for sexual offenses. The State’s contention that the increase operates only as a “procedural guidepost” for the exercise of judicial discretion within the same statutorily imposed sentencing limits is not persuasive. The Court of Appeals decisions cited as authority, which sustained the United States Parole Commission’s guidelines against ex post facto claims, are inapposite. Unlike the federal guidelines, Florida’s revised sentencing law was enacted by the state legislature and has the force and effect of law. Nor do the revised guidelines simply provide flexible “guideposts,” but instead create strict standards that must be met before the sentencing judge can depart from the presumptive sentence range. Moreover, the revised guidelines directly and adversely affect the sentence petitioner receives. Pp. 433-435. 488 So. 2d 820, reversed and remanded. O’Connor, J., delivered the opinion for a unanimous Court. Anthony Calvello argued the cause for petitioner. With him on the briefs were Richard L. Jorandby and Craig S. Barnard. Joy B. Shearer, Assistant Attorney General of Florida, argued the cause for respondent. With her on the brief was Robert A. Butterworth, Attorney General.* Justice O’Connor delivered the opinion of the Court. At the time petitioner committed the crime for which he was convicted, Florida’s sentencing guidelines would have resulted in a presumptive sentence of 3% to 4% years’ imprisonment. At the time petitioner was sentenced, the revised guidelines called for a presumptive sentence of 5% to 7 years in prison. The trial court applied the guidelines in effect at the time of sentencing and imposed a 7-year sentence. The question presented is whether application of these amended *Gerald D. Stem and Alvin Bronstein filed a brief for the American Civil Liberties Union et al. as amici curiae urging reversal. MILLER v. FLORIDA 425 423 Opinion of the Court guidelines in petitioner’s case is unconstitutional by virtue of the Ex Post Facto Clause. I In 1983, the Florida Legislature enacted legislation replacing Florida’s system of indeterminate sentencing with a sentencing guidelines scheme intended “to eliminate unwarranted variation in the sentencing process.” Fla. Rule Crim. Proc. 3.701(b) (1983). See 1983 Fla. Laws, ch. 83-216. Under the sentencing statute, a guidelines commission was responsible for “the initial development of a statewide system of sentencing guidelines.” Fla. Stat. §921.001(1) (1983). Once the commission had made its recommendation, the Supreme Court of Florida was to develop a final system of guidelines. These guidelines were to become effective for crimes committed on or after October 1, 1983. Fla. Stat. §921.001(4)(a) (1983). The sentencing statute authorized the guidelines commission to “meet annually or at the call of the chairman to review sentencing practices and recommend modifications to the guidelines.” Fla. Stat. §921.001(3) (1983). Before the convening of the legislature each year, the commission was to make its recommendations regarding the need for changes in the guidelines. The Supreme Court of Florida then could revise the sentencing guidelines to conform to all or part of the commission’s recommendations. The sentencing law provided, however, that such revisions would become effective “only upon the subsequent adoption by the Legislature of legislation implementing the guidelines as then revised.” Fla. Stat. §921.001(4)(b) (1983). In accordance with this legislation, the Supreme Court of Florida developed sentencing guidelines that went into effect on October 1, 1983. See In re Rules of Criminal Procedure (Sentencing Guidelines), 439 So. 2d 848 (1983). Under the scheme, offenses were grouped into nine “offense categories” 9-, “robbery” and “sexual offenses”). A single sentencing 426 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. “scoresheet” would be prepared based on the defendant’s “primary offense,” defined as the crime “with the highest statutory degree” at the time of conviction. Fla. Rule Crim. Proc. 3.701(d) (1983). In scoring a defendant’s guidelines sentence, points would be assigned based on the primary offense, additional offenses at the time of conviction, prior record, legal status at the time of the offense, and victim injury. The defendant’s total point score then would be compared to a chart for that offense category, which provided a presumptive sentence for that composite score. The presumptive sentence range was “assumed to be appropriate for the composite score of the offender.” Fla. Rule Crim. Proc. 3.701(d)(8) (1983). Within the recommended range, the sentencing judge had discretion to fix the sentence “without the requirement of a written explanation.” Ibid. If the sentencing judge wished to depart from the guideline range, however, the judge had to give clear and convincing reasons in writing for doing so: “Departures from the presumptive sentence should be avoided unless there are clear and convincing reasons to warrant aggravating or mitigating the sentence. Any sentence outside of the guidelines must be accompanied by a written statement delineating the reasons for the departure. Reasons for deviating from the guidelines shall not include factors relating to either instant offense or prior arrests for which convictions have not been obtained.” Fla. Rule Crim. Proc. 3.701(d)(ll) (1983). The “clear and convincing” standard was construed as requiring reasons “of such weight as to produce in the mind of the judge a firm belief or conviction, without hesitancy, that departure is warranted.” State v. Mischler, 488 So. 2d 523, 525 (Fla. 1986). Only those sentences that fall outside the guidelines’ range are subject to appellate review. See Fla. Stat. §921.001(5) (1983). Petitioner was convicted in August 1984 on counts of sexual battery with slight force, a second-degree felony, Fla. MILLER v. FLORIDA 427 423 Opinion of the Court Stat. §794.011(5) (Supp. 1984); burglary with an assault, a felony of the “first degree punishable by . . . life,” Fla. Stat. §810.02 (1983); and petit theft, a misdemeanor, Fla. Stat. §812.014(2)(c) (1983). On April 25, 1984, when these offenses were committed, the sentencing guidelines adopted October 1,1983, were still in effect. On May 8,1984, however, the Supreme Court of Florida proposed several revisions to the sentencing guidelines. See Florida Bar: Amendment to Rules of Criminal Procedure (3.701, 3.988—Sentencing Guidelines), 451 So. 2d 824 (1984). In June 1984 the Florida Legislature adopted the recommended changes, see 1984 Fla. Laws, ch. 84-328, and the legislation implementing the revised guidelines became effective July 1, 1984. When petitioner was sentenced on October 2, 1984, therefore, these revised sentencing guidelines were the guidelines then in effect. Only two changes made in the revised guidelines are relevant here. First, the guidelines changed the definition of “primary offense” from the offense with “the highest statutory degree,” to the offense which results in “the most severe sentence range.” See 451 So. 2d, at 824, n. This changed petitioner’s primary offense from burglary with assault—the offense with the higher statutory degree—to sexual battery. Petitioner does not argue here that the new definition itself changed his presumptive sentence. See Tr. of Oral Arg. 6. As a result of the new definition, however, petitioner was affected by another change in the revised guidelines law: a 20% increase in the number of primary offense points assigned to sexual offenses. The Supreme Court of Florida, in its com-ments accompanying the revised guidelines, described the change: “The revision increases the primary offense points by 20% and will result in both increased rates and length of incarceration for sexual offenders.” 451 So. 2d, at 824, n. As a result of the point increase, petitioner’s total point score jumped to a presumptive sentence of 5/2 to 7 years. See App. 12. 428 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. At petitioner’s sentencing hearing on October 2, 1984, the State contended that the revised guidelines should apply in determining petitioner’s sentence. Alternatively, the State argued that if the sentencing judge applied the earlier guidelines, he should depart from the guidelines’ range and impose a 7-year sentence. Id., at 8-9. The sentencing judge, rejecting petitioner’s ex post facto argument, ruled that the revised guidelines should apply. Concluding that he would “stay within the new guidelines,” the judge imposed a 7-year term of imprisonment for the sexual assault count. Id., at 10. Petitioner received a concurrent 7-year sentence on the burglary count, and time served on the misdemeanor charge. Id., at 6, 11. On appeal, the Florida District Court of Appeal, relying on this Court’s decision in Weaver v. Graham, 450 U. S. 24 (1981), vacated petitioner’s sentence and remanded for resentencing in accordance with the sentencing guidelines in effect at the time the offense was committed. 468 So. 2d 1018 (1985). In remanding the case, the court noted that “the same sentence is possible if clear and convincing reasons for departure from the then applicable guidelines are stated in writing.” Ibid. The Supreme Court of Florida reversed. 488 So. 2d 820 (1986). In a summary opinion, the court concluded that its decision in State v. Jackson, 478 So. 2d 1054 (1985), established that “the trial court may sentence a defendant pursuant to the guidelines in effect at the time of sentencing.” 488 So. 2d, at 820. In Jackson, the Supreme Court of Florida had emphasized that “the presumptive sentence established by the guidelines does not change the statutory limits of the sentence imposed for a particular offense.” 478 So. 2d, at 1056. On that basis, it had concluded that a modification in sentencing guidelines procedure was “merely a procedural change, not requiring the application of the ex post facto doctrine” under Dobbert v. Florida, 432 U. S. 282 (1977). 478 So. 2d, at 1056. MILLER v. FLORIDA 429 423 Opinion of the Court We granted certiorari, 479 U. S. 960 (1986), and now reverse. II Article I of the United States Constitution provides that neither Congress nor any State shall pass any “ex post facto Law.” See Art. I, §9, cl. 3; Art. I, §10, cl. 1. Our understanding of what is meant by ex post facto largely derives from the case of Calder v. Bull, 3 Dall. 386 (1798), in which this Court first considered the scope of the ex post facto prohibition. In Calder, Justice Chase, noting that the expression “ex post facto” “had been in use long before the revolution,” id., at 391, summarized his understanding of what fell “within the words and the intent of the prohibition”: “1st. Every law that makes an action done before the passing of the law, and which was innocent when done, criminal; and punishes such action. 2d. Every law that aggravates a crime, or makes it greater than it was, when committed. 3d. Every law that changes the punishment, and inflicts a greater punishment, than the law annexed to the crime, when committed. 4th. Every law that alters the legal rules of evidence, and receives less, or different testimony, than the law required at the time of the commission of the offense, in order to convict the offender.” Id., at 390 (emphasis omitted). Accord, Dobbert v. Florida, supra, at 292, quoting Beazell v. Ohio, 269 U. S. 167, 169-170 (1925). Justice Chase explained that the reason the Ex Post Facto Clauses were included in the Constitution was to assure that federal and state legislatures were restrained from enacting arbitrary or vindictive legislation. See 3 Dall., at 389. Justices Paterson and Iredell, in their separate opinions in Calder, likewise emphasized that the Clauses were aimed at preventing legislative abuses. See id., at 396 (Paterson, J.); Id., at 399-400 (Iredell, J.). See also Malloy v. South Caro-tina, 237 U. S. 180, 183 (1915); James v. United States, 366 430 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. U. S. 213, 247, n. 3 (1961) (separate opinion of Harlan, J.). In addition, the Justices’ opinions in Calder, as well as other early authorities, indicate that the Clauses were aimed at a second concern, namely, that legislative enactments “give fair warning of their effect and permit individuals to rely on their meaning until explicitly changed.” Weaver v. Graham, supra, at 28-29. See Calder v. Bull, 3 Dall., at 388 (Chase, J.); id., at 396 (Paterson, J.); 1W. Blackstone, Commentaries *46. Thus, almost from the outset, we have recognized that central to the ex post facto prohibition is a concern for “the lack of fair notice and governmental restraint when the legislature increases punishment beyond what was prescribed when the crime was consummated.” Weaver, 450 U. S., at 30. Our test for determining whether a criminal law is ex post facto derives from these principles. As was stated in Weaver, to fall within the ex post facto prohibition, two critical elements must be present: first, the law “must be retrospective, that is, it must apply to events occurring before its enactment”; and second, “it must disadvantage the offender affected by it.” Id., at 29. We have also held in Dobbert v. Florida, supra, that no ex post facto violation occurs if a change does not alter “substantial personal rights,” but merely changes “modes of procedure which do not affect matters of substance.” Id., at 293. See Beazell n. Ohio, supra, at 170-171. Respondent contends that the revised sentencing law is neither impermissibly retrospective, nor to petitioner’s disadvantage; respondent also contends that the revised sentencing law is merely a procedural change. We consider these claims in turn. A law is retrospective if it “changes the legal consequences of acts completed before its effective date.” Weaver, supra, at 31. Application of the revised guidelines law in petitioner’s case clearly satisfies this standard. Respondent nevertheless contends that the ex post facto concern for retrospective laws is not violated here because Florida’s sen MILLER v. FLORIDA 431 423 Opinion of the Court tencing statute “on its face provides for continuous review and recommendation of changes to the guidelines. ” Brief for Respondent 27-28. Relying on our decision in Dobbert, respondent argues that it is sufficient that petitioner was given “fair warning” that he would be sentenced pursuant to the guidelines then in effect on his sentencing date. Brief for Respondent 28. In our view, Dobbert provides scant support for such a pinched construction of the ex post facto prohibition. In Dobbert, the capital sentencing statute in effect at the time the murders took place later was held to be invalid. In rejecting the defendant’s argument that imposition of the death penalty therefore was a change in punishment from the punishment “in effect” when the crimes were committed, the Court concluded that ex post facto concerns were satisfied because the statute on the books at the time Dobbert committed the crimes warned him of the specific punishment Florida prescribed for first-degree murders. See 432 U. S., at 298. Here, by contrast, the statute in effect at the time petitioner acted did not warn him that Florida prescribed a 5%- to 7-year presumptive sentence for that crime. Petitioner simply was warned of the obvious fact that the sentencing guidelines law—like any other law—was subject to revision. The constitutional prohibition against ex post facto laws cannot be avoided merely by adding to a law notice that it might be changed. It is “axiomatic that for a law to be ex post facto it must be more onerous than the prior law.” Dobbert, supra, at 294. Looking only at the change in primary offense points, the revised guidelines law clearly disadvantages petitioner and similarly situated defendants. See 451 So. 2d, at 824, n. (the purpose and effect of the change in primary offense points was to increas[e] [the] rates and length of incarceration for sexual offenders”). Considering the revised guidelines law as a whole does not change this result. Unlike Dobbert, where we found that the “totality of the procedural changes 432 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. wrought by the new statute . . . did not work an onerous application of an ex post facto change,” 432 U. S., at 296-297, here respondent has not been able to identify any feature of the revised guidelines law that could be considered ameliorative. Respondent maintains that the change in guidelines laws is not disadvantageous because petitioner “cannot show definitively that he would have gotten a lesser sentence.” Tr. of Oral Arg. 29. This argument, however, is foreclosed by our decision in Lindsey v. Washington, 301 U. S. 397 (1937). In Lindsey, the law in effect at the time the crime was committed provided for a maximum sentence of 15 years, and a minimum sentence of not less than six months. At the time Lindsey was sentenced, the law had been changed to provide for a mandatory 15-year sentence. Finding that retrospective application of this change was ex post facto, the Court determined that “we need not inquire whether this is technically an increase in the punishment annexed to the crime,” because “[i]t is plainly to the substantial disadvantage of petitioners to be deprived of all opportunity to receive a sentence which would give them freedom from custody and control prior to the expiration of the 15-year term.” Id., at 401-402. Thus, Lindsey establishes “that one is not barred from challenging a change in the penal code on ex post facto grounds simply because the sentence he received under the new law was not more onerous than that which he might have received under the old.” Dobbert, supra, at 300. Petitioner plainly has been “substantially disadvantaged” by the change in sentencing laws. To impose a 7-year sentence under the old guidelines, the sentencing judge would have to depart from the presumptive sentence range of 3% to 4% years. As a result, the sentencing judge would have to provide clear and convincing reasons in writing for the departure, on facts proved beyond a reasonable doubt, and his determination would be reviewable on appeal. By contrast, because a 7-year sentence is within the presumptive range MILLER v. FLORIDA 433 423 Opinion of the Court under the revised law, the trial judge did not have to provide any reasons, convincing or otherwise, for imposing the sentence, and his decision was unreviewable. Thus, even if the revised guidelines law did not “technically . . . increase . . . the punishment annexed to [petitioner’s] crime,” Lindsey, supra, at 401, it foreclosed his ability to challenge the imposition of a sentence longer than his presumptive sentence under the old law. Petitioner therefore was “substantially disadvantaged” by the retrospective application of the revised guidelines to his crime. Finally, even if a law operates to the defendant’s detriment, the ex post facto prohibition does not restrict “legislative control of remedies and modes of procedure which do not affect matters of substance.” Dobbert, 432 U. S., at 293. Hence, no ex post facto violation occurs if the change in the law is merely procedural and does “not increase the punishment, nor change the ingredients of the offence or the ultimate facts necessary to establish guilt.” Hopt n. Utah, 110 U. S. 574, 590 (1884). See Dobbert, supra, at 293-294 (“The new statute simply altered the methods employed in determining whether the death penalty was to be imposed; there was no change in the quantum of punishment attached to the crime”). On the other hand, a change in the law that alters a substantial right can be ex post facto “even if the statute takes a seemingly procedural form.” Weaver, 450 U. S., at 29, n. 12. Although tile distinction between substance and procedure might sometimes prove elusive, here the change at issue appears to have little about it that could be deemed procedural. The 20% increase in points for sexual offenses in no wise alters the method to be followed in determining the appropriate sentence; it simply inserts a larger number into the same equation. The comments of the Florida Supreme Court acknowledge that the sole reason for the increase was to punish sex offenders more heavily: the amendment was intended to, 434 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. and did, increase the “quantum of punishment” for category 2 crimes. See 451 So. 2d, at 824, n. Respondent objects that it is misleading to view the change in the revised guidelines apart from the sentencing scheme as a whole. Relying largely on decisions by the Courts of Appeals sustaining the United States Parole Commission’s guidelines against ex post facto claims, respondent urges that the revised guidelines “merely guide and channel” the sentencing judge’s discretion. Brief for Respondent 35. See, e. g., Wallace v. Christensen, 802 F. 2d 1539 (CA9 1986) (en banc); Yamamoto v. United States Parole Common, 794 F. 2d 1295 (CA8 1986); Dufresne n. Baer, 744 F. 2d 1543 (CA11 1984), cert, denied, 474 U. S. 817 (1985); Warren v. United States Parole Common, 212 U. S. App. D. C. 137, 659 F. 2d 183 (1981), cert, denied, 455 U. S. 950 (1982). See also Portley v. Grossman, 444 U. S. 1311 (1980) (Rehnquist, J., in chambers). Invoking the reasoning of these cases, respondent contends that an increase in the guidelines sentence operates only as a “procedural guidepost” for the exercise of discretion within the same statutorily imposed sentencing limits. We find the federal parole guidelines cases inapposite. The courts that have upheld the retrospective application of federal parole guidelines have articulated several reasons why the ex post facto prohibition does not apply. The majority of these courts have held that the federal parole guidelines are not “laws” for purposes of the Ex Post Facto Clause. See, e. g., Wallace v. Christensen, supra, at 1553-1554 (citing cases). Other courts have found that the guidelines merely rationalize the exercise of statutory discretion, and that retrospective application of the guidelines thus does not violate the Ex Post Facto Clause. See, e. g., Warren v. United States Parole Common, supra, at 149, 659 F. 2d, at 195; Portley v. Grossman, supra, at 1312. Finally, some of the cases have held that retrospective application of the guidelines does not result in a more onerous punishment and MILLER v. FLORIDA 435 423 Opinion of the Court thus does not constitute an ex post facto violation. See, e. g., Dufresne n. Baer, supra, at 1549-1550. None of the reasons given in the federal parole cases even arguably applies here. First, the revised sentencing law is a law enacted by the Florida Legislature, and it has the force and effect of law. Cf. Williams v. State, 500 So. 2d 501, 503 (Fla. 1986) (departure sentence not supported by clear and convincing reasons was erroneous even though defendant consented, because “a defendant cannot . . . confer on the court the authority to impose an illegal sentence”). Nor do the revised guidelines simply provide flexible “guideposts” for use in the exercise of discretion: instead, they create a high hurdle that must be cleared before discretion can be exercised, so that a sentencing judge may impose a departure sentence only after first finding “clear and convincing reasons” that are “credible,” “proven beyond a reasonable doubt,” and “not... a factor which has already been weighed in arriving at a presumptive sentence.” See State v. Mischler, 488 So. 2d, at 525; Williams v. State, 492 So. 2d 1308, 1309 (Fla. 1986). Compare S. Rep. No. 98-225, p. 38 (1983) (describing the “unfettered discretion” of the Parole Commission under the system of parole guidelines). Finally, the revised guidelines directly and adversely affect the sentence petitioner receives. Thus, this is not a case where we can conclude, as we did in Dobbert, that “[t]he crime for which the present defendant was indicted, the punishment prescribed therefor, and the quantity or the degree of proof necessary to establish his guilt, all remained unaffected by the subsequent statute.” 432 U. S., at 294. Ill The law at issue in this case, like the law in Weaver, makes more onerous the punishment for crimes committed before its enactment.” Weaver, supra, at 36. Accord-My, we find that Florida’s revised guidelines law, 1984 Fla. Laws, ch. 84-328, is void as applied to petitioner, whose 436 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. crime occurred before the law’s effective date. We reverse the judgment of the Supreme Court of Florida, and remand the case for further proceedings not inconsistent with this opinion. It is so ordered. CRAWFORD FITTING CO. v. J. T. GIBBONS, INC. 437 Syllabus CRAWFORD FITTING CO. ET AL. v. J. T. GIBBONS, INC. CERTIORARI TO THE UNITED STATES COURT OF APPEALS FOR THE FIFTH CIRCUIT No. 86-322. Argued April 29, 1987—Decided June 15, 1987* Title 28 U. S. C. § 1920 provides that a federal court “may tax” specified items, including witness fees, as costs against the losing party, and § 1821(b) states that a witness “shall be paid” a fee of $30 per day for court attendance. Federal Rule of Civil Procedure 54(d) provides in part: “Except when express provision therefor is made either in a statute of the United States or in these rules, costs shall be allowed as of course to the prevailing party unless the court otherwise directs.” In No. 86-322, petitioners prevailed as the defendants in an antitrust action filed by respondent, and the Federal District Court awarded, as part of petitioners’ costs, an amount for expert witness fees in excess of § 1821(b)’s $30-per-day limit, holding that Rule 54(d) granted it discretion to exceed such limit. The Court of Appeals reversed, concluding that § 1821(b)’s limit controlled. In No. 86-328, petitioner prevailed in an action against it by respondents for alleged violations of federal civil rights statutes. The Federal District Court refused to order respondents to reimburse petitioner for its expert witness fees to the extent they exceeded the $30-per-day limit, and the Court of Appeals affirmed. Held: When a prevailing party seeks reimbursement for fees paid to its expert witnesses, a federal court is bound by the limits of § 1821(b), absent contract or explicit statutory authority to the contrary. There is no merit to petitioners’ contentions that, since § 1920 lists expenses which a court “may” tax as costs, it only authorizes taxation of such items and does not preclude taxation for other items or amounts in excess of the § 1821(b) fee; and that the discretion granted by Rule 54(d) is a separate power to tax expenses as costs. If Rule 54(d) were so construed, § 1920 would serve no role whatsoever. The better view is that § 1920 defines the term “costs” as used in Rule 54(d) and enumerates expenses that a federal court may tax as costs under the discretionary authority found in Rule 54(d). Section 1920 is phrased permissibly because Rule 54(d) generally grants a federal court discretion to refuse to tax costs in favor of *Together with No. 86-328, Champion International Corp. v. International Woodworkers of America, AFL-CIO, CLC, et al., also on certiorari to the same court. 438 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. the prevailing party. Such discretion is not a power to evade the specific congressional command limiting the amount of witness fees. Rather, it is solely a power to decline to tax, as costs, the items enumerated in § 1920. The dictum to the contrary in Farmer v. Arabian American Oil Co., 379 U. S. 227, is disapproved. Henkel v. Chicago, S. P., M. & 0. R. Co., 284 U. S. 444—which held that federal courts had no authority to award expert witness fees in excess of the 1853 statutory limit—controls here, even though it was decided before the adoption of the Federal Rules of Civil Procedure and the merger of law and equity in the federal courts. Cf. Alyeska Pipeline Service Co. v. Wilderness Society, 421 U. S. 240. Pp. 441-445. 790 F. 2d 1193, affirmed and remanded; and 790 F. 2d 1174, affirmed. Rehnquist, C. J., delivered the opinion of the Court, in which White, Blackmun, Powell, Stevens, O’Connor, and Scalia, JJ., joined. Blackmun, J. filed a concurring opinion, post, p. 445. Marshall, J., filed a dissenting opinion, in which Brennan, J., joined, post, p. 445. Ernest P. Mansour argued the cause for petitioners in No. 86-322. With him on the briefs were Dando B. Cellini and Victoria Knight McHenry. Jeffrey A. Walker argued the cause for petitioner in No. 86-328. With him on the briefs was Miles Curtiss McKee. William H. Block argued the cause and filed a brief for respondent in No. 86-322. James E. Youngdahl argued the cause and filed a brief for respondents in No. 86-328. t Chief Justice Rehnquist delivered the opinion of the Court. In these two consolidated cases we address the power of federal courts to require a losing party to pay the compensation of the winner’s expert witnesses. In No. 86-322, respondent J. T. Gibbons, Inc., sued petitioner Crawford Fitting Co. and other petitioners for alleged violations of the ÌJames Robertson, Harold R. Tyler, Jr., Norman Redlich, William L. Robinson, Richard T. Seymour, Antonia Hernandez, E. Richard Larson, Steven L. Winter, Julius LeVonne Chambers, and Charles Stephen Ralston filed a brief for the NAACP Legal Defense and Educational Fund, Inc., et al. as amici curiae urging affirmance in No. 86-328. CRAWFORD FITTING CO. v. J. T. GIBBONS, INC. 439 437 Opinion of the Court antitrust laws. The District Court directed a verdict in favor of petitioners. 565 F. Supp. 167 (ED La. 1981), aff’d, 704 F. 2d 787 (CA5 1983). Petitioners then filed a bill of costs with the Clerk of that court, seeking reimbursement from respondent for over $220,000 in litigation expenses, including substantial expert witness fees. The District Court held that Federal Rule of Civil Procedure 54(d) granted it discretion to exceed the $30-per-day witness fee limit found in 28 U. S. C. § 1821(b). It accordingly awarded petitioners $86,480.70 for their expert witnesses. 102 F. R. D. 73 (ED La. 1984). En banc, the Court of Appeals for the Fifth Circuit reversed, holding that the limit of § 1821(b) controlled. 790 F. 2d 1193 (1986). In No. 86-328, respondent International Woodworkers of America (IWA) sued petitioner Champion International, alleging racial discrimination in violation of Title VII of the Civil Rights Act of 1964 and 42 U. S. C. § 1981. After a trial on the merits, the District Court dismissed all of respondent’s claims. Petitioner thereafter filed a bill of costs, including $11,807 in expert witness fees. The District Court declined to order respondent to reimburse petitioner for these fees to the extent they exceeded the $30-per-day limit. The en banc Court of Appeals for the Fifth Circuit affirmed, finding the limit set forth in § 1821(b) dispositive. 790 F. 2d 1174 (1986). We agree and hold that when a prevailing party seeks reimbursement for fees paid to its own expert witnesses, a federal court is bound by the limit of § 1821(b), absent contract or explicit statutory authority to the contrary. In 1793 Congress enacted a general provision linking some taxable costs in most cases in federal courts to the practice of the courts of the State in which the federal court sat. Act of Mar. 1, 1793, §4, 1 Stat. 333. This provision expired in 1799. Apparently from 1799 until 1853 federal courts continued to refer to state rules governing taxable costs. See ^yeska Pipeline Service Co. v. Wilderness Society, 421U. S. 240, 250 (1975). By 1853 there was a “great diversity 440 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. in practice among the courts” and “losing litigants were being unfairly saddled with exorbitant fees.” Id., at 251. Accordingly, Congress returned to the issue and comprehensively regulated fees and the taxation of fees as costs in the federal courts. The resulting 1853 Fee Act “was a far-reaching Act specifying in detail the nature and amount of the taxable items of cost in the federal courts.” 421 U. S., at 251-252. It provided, in part, “That in lieu of the compensation now allowed by law to attorneys, solicitors, . . . and . . . witnesses ... in the several States, the following and no other compensation shall be taxed and allowed.” Act of Feb. 26, 1853, 10 Stat. 161. The rate for witnesses was set at $1.50 per day. 10 Stat. 167. The sweeping reforms of the 1853 Act have been carried forward to today, “without any apparent intent to change the controlling rules.” Alyeska Pipeline, supra, at 255. Title 28 U. S. C. § 1920 now embodies Congress’ considered choice as to the kinds of expenses that a federal court may tax as costs against the losing party: “A judge or clerk of any court of the United States may tax as costs the following: “(1) Fees of the clerk and marshal; “(2) Fees of the court reporter for all or any part of the stenographic transcript necessarily obtained for use in the case; “(3) Fees and disbursements for printing and witnesses; “(4) Fees for exemplification and copies of papers necessarily obtained for use in the case; “(5) Docket fees under section 1923 of this title; “(6) Compensation of court appointed experts, compensation of interpreters, and salaries, fees, expenses, and costs of special interpretation services under section 1828 of this title.” The witness fee specified in § 1920(3) is defined in 28 U. S. C. § 1821: CRAWFORD FITTING CO. u J. T. GIBBONS, INC. 441 437 Opinion of the Court “(a)(1) Except as otherwise provided by law, a witness in attendance at any court of the United States . . . shall be paid the fees and allowances provided by this section. “(b) A witness shall be paid an attendance fee of $30 per day for each day’s attendance. A witness shall also be paid the attendance fee for the time necessarily occupied in going to and returning from the place of attendance at the beginning and end of such attendance or at any time during such attendance.” Federal Rule of Civil Procedure 54(d) in turn provides in part: “Except when express provision therefor is made either in a statute of the United States or in these rules, costs shall be allowed as of course to the prevailing party unless the court otherwise directs.” The logical conclusion from the language and interrelation of these provisions is that § 1821 specifies the amount of the fee that must be tendered to a witness, § 1920 provides that the fee may be taxed as a cost, and Rule 54(d) provides that the cost shall be taxed against the losing party unless the court otherwise directs. Petitioners argue that since § 1920 lists which expenses a court “may” tax as costs, that section only authorizes taxation of certain items. In their view, § 1920 does not preclude taxation of costs above and beyond the items listed, and more particularly, amounts in excess of the § 1821(b) fee. Thus, the discretion granted by Rule 54(d) is a separate source of power to tax as costs expenses not enumerated in §1920. We think, however, that no reasonable reading of these provisions together can lead to this conclusion, for petitioners’ view renders § 1920 superfluous. If Rule 54(d) grants courts discretion to tax whatever costs may seem appropriate, then § 1920, which enumerates the costs that may be taxed, serves no role whatsoever. We think the better view is that § 1920 defines the term “costs” as used in Rule 54(d). Section 1920 enumerates expenses that a federal court may tax as a cost under the discretionary authority 442 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. found in Rule 54(d). It is phrased permissively because Rule 54(d) generally grants a federal court discretion to refuse to tax costs in favor of the prevailing party. One of the items enumerated in § 1920 is the witness fee, set by § 1821(b) at $30 per day. We cannot accept an interpretation of Rule 54(d) that would render any of these specific statutory provisions entirely without meaning. Repeals by implication are not favored, and petitioners proffer the ultimate in implication, for Rule 54(d) and §§ 1920 and 1821 are not even inconsistent. We think that it is clear that in §§ 1920 and 1821, Congress comprehensively addressed the taxation of fees for litigants’ witnesses. This conclusion is all the more compelling when we consider that § 1920(6) allows the taxation, as a cost, of the compensation of court-appointed expert witnesses. There is no provision that sets a limit on the compensation for court-appointed expert witnesses in the way that § 1821(b) sets a limit for litigants’ witnesses. It is therefore clear that when Congress meant to set a limit on fees, it knew how to do so. We think that the inescapable effect of these sections in combination is that a federal court may tax expert witness fees in excess of the $30-per-day limit set out in § 1821(b) only when the witness is court-appointed. The discretion granted by Rule 54(d) is not a power to evade this specific congressional command. Rather, it is solely a power to decline to tax, as costs, the items enumerated in § 1920. The logic of this conclusion notwithstanding, petitioners place heavy weight on a single sentence found in our opinion in Farmer n. Arabian American Oil Co., 379 U. S. 227 (1964). In that case this Court held that the District Court had not abused its discretion in refusing to tax against the losing plaintiff the travel expenses of witnesses for the defendant. In the course of so ruling, the Court stated: “[T]he discretion given district judges [by Rule 54(d)] to tax costs should be sparingly exercised with reference to CRAWFORD FITTING CO. v. J. T. GIBBONS, INC. 443 437 Opinion of the Court expenses not specifically allowed by statute.” Id., at 235. Applying this language to the present case, petitioners argue that courts therefore have discretion to tax as costs expenses incurred beyond those specified by Congress as fees in § 1821, and made taxable by § 1920. The sentence relied upon is classic obiter: something mentioned in passing, which is not in any way necessary to the decision of the issue before the Court. We think the dictum is inconsistent with the foregoing analysis, and we disapprove it. The argument petitioners present today was squarely rejected in Henkel v. Chicago, S. P., M. & 0. R. Co., 284 U. S. 444 (1932). In that case, the Court held that federal courts have no authority to award expert witness fees in excess of the statutory limit set by Congress in the Fee Act of 1853. The Court’s reasoning was straightforward: “Specific provision as to the amounts payable and taxable as witness fees was made by Congress as early as the Act of February 28, 1799 .... Under these provisions, additional amounts paid as compensation, or fees, to expert witnesses cannot be allowed or taxed as costs in cases in the federal courts. “. . . Congress has dealt with the subject comprehensively and has made no exception of the fees of expert witnesses. Its legislation must be deemed controlling ... Id., at 446-447. Petitioners contend that because Henkel was decided before the merger of law and equity in the federal courts, it is no longer good law. Petitioners’ argument proceeds along the following lines: Prior to the adoption of the Federal Rules of Civil Procedure, federal courts could sit in law or in equity, n petitioners’ view, courts sitting in equity had broad discretion to award fees not specified by statute. Henkel, decided 444 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. under this regime, held that courts at law had no power to exceed the limits set by statute. Now that the federal courts’ legal and equitable powers are combined, petitioners conclude that Henkel cannot control the scope of a federal court’s powers to exceed the limits set by statute. We cannot agree. Henkel rested on statutory interpretation. Whatever the effect of the merger of law and equity in federal courts, it did not repeal any part of the Fee Act. Title 28 U. S. C. §§ 1920 and 1821, today’s counterparts to the provisions of the Fee Act at issue in Henkel, are still law, and when not overridden by contract or explicit statutory authority, they control a federal court’s power to hold a losing party responsible for the opponent’s witness fees. Our conclusion conforms to our prior interpretations of the 1853 Fee Act. In Alyeska Pipeline Service Co. v. Wilderness Society, 421 U. S. 240 (1975), we considered the general role of the Act in federal courts. The Act “specified] in detail the nature and amount of the taxable items of cost in the federal courts.” Id., at 252. The comprehensive scope of the Act and the particularity with which it was drafted demonstrated to us that Congress meant to impose rigid controls on cost-shifting in federal courts. Thus, we rejected an argument similar to the one posited by petitioners today: “Nor has [Congress] extended any roving authority to the Judiciary to allow counsel fees as costs or otherwise whenever the courts might deem them warranted.” Id., at 260. Although Congress responded to our decision in Alyeska by broadening the availability of attorney’s fees in the federal courts, see the Civil Rights Attorney’s Fees Awards Act of 1976, 90 Stat. 2641, 42 U. S. C. § 1988, it has not otherwise “retracted, repealed, or modified the limitations on taxable fees contained in the 1853 statute and its successors.” 421 U. S., at 260. Thus, we are once again asked to hold that a specific congressional enactment on the shifting of litigation costs is of no moment. We think that, as in Alyeska, Congress has made its intent plain in its detailed treatment of CRAWFORD FITTING CO. v. J. T. GIBBONS, INC. 445 437 Marshall, J., dissenting witness fees. We will not lightly infer that Congress has repealed §§1920 and 1821, either through Rule 54(d) or any other provision not referring explicitly to witness fees. As always, “ ‘[w]here there is no clear intention otherwise, a specific statute will not be controlled or nullified by a general one, regardless of the priority of enactment.’” Radzanower n. Touche Ross & Co., 426 U. S. 148, 153 (1976), quoting Morton v. Mancari, 417 U. S. 535, 550-551 (1974) (emphasis added). Any argument that a federal court is empowered to exceed the limitations explicitly set out in §§ 1920 and 1821 without plain evidence of congressional intent to supersede those sections ignores our longstanding practice of construing statutes in pari materia. See United States v. United Continental Tuna Corp., 425 U. S. 164, 168-169 (1976); Train v. Colorado Public Interest Research Group, 426 U. S. 1, 24 (1976). We hold that absent explicit statutory or contractual authorization for the taxation of the expenses of a litigant’s witness as costs, federal courts are bound by the limitations set out in 28 U. S. C. § 1821 and § 1920. The judgments of the Court of Appeals are affirmed, and No. 86-322 is remanded for further proceedings consistent with this opinion. It is so ordered. Justice Blackmun, concurring. I join the Court’s opinion and its judgment but upon the understanding that it does not reach the question whether, under 42 U. S. C. § 1988, a district court may award fees for an expert witness. See post, at 446, n. 1 (Marshall, J., dissenting). Justice Marshall, with whom Justice Brennan joins, dissenting. In these two cases, prevailing defendants sought reimbursement for expert witness fees pursuant to Federal Rule of Civil Procedure 54(d). The Rule provides that “[e]x- 446 OCTOBER TERM, 1986 Marshall, J., dissenting 482 U. S. cept when express provision therefor is made either in a statute of the United States or in these rules, costs shall be allowed as of course to the prevailing party unless the court otherwise directs.” In No. 86-322, the District Court found that some of the expert testimony was “indispensable to the determination of [the] issues in the case” and taxed against the plaintiff the portion of witness fees attributable to that testimony. 102 F. R. D. 73, 86 (ED La. 1984). In No. 86-328, even though the District Court found the defendant’s expert was “helpful and perhaps necessary to its case,” the court declined to award fees in excess of the amounts specified in 28 U. S. C. § 1821. Civ. Action No. WC 78-33-WK-P (ND Miss., Aug. 24, 1983), p. 9. The Court now informs us that the District Courts had no power to award costs not expressly authorized by statute.1 In its haste to extinguish all discretion to award these non-statutory costs, however, the Court has rendered Rule 54(d) a nullity. Before today, it was generally recognized that the “unless the court otherwise directs” language in Rule 54(d) was in- 11 do not understand today’s decision to decide the question whether a district court may award expert witness fees under 42 U. S. C. § 1988. No. 86-322 is an antitrust case; obviously, § 1988 is not at issue in that case. And, as an examination of the record reveals, the issue is not properly before the Court in No. 86-328, either. In that case, petitioner, a prevailing civil rights defendant, made a motion for attorney’s fees "and expenses” under § 1988 and filed a bill of costs under Rule 54(d). The bill of costs included $31,333.87 for "expert witness fees and expenses.” Record 38. On December 30, 1982, the District Court summarily denied the motion for attorney’s fees and expenses, based on its conclusion that, under Christiansburg Garment Co. v. EEOC, 434 U. S. 412 (1978), “the lawsuit was brought in good faith and was neither frivolous, unreasonable, nor without foundation.” Record 1. The court referred all other questions concerning the taxing of costs to a Magistrate. Id., at 2. Petitioner did not appeal the District Court’s order denying attorney’s fees under § 1988. It appealed only the District Court’s order of August 24, 1983, denying its application for expert witness fees under Rule 54(d). See Record 33. CRAWFORD FITTING CO. v. J. T. GIBBONS, INC. 447 437 Marshall, J., dissenting tended as a grant of discretion to the district courts. See, e. g., 10 C. Wright, A. Miller, & M. Kane, Federal Practice and Procedure §2665, p. 171 (2d ed. 1983). Except where expressly prohibited by statute from doing so, the Rule “vests in the district court a sound discretion over the allowance, disallowance, or apportionment of costs in all civil actions.” 6 J. Moore, W. Toeggart, & J. Wicker, Moore’s Federal Practice 1J54.70[5], p. 54-331 (2d ed. 1987). This is because Rule 54(d) adopts the practice formerly followed in equity, see 6 Moore 5I54.70[3], p. 54-321; 10 Wright §2668, pp. 197-200, where courts possessed the power to award costs not expressly provided by statute, as “part of the original authority of the chancellor to do equity in a particular situation.” Sprague n. Ticonic National Bank, 307 U. S. 161, 166 (1939) (footnote omitted). See generally Newton v. Consolidated Gas Co., 265 U. S. 78, 83 (1924); Ex parte Peterson, 253 U. S. 300, 317-318 (1920). Since the adoption of the Federal Rules, this Court has addressed the scope of the district courts’ power to tax costs on only one occasion. In Farmer n. Arabian American Oil Co., 379 U. S. 227 (1964), the Court held that a District Court acted within its discretion in refusing to tax a witness’ expenses for travel in excess of 100 miles as costs against an unsuccessful plaintiff.2 It expressly rejected the argument that a district court lacks the power to award these expenses as costs. The Court noted: “While this Rule could be far more definite as to what ‘costs shall be allowed,’ the words ‘unless the court otherwise directs’ quite plainly vest some power in the court to allow some ‘costs.’” Id., at 232. 2 Under Federal Rule of Civil Procedure 45(e), a district court’s power to compel attendance of witnesses extends only 100 miles. Relying on this Rule, District Courts had traditionally declined to tax as costs expenses of witnesses traveling more than 100 miles. See Farmer v. Arabian American Oil Co., 379 U. S., at 231-232. 448 OCTOBER TERM, 1986 Marshall, J., dissenting 482 U. S. In a sentence labeled dictum by the majority, the Court sought to provide guidance to the lower courts by explaining that this discretion “should be sparingly exercised with reference to expenses not specifically allowed by statute.” Id., at 235. As Judge Rubin observed below, “[t]he Court’s conclusion reveals its premise: Rule 54(d) gives the district court discretion to award costs not enumerated in § 1920.” 790 F. 2d 1174, 1190 (CA5 1986) (en banc) (concurring and dissenting). This is certainly how Justice Harlan, author of the dissent in Farmer, viewed the case. See 379 U. S., at 240 (“the foundation of today’s decision” is the “scope of the discretion of a district judge acting within his powers”). Rather than following Farmer, as it should, the majority relies on Henkel n. Chicago, S. P., M. & 0. R. Co., 284 U. S. 444 (1932). But Henkel provides no support for a restrictive interpretation of a district court’s power to award fees under Rule 54(d). The opinion in Henkel addressed the narrow question whether district courts had authority to tax expert witness fees as costs in an action at law. At the time, courts of law lacked the power to award costs not expressly granted by statute, see Ex parte Peterson, supra, at 317-318, although those sitting in equity could award such costs, as justice required, without regard to the fee statutes. Approaching the issue purely as a matter of statutory construction, the Court concluded that expert witness fees were included in and limited to the amounts prescribed by the predecessors to 28 U. S. C. §§1920 and 1821. 284 U. S., at 446-447. The majority acknowledges, as it must, that Henkel was decided before the Federal Rules of Civil Procedure effected a merger of law and equity. What the majority ignores, however, is the vital significance of that fact. As noted above, Rule 54(d) adopts the practice in equity, thereby giving federal courts in all actions the broad discretion previously afforded only to courts exercising equitable powers. CRAWFORD FITTING CO. v. J. T. GIBBONS, INC. 449 437 Marshall, J., dissenting The majority’s assertion that discretion can be exercised only “to refuse to tax costs in favor of the prevailing party,” ante, at 442, is plainly inconsistent with the equitable principles on which Rule 54(d) is based. Moreover, it reinforces the fact that the Rule is now entirely superfluous. Because the language of §1920 is permissive—“[a] judge or clerk of any court of the United States may tax as costs the following”—courts already have discretion to disallow the costs listed therein.3 As the Court noted in Farmer, Rule 54(d) does not define “costs.” 379 U. S., at 232. Seizing on this “omission,” the Court now declares that §1920 sets forth the universe of “costs” taxable under the Rule. Ante, at 441-442. Any contrary interpretation, it claims, “renders § 1920 superfluous.” Ante, at 441. This misreads § 1920. That section does not purport to be exclusive. It does not direct that “the following costs and no others may be taxed.”4 By contrast, the predecessor to § 1920, the 1853 Fee Act, provided that “the following and no other compensation shall be taxed and allowed,” Act of Feb. 26,1853,10 Stat. 161 (emphasis added); this language was omitted from the 1948 revision. Despite this seemingly significant deletion, the majority contends that “[t]he sweeping reforms of the 1853 Act have been carried forward to today, ‘without any apparent intent to change the controlling rules.’” Ante, at 440, quoting Alyeska Pipeline Co. v. Wilderness Society, 421 U. S. 240, 255 (1975). In Alyeska, this Court held that the same fee statutes did not 8 The legislative history of § 1920 supports this view of Rule 54(d). Congress replaced the mandatory language found in the earlier version— shall tax costs”—to conform to the discretion afforded by Rule 54(d). See H. R. Rep. No. 308, 80th Cong., 1st Sess., App. A162 (1947) (Reviser’s Note). ‘Despite the majority’s protestations, refusing to construe § 1920 as the exclusive definition of costs would not render the statute superfluous. Its principal purpose is to set forth those routine, readily determinable costs which, in ordinary cases, will automatically be taxed by the clerk of the court. 450 OCTOBER TERM, 1986 Marshall, J., dissenting 482 U. S. authorize recovery of attorney’s fees by a prevailing party. Even in Alyeska, however, the Court recognized that the fee statutes had never been entirely exclusive: “To be sure, the fee statutes have been construed to allow, in limited circumstances, a reasonable attorney’s fee to the prevailing party in excess of the small sums [for docket fees] permitted by §1923.” Id., at 257.6 Not only is the Court’s holding inconsistent with the language and history of Rule 54(d) and § 1920, but it is also ill advised as a policy matter. As Judge Rubin stated in his opinion below: “The costs of litigation, as we all know, have become staggering. A plaintiff may put a defendant or a defendant may put a plaintiff to a tremendous amount of expense, apart from the cost of obtaining an attorney’s services, in defending or prosecuting a case. One cause of this expense is the unavoidable necessity of expert witness testimony to establish or rebut many legal claims. “Although the victor in litigation is not entitled to spoils, he ought at least to be able to invoke the court’s discretion to make him whole.” 790 F. 2d, at 1192-1193. For the foregoing reasons, I dissent. 6 With respect to fees, Alyeska identified three circumstances appropriate for such “assertions of inherent power in the courts,” Alyeska Pipeline Co., v. Wilderness Society, 421 U. S., at 259: when the trustee of a fund preserved or recovered the fund for the benefit of others in addition to him or herself; when a party acted in willful disobedience to a court order; or when the losing party acted in bad faith or vexatiously. Id., at 257-259. HOUSTON v. HILL 451 Syllabus CITY OF HOUSTON, TEXAS v. HILL APPEAL FROM THE UNITED STATES COURT OF APPEALS FOR THE FIFTH CIRCUIT No. 86-243. Argued March 23, 1987—Decided June 15, 1987 Upon shouting at police in an attempt to divert their attention from his friend during a confrontation, appellee was arrested for “wilfully . . . interrupt[ing] a city policeman ... by verbal challenge during an investigation” in violation of a municipal ordinance making it unlawful for any person “to assault, strike or in any manner oppose, molest, abuse or interrupt any policeman in the execution of his duty.” After his acquittal in Municipal Court, appellee brought suit in Federal District Court challenging the ordinance’s constitutionality and seeking, inter alia, damages and attorney’s fees. The District Court held that the ordinance was not unconstitutionally vague or overbroad on its face, but the Court of Appeals reversed, finding that the ordinance was substantially overbroad since its literal wording punished and might deter a significant range of protected speech. Held: 1. A municipal ordinance that makes it unlawful to interrupt a police officer in the performance of his duty is substantially overbroad and therefore invalid on its face under the First Amendment. The ordinance in question criminalizes a substantial amount of, and is susceptible of regular application to, constitutionally protected speech, and accords the police unconstitutional enforcement discretion, as is demonstrated by evidence indicating that, although the ordinance’s plain language is violated scores of times daily, only those individuals chosen by police in their unguided discretion are arrested. Appellant’s argument that the ordinance is not substantially overbroad because it does not inhibit the exposition of ideas, but simply bans unprotected “core criminal conduct,” is not persuasive. Since the ordinance’s language making it unlawful to “assault” or “strike” a police officer is expressly pre-empted by the State Penal Code, its enforceable portion prohibits verbal interruptions of police and thereby deals with speech rather than with core criminal conduct. Moreover, although speech might be prohibited if it consists of fighting words” that by their very utterance inflict injury or tend to incite an immediate breach of the peace, the ordinance in question is not limited to such expressions but broadly applies to speech that “in any manner . . . interrupt[s] any policeman” and thereby impermissibly infringes the constitutionally protected freedom of individuals verbally to 452 OCTOBER TERM, 1986 Syllabus 482 U. S. oppose or challenge police action. Appellant’s contention that the ordinance’s sweeping nature is both inevitable and essential to maintain public order is also without merit, since the ordinance is not narrowly tailored to prohibit only disorderly conduct or fighting words, but impermissibly provides police with unfettered discretion to arrest individuals for words or conduct that are simply annoying or offensive. Pp. 458-467. 2. Abstention—assertedly to allow the state courts to reach a readily available limiting construction that would eliminate the ordinance’s overbreadth—would be inappropriate here. Even if this case did not involve a First Amendment facial challenge, for which abstention is particularly inappropriate, the ordinance in question is plain and unambiguous and thus is not susceptible to a limiting construction. Moreover, it cannot be limited by severing discrete unconstitutional subsections since its enforceable portion is unconstitutional in its entirety. Even if the municipal courts had not had many opportunities to narrow the ordinance’s scope, appellant’s claim that state courts had not had the chance to construe the ordinance would be unavailing in light of the ordinance’s nonambiguity. Nor does the availability of certification to state courts under state law in itself render abstention appropriate where, as here, there is no uncertain question of state law to be resolved. Pp. 467-471. 3. Although the preservation of liberty depends in part upon the maintenance of social order, the First Amendment requires that officers and municipalities respond with restraint in the face of verbal challenges to police action, since a certain amount of expressive disorder is inevitable in a society committed to individual freedom and must be protected if that freedom would survive. Pp. 471-472. 789 F. 2d 1103, affirmed. Brennan, J., delivered the opinion of the Court, in which White, Marshall, Blackmun, and Stevens, JJ., joined. Blackmun, J., filed a concurring opinion, post, p. 472. Scalia, J., filed an opinion concurring in the judgment, post, p. 472. Powell, J., filed an opinion concurring in the judgment in part and dissenting in part, in which O’Connor, J., joined, in Parts I and II of which Rehnquist, C. J., joined, and in Parts II and III of which Scalia, J., joined, post, p. 473. Rehnquist, C. J., filed a dissenting opinion, post, p. 481. Robert J. Collins argued the cause for appellant. With him on the briefs was Jerry Edwin Smith. HOUSTON v. HILL 453 451 Opinion of the Court Charles Alan Wright argued the cause for appellee. With him on the brief were Michael A. Maness and Gerald M. Bimberg. * Justice Brennan delivered the opinion of the Court. This case presents the question whether a municipal ordinance that makes it unlawful to interrupt a police officer in the performance of his or her duties is unconstitutionally overbroad under the First Amendment. I Appellee Raymond Wayne Hill is a lifelong resident of Houston, Texas. At the time this lawsuit began, he worked as a paralegal and as executive director of the Houston Human Rights League. A member of the board of the Gay Political Caucus, which he helped found in 1975, Hill was also affiliated with a Houston radio station, and had carried city and county press passes since 1975. He lived in Montrose, a “diverse and eclectic neighborhood” that is the center of gay political and social life in Houston. App. 26-27. The incident that sparked this lawsuit occurred in the Montrose area on February 14, 1982. Hill observed a friend, Charles Hill, intentionally stopping traffic on a busy street, evidently to enable a vehicle to enter traffic. Two Houston police officers, one of whom was named Kelley, approached Charles and began speaking with him. According to the District Court, “shortly thereafter” Hill began shouting at the officers “in an admitted attempt to divert Kelley’s attention from Charles Hill.” App. to Juris. Statement B-2.1 Hill Alvin Bronstein, David Goldstein, Burt Neubome, James Harring-lon, and Bruce Griffiths filed a brief for the American Civil Liberties Union et al. as amici curiae urging affirmance. 1 Hill testified that his “motivation was to stop [the officers] from hitting Charles.” App. 37, 40. See n. 2, infra. He also explained: “I would rather that I get arrested than those whose careers can be damaged; I would rather that I get arrested than those whose families wouldn’t understand; I would rather that I get arrested than those who couldn’t spend a long time in jail. I am prepared to respond in any legal, nonaggressive or 454 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. first shouted: “Why don’t you pick on somebody your own size?” After Officer Kelley responded: “[A]re you interrupting me in my official capacity as a Houston police officer?” Hill then shouted: “Yes, why don’t you pick on somebody my size?” App. 40-41, 58, 71-74. Hill was arrested under Houston Code of Ordinances, §34-11(a), for “wilfully or intentionally interrupt[ing] a city policeman ... by verbal challenge during an investigation.” App. 2. Charles Hill was not arrested. Hill was then acquitted after a nonjury trial in Municipal Court.2 nonviolent way, to any illegal police activity, at any time, under any circumstances.” Id., at 29. 2 The District Court stated that Hill “shout[ed] abuses” at the officers, App. to Juris. Statement B-2 (emphasis added). As the Court of Appeals held, however, there is “no evidence to support the district court’s finding that Raymond [Hill] ‘shout[ed] abuses’ at Officer Kelley.” 789 F. 2d 1103, 1105 (CA5 1986). See App. 73-74 (testimony of Officer Kelley that Hill did not use “abusive” language). The testimony of Hill and Kelley is consistent in other ways ignored by the District Court. Both agree, for example, that Charles attempted to leave after an initial conversation with the officers, and that Kelley then grabbed Charles by the arm, turned him around, and told him not to walk away. Id., at 14, 57. According to Hill, Charles, who “has a nervous tic,” then went “into these spasms,” which prompted one of the officers to “screa[m]” at Charles “Are you making fun of me? ” Id., at 14-15. Kelley stated that Charles was “twitching” in an “erratic and strange” manner, and that Kelley “didn’t know if [Charles] was about to have a seizure or if he was being insolent or what.” Id., at 56-57. At this point, however, the testimony substantially diverges. Kelley states that Hill then “interrupte[d]” him with the verbal challenge quoted in text, and that a crowd was beginning to form. Id., at 57-58, 61, 68-69. Hill testified that both officers grabbed Charles, placed him up against a wall, and threatened to hit him with a large flashlight. Id., at 14. Only then, according to Hill, did he call out: “[T]he kid has done nothing wrong. If you want to pick on somebody, pick on me.” Id., at 16. We note the applicability of Justice Powell’s observation that there is a “possibility of abuse” where convictions under an ordinance frequently turn on the resolution of a “direct conflict of testimony as to ‘who said what.’ ” Lews v. City of New Orleans, 415 U. S. 130, 135, n. (1974) (Powell, J., concurring in result). See infra, at 466. HOUSTON v. HILL 455 451 Opinion of the Court Code of Ordinances, City of Houston, Texas, §34-ll(a) (1984), reads: “Sec. 34-11. Assaulting or interfering with policemen. “(a) It shall be unlawful for any person to assault, strike or in any manner oppose, molest, abuse or interrupt any policeman in the execution of his duty, or any person summoned to aid in making an arrest.”3 Following his acquittal in the Charles Hill incident, Hill brought the suit in the Federal District Court for the Southern District of Texas, seeking (1) a declaratory judgment that §34-11(a) was unconstitutional both on its face and as it had been applied to him, (2) a permanent injunction against any attempt to enforce the ordinance, (3) an order expunging the records of his arrests under the ordinance, and (4) damages and attorney’s fees under 42 U. S. C. §§ 1983 and 1988. At trial, Hill introduced records provided by the city regarding both the frequency with which arrests had been made for violation of the ordinance and the type of conduct with which those arrested had been charged. He also introduced evidence and testimony concerning the arrests of several reporters under the ordinance. Finally, Hill introduced evidence regarding his own experience with the ordinance, under which he has been arrested four times since 1975, but never convicted. The District Court held that Hill’s evidence did not demonstrate that the ordinance had been unconstitutionally applied.4 The court also rejected Hill’s contention that the A conviction under the ordinance is a misdemeanor punishable by a fine of not more than $200. App. to Juris. Statement B-l. 4 The facts of Hill’s other three arrests as found by the District Court are as follows. On August 31, 1975, Hill intentionally interrupted two Houston police officers as they made a traffic arrest. During the arrest, Hill wrote down license plate numbers, and then walked to within an arm’s length of one of the officers on the side nearest the officer’s revolver. The officer asked Hill to leave, but Hill instead moved closer. Hill was ar-rested, tried, and found not guilty. In 1977, after observing vice-squad cars parked near a bookstore, Hill entered the store and announced on the public address system that police 456 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. ordinance was unconstitutionally vague or overbroad on its face. The ordinance was not vague, the court stated, because: “[t]he wording of the ordinance is sufficiently definite to put a person of reasonable intelligence on fair notice of what actions are forbidden. In particular, the Court finds that the use of words such as ‘interrupt’ are sufficiently clear by virtue of their commonly-understood, everyday definitions. Interrupt commonly means to cause one to cease, such as stopping someone in the middle of something. The Plaintiff, for example, clearly ‘interrupted’ the police officers regarding the Charles Hill incident.” App. to Juris. Statement B-8. The court also held that the statute was not overbroad because “the ordinance does not, at least facially, proscribe speech or conduct which is protected by the First Amendment.” Id., at B-12. A panel of the Court of Appeals reversed. 764 F. 2d 1156 (CA5 1985). The city’s suggestion for rehearing en banc was granted, and the Court of Appeals, by a vote of 8-7, upheld the judgment of the panel. 789 F. 2d 1103 (1986). The Court of Appeals agreed with the District Court’s conclusion that the ordinance was not vague, and that it “plainly encom-passe[d] mere verbal as well as physical conduct.” Id., at 1109. Applying the standard established in Broadrick v. Oklahoma, 413 U. S. 601 (1973), however, the Court of Appeals concluded that the ordinance was substantially officers were present and that patrons should prepare to show their identification. The patrons promptly left the store, thereby frustrating the investigation. Hill was arrested for interfering with the investigation, but the case was subsequently dismissed. Finally, on October 3, 1982, eight months after the lawsuit began, Hill was arrested for refusing to leave the immediate area of a car with an unknown and unconscious person inside. The arresting officers failed to appear in Municipal Court, however, so the charge against Hill was dismissed. HOUSTON v. HILL 457 451 Opinion of the Court overbroad. It found that “[a] significant range of protected speech and expression is punishable and might be deterred by the literal wording of the statute.” 789 F. 2d, at 1110. The Court of Appeals also reviewed the evidence of the unconstitutional application of the ordinance which Hill had introduced at trial. The court did not disturb the District Court’s ruling that the statute had not been unconstitutionally applied to Hill or to the reporters. It did conclude, however, that other evidence not mentioned by the District Court revealed “a realistic danger of, and a substantial potential for, the unconstitutional application of the ordinance.” Ibid. This evidence showed that the ordinance “is officially regarded as penalizing the mere interruption of a policeman while in the line of duty,” id., at 1109, and has been employed to make arrests for, inter alia, “arguing,” “[t]alking,” “[i]n-terfering,” “[f Jailing to remain quiet,” “[r]efusing to remain silent,” “[v]erbal abuse,” “[c]ursing,” “[v]erbally yelling,” and “[t]alking loudly, [w]alking through scene.” Id., at 1113—1114.6 The city appealed, claiming that the Court of Appeals erred in holding the ordinance facially overbroad and in not abstaining until the ordinance had been construed by the ’These charges are summarized in an appendix to the opinion of the Court of Appeals, 789 F. 2d, at 1113-1114. The court noted that “[appellee] offered evidence of over 200 arrests that had been made for violation of the ordinance between November 1981 and March 1982. Violations are apparently so frequent that the City uses a printed form to report charges.” Id., at 1107. The form, entitled “Complaint: Interrupting a Policeman,” contains the preprinted charge of “wilfully or intentionally interrupting] a city policeman” that is followed by a blank in which the officer fills in a description of the basis for the charge. Id., at 1108-1109. While noting that the majority of those arrested are charged with conduct that is “patently unlawful,” the Court of Appeals observed that [i]n many instances . . . the malefactor is described [in the handwritten portion] as having done nothing more offensive to the public order than speaking or failing to remain silent.” Id., at 1109. Over a third of these arrests were never prosecuted. Id., at 1110. 458 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. state courts.6 We noted probable jurisdiction, 479 U. S. 811 (1986), and now affirm. II The elements of First Amendment overbreadth analysis are familiar. Only a statute that is substantially overbroad may be invalidated on its face. New York v. Ferber, 458 U. S. 747, 769 (1982); Broadrick v. Oklahoma, supra. “We have never held that a statute should be held invalid on its face merely because it is possible to conceive of a single impermissible application . .. .” Id., at 630 (Brennan, J., dissenting). Instead, “[i]n a facial challenge to the overbreadth and vagueness of a law, a court’s first task is to determine whether the enactment reaches a substantial amount of constitutionally protected conduct.” Hoffman Estates v. The Flipside, Hoffman Estates, Inc., 455 U. S. 489, 494 (1982); 6 The city also claims that the Court of Appeals engaged in improper factfinding. The city notes that the District Court found that the ordinance had not been unconstitutionally applied, and argues that the Court of Appeals erred in reviewing Hill’s evidence and concluding that it showed a potential for unconstitutional application. Such a conclusion was foreclosed, according to the city, by the “clearly erroneous” standard of Federal Rule of Civil Procedure 52(a). Brief for Appellant 40. This argument is without merit. An independent review of the record is appropriate where the activity in question is arguably protected by the Constitution. NAACP v. Claiborne Hardware Co., 458 U. S. 886, 915— 916, n. 50 (1982). Moreover, the Court of Appeals accepted as “not challenged on appeal” the District Court’s finding that the ordinance had not been unconstitutionally applied to Hill or to the reporters, 789 F. 2d, at 1107, 1110. The disagreement between the lower courts was therefore limited to a question of law—whether the ordinance on its face was substantially overbroad. In concluding that the ordinance was overbroad, the Court of Appeals did not err in reviewing evidence ignored by the District Court Concerning the application of the ordinance, and in concluding that this evidence demonstrated a significant potential for unconstitutional application of the ordinance. The question whether the ordinance has been unconstitutionally applied to Hill is neither presented by this appeal nor essential to our decision, and we do not address it. HOUSTON v. HILL 459 451 Opinion of the Court Kolender v. Lawson, 461 U. S. 352, 359, n. 8 (1983). Criminal statutes must be scrutinized with particular care, e. g., Winters v. New York, 333 U. S. 507, 515 (1948); those that make unlawful a substantial amount of constitutionally protected conduct may be held facially invalid even if they also have legitimate application. E. g., Kolender, supra, at 359, n. 8. The city’s principal argument is that the ordinance does not inhibit the exposition of ideas, and that it bans “core criminal conduct” not protected by the First Amendment. Brief for Appellant 12. In its view, the application of the ordinance to Hill illustrates that the police employ it only to prohibit such conduct, and not “as a subterfuge to control or dissuade free expression.” Ibid. Since the ordinance is “contentneutral,” and since there is no evidence that the city has applied the ordinance to chill particular speakers or ideas, the city concludes that the ordinance is not substantially overbroad.7 7 The city’s threshold argument that Hill lacks standing is without merit. The basis for the argument is the District Court’s finding that the ordinance has been constitutionally applied to Hill in the past. This finding is irrelevant, however, to the question of Hill’s standing to seek prospective relief. Hill has shown “a genuine threat of enforcement” of the ordinance against his future activities, Steffel v. Thompson, 415 U. S. 452, 475 (1974). Compare, e. g., n. 1, supra (testimony of Hill’s willingness to interrupt officers in the future), with Golden v. Zunckler, 394 U. S. 103 (1969) (intervening event rendered unlikely any future application of statute to appellee); see also App. to Juris. Statement B-3, n. 1 (District Court finding that Hill “is a gay rights activist who claims that the Houston police have ‘systematically’ harassed him ‘as the direct result’ of his sexual preferences”). Moreover, although we have never required that a plaintiff undergo a criminal prosecution” to obtain standing to challenge the facial validity of a statute, Doe v. Bolton, 410 U. S. 179, 188 (1973), the fact that Hill has already been arrested four times under the ordinance lends compelling support to the threat of future enforcement. We therefore agree with the Court of Appeals that “Hill’s record of arrests under the ordinance and his adopted role as citizen provocateur” give Hill standing to challenge the facial validity of the ordinance. 789 F. 2d, at 1107. Cf. Ellis v. tyson, 421 U. S. 426 (1975). 460 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. We disagree with the city’s characterization for several reasons. First, the enforceable portion of the ordinance deals not with core criminal conduct, but with speech. As the city has conceded, the language in the ordinance making it unlawful for any person to “assault” or “strike” a police officer is pre-empted by the Texas Penal Code. Reply Brief for Appellant 10. The city explains, ibid., that “any species of physical assault on a police officer is encompassed within the provisions [§§22.01, 22.02] of the Texas Penal Code,”8 and under §1.08 of the Code, “[n]o governmental subdivision or agency may enact or enforce a law that makes any conduct covered by this code an offense subject to a criminal penalty.” 8 One who assaults or strikes either a police officer or “any person summoned to aid in making the arrest” may be arrested and prosecuted either under Tex. Penal Code Ann. § 22.01 (1974 and Supp. 1987), which renders unlawful any provocative contact with (or assault or threatened assault against) any person, or under Tex. Penal Code Ann. § 22.02 (1974), which renders unlawful conduct causing bodily injury to a peace officer. These sections provide in pertinent part: “Section 22.01. Assault. “(a) A person commits an offense if the person: “(1) intentionally, knowingly, or recklessly causes bodily injury to another including the person’s spouse; or “(2) intentionally or knowingly threatens another with imminent bodily injury including the person’s spouse; or “(3) intentionally or knowingly causes physical contact with another when the person knows or should reasonably believe that the other will regard the contact as offensive or provocative.” “Section 22.02. Aggravated Assault. “(a) A person commits an offense if he commits assault as defined in Section 22.01 of this code and he: “(1) causes serious bodily injury to another; “(2) causes bodily injury to a peace officer in the lawful discharge of official duty when he knows or has been informed the person is a peace officer; or “(3) uses a deadly weapon. “(b) The actor is presumed to have known the person assaulted was a peace officer if he was wearing a distinctive uniform indicating his employment as a peace officer.” HOUSTON v. HILL 461 451 Opinion of the Court Tex. Penal Code Ann. § 1.08 (1974). See Knott v. State, 648 S. W. 2d 20 (Tex. App. 1983) (reversing conviction obtained under municipal ordinance pre-empted by state penal code). Accordingly, the enforceable portion of the ordinance makes it “unlawful for any person to ... in any manner oppose, molest, abuse or interrupt any policeman in the execution of his duty,” and thereby prohibits verbal interruptions of police officers.9 Second, contrary to the city’s contention, the First Amendment protects a significant amount of verbal criticism and challenge directed at police officers. “Speech is often provocative and challenging. . . . [But it] is nevertheless protected against censorship or punishment, unless shown likely to produce a clear and present danger of a serious substantive evil that rises far above public inconvenience, annoyance, or unrest.” Terminiello v. Chicago, 337 U. S. 1, 4 (1949). In Lewis n. City of New Orleans, 415 U. S. 130 (1974), for example, the appellant was found to have yelled obscenities and threats at an officer who had asked appellant’s husband to produce his driver’s license. Appellant was convicted under a municipal ordinance that made it a crime “ ‘for any person wantonly to curse or revile or to use obscene or opprobrious language toward or with reference to any member of the city police while in the actual performance of his duty.’” Id., at 132 (citation omitted). We vacated the conviction and invalidated the ordinance as facially overbroad. Critical to our decision was the fact that the ordinance “punishe[d] only spoken words” and was not limited m scope to fighting words that “‘by their very utterance 9 It is this portion of the ordinance to which Hill directed his constitutional challenge, see 6 and 27 of his complaint. Record 138, 144-145. The Court of Appeals did not address the pre-emption issue; it assumed that the ordinance prohibited physical as well as verbal assaults, and still found the ordinance substantially overbroad. 789 F. 2d, at 1109. Because the city conceded pre-emption in this Court, see Reply Brief for Appellant 10, we need not address the question whether the ordinance, if not partially pre-empted, would be substantially overbroad. 462 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. inflict injury or tend to incite an immediate breach of the peace.’” Id., at 133, quoting Gooding n. Wilson, 405 U. S. 518, 525 (1972); see also ibid. (Georgia breach-of-peace statute not limited to fighting words held facially invalid). Moreover, in a concurring opinion in Lewis, Justice Powell suggested that even the “fighting words” exception recognized in Chaplinsky v. New Hampshire, 315 U. S. 568 (1942), might require a narrower application in cases involving words addressed to a police officer, because “a properly trained officer may reasonably be expected to ‘exercise a higher degree of restraint’ than the average citizen, and thus be less likely to respond belligerently to ‘fighting words.’” 415 U. S., at 135 (citation omitted). The Houston ordinance is much more sweeping than the municipal ordinance struck down in Lewis. It is not limited to fighting words nor even to obscene or opprobrious language, but prohibits speech that “in any manner . . . inter-rupt[s]” an officer.10 The Constitution does not allow such speech to be made a crime.11 The freedom of individuals ver 10 To the extent the ordinance could be interpreted to ban fighting words, it is pre-empted by Tex. Penal Code Ann. § 1.08 (1974), which preempts municipal laws that prohibit conduct subject to penalty under the Code, see supra, at 460-461, and by § 42.01, the State’s comprehensive disorderly conduct provision. Subsection § 42.01(a)(1), which makes unlawful “abusive, indecent, profane or vulgar language” only if “by its very utterance [it] tends to incite an immediate breach of the peace,” prohibits the use of fighting words. The “practice commentary” in the annotated Code confirms that this section is designed to track the “fighting words” exception set forth in Chaplinsky v. New Hampshire, 315 U. S. 568 (1942). Tex. Penal Code Ann. §42.01, pp. 124-125 (1974 and Supp. 1987). 11 Justice Powell suggests that our analysis of protected speech sweeps too broadly. But if some constitutionally unprotected speech must go unpunished, that is a price worth paying to preserve the vitality of the First Amendment. “ ‘[I]f absolute assurance of tranquility is required, we may as well forget about free speech. Under such a requirement, the only “free” speech would consist of platitudes. That kind of speech does not HOUSTON v. HILL 463 451 Opinion of the Court bally to oppose or challenge police action without thereby risking arrest is one of the principal characteristics by which we distinguish a free nation from a police state.12 need constitutional protection.’” Spence v. Washington, 418 U. S. 405, 416 (1974) (Douglas, J., concurring) (citation omitted). In any case, today’s decision does not leave municipalities powerless to punish physical obstruction of police action. For example, Justice Powell states that “a municipality constitutionally may punish an individual who chooses to stand near a police officer and persistently attempt to engage the officer in conversation while the officer is directing traffic at a busy intersection.” Post, at 479. We agree, however, that such conduct might constitutionally be punished under a properly tailored statute, such as a disorderly conduct statute that makes it unlawful to fail to disperse in response to a valid police order or to create a traffic hazard. E. g., Gotten v. Kentucky, 407 U. S. 104 (1972). What a municipality may not do, however, and what Houston has done in this case, is to attempt to punish such conduct by broadly criminalizing speech directed to an officer—in this case, by authorizing the police to arrest a person who in any manner verbally interrupts an officer. Justice Powell also observes that “contentious and abusive” speech can interrupt an officer’s investigation, and offers as an example a person who “run[s] beside [an officer pursuing a felon] in a public street shouting at the officer.” Post, at 479. But what is of concern in that example is not simply contentious speech, but rather the possibility that by shouting and running beside the officer the person may physically obstruct the officer’s investigation. Although that person might constitutionally be punished under a tailored statute that prohibited individuals from physically obstructing an officer’s investigation, he or she may not be punished under a broad statute aimed at speech. 12 This conclusion finds a familiar echo in the common law. See, e. g., The King v. Cook, 11 Can. Crim. Cas. Ann. 32, 33 (B. C. County Ct. 1906) (“Cook ... a troublesome, talkative individual, who evidently regards the police with disfavour and makes no secret of his opinions on the subject . . . [told] some persons in a tone of voice undoubtedly intended for [the officer’s] ears, that the arrested man was not drunk and the arrest was unjustifiable. Now up to this point he had committed no crime, as in a free country like this citizens are entitled to express their opinions without thereby rendering themselves liable to arrest unless they are inciting others to break the law; and policemen are not exempt from criticism any more than Cabinet Ministers”); Levy v. Edwards, 1 Car. & P. 40, 171 464 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. The city argues, however, that even if the ordinance encompasses some protected speech, its sweeping nature is both inevitable and essential to maintain public order. The city recalls this Court’s observation in Smith v. Goguen, 415 U. S. 566, 581 (1974): “There are areas of human conduct where, by the nature of the problems presented, legislatures simply cannot establish standards with great precision. Control of the broad range of disorderly conduct that may inhibit a policeman in the performance of his official duties may be one such area requiring as it does an on-the-spot assessment of the need to keep order.” The city further suggests that its ordinance is comparable to the disorderly conduct statute upheld against a facial challenge in Colten v. Kentucky, 407 U. S. 104 (1972). Eng. Rep. 1094 (Nisi Prius 1823) (where constable breaks up fight between two boys and proceeds to handcuff one of them, third party who objects by telling constable “ ‘you have no right to handcuff the boy’ ” has done no wrong and may not be arrested); cf. Ruthenbeck v. First Criminal Judicial Court of Bergen Cty., 7 N. J. Mise. 969, 147 A. 625 (1929) (vacating conviction for saying to police officer “You big muttonhead, do you think you are a czar around here?”). See generally Note, Obstructing A Public Officer, 108 U. Pa. L. Rev. 388, 390-392, 406-407 (1960) (“[C]onduct involving only verbal challenge of an officer’s authority or criticism of his actions . . . operates, of course, to impair the working efficiency of government agents. ... Yet the countervailing danger that would lie in the stifling of all individual power to resist—the danger of an omnipotent, unquestionable officialdom—demands some sacrifice of efficiency ... to the forces of private opposition. . . . [T]he strongest case for allowing challenge is simply the imponderable risk of abuse—to what extent realized it would never be possible to ascertain—that lies in the state in which no challenge is allowed”). The freedom verbally to challenge police action is not without limits, of course; we have recognized that “fighting words” which “by their very utterance inflict injury or tend to incite an immediate breach of the peace” are not constitutionally protected. Chaplinsky v. New Hampshire, supra, at 572; Gooding v. Wilson, 405 U. S. 518, 522-525 (1972). See also supra, at 461-462, and n. 10. HOUSTON v. HILL 465 451 Opinion of the Court This Houston ordinance, however, is not narrowly tailored to prohibit only disorderly conduct or fighting words,13 and in no way resembles the law upheld in Colten.14 Although we appreciate the difficulties of drafting precise laws, we have repeatedly invalidated laws that provide the police with unfettered discretion to arrest individuals for words or conduct that annoy or offend them.16 As the Court observed over a 13 To the extent the ordinance did extend to disorderly conduct, it would be pre-empted by Tex. Penal Code Ann. § 42.01 (1974 and Supp. 1987), the comprehensive state disorderly conduct provision. See n. 10, supra. 14 The ordinance challenged in Colten v. Kentucky stated: “(1) A person is guilty of disorderly conduct if, with intent to cause public inconvenience, annoyance, or alarm, or recklessly creating a risk thereof, he: “(f) Congregates with other persons in a public place and refuses to comply with a lawful order of the police to disperse . . . .” Ky. Rev. Stat. §437.016(l)(f) (Supp. 1968); see 407 U. S., at 108. The Court upheld the ordinance against overbreadth challenge because the Kentucky Supreme Court had construed it so that it “infringe[d] no protected speech or conduct.” Id., at 111. 16 See, e. g., Kolenderv. Lawson, 461 U. S. 352, 360-361 (1983) (identification requirement unconstitutional because it accords police “full discretion”); Smith v. Goguen, 415 U. S. 566, 575 (1974) (“Statutory language of such a standardless sweep allows policemen, prosecutors, and juries to pursue their personal predilections . . . [thereby] entrusting lawmaking ‘to the moment-to-moment judgment of the policeman on his beat’ ”), quoting Gregory v. Chicago, 394 U. S. Ill, 120 (1969) (Black, J., concurring); Papachristou v. City of Jacksonville, 405 U. S. 156, 170 (1972) (vagrancy ordinance “furnishes a convenient tool for ‘harsh and discriminatory enforcement by local prosecuting officials, against particular groups deemed to merit their displeasure’ ”), quoting Thornhill v. Alabama, 310 U. S. 88, 97-98 (1940); Coates v. Cincinnati, 402 U. S. 611, 615-616 (1971) (statute prohibiting “annoying” conduct “contains an obvious invitation to discriminatory enforcement”). Like many of the ordinances in these cases, Houston’s effectively grants police the discretion to make arrests selectively on the basis of the content of the speech. Such discretion is particularly repugnant given “[t]he eternal temptation ... to arrest the speaker rather than to correct the conditions about which he complains.” Younger v. Harris, 401 U. S. 37, 65 (1971) (Douglas, J., dissenting). 466 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. century ago, “[i]t would certainly be dangerous if the legislature could set a net large enough to catch all possible offenders, and leave it to the courts to step inside and say who could be rightfully detained, and who should be set at large.” United States v. Reese, 92 U. S. 214, 221 (1876). In Lewis, Justice Powell elaborated the basis for our concern with such sweeping, dragnet laws: “This ordinance, as construed by the Louisiana Supreme Court, confers on police a virtually unrestrained power to arrest and charge persons with a violation. Many arrests are made in ‘one-on-one’ situations where the only witnesses are the arresting officer and the person charged. All that is required for conviction is that the court accept the testimony of the officer that obscene or opprobrious language had been used toward him while in the performance of his duties.*. . . “Contrary to the city’s argument, it is unlikely that limiting the ordinance’s application to genuine ‘fighting words’ would be incompatible with the full and adequate performance of an officer’s duties. . . . [I]t is usually unnecessary [to charge a person] with the less serious offense of addressing obscene words to the officer. The present type of ordinance tends to be invoked only where there is no other valid basis for arresting an objectionable or suspicious person. The opportunity for abuse, especially where a statute has received a virtually open-ended interpretation, is self-evident. “ *The facts in this case, and particularly the direct conflict of testimony as to ‘who said what,’ well illustrate the possibility of abuse. 415 U. S., at 135-136, and n. Houston’s ordinance criminalizes a substantial amount of constitutionally protected speech, and accords the police unconstitutional discretion in enforcement. The ordinance s plain language is admittedly violated scores of times daily, App. 77, yet only some individuals—those chosen by the po- HOUSTON v. HILL 467 451 Opinion of the Court lice in their unguided discretion—are arrested. Far from providing the “breathing space” that “First Amendment freedoms need ... to survive,” NAACP v. Button, 371 U. S. 415, 433 (1963), the ordinance is susceptible of regular application to protected expression. We conclude that the ordinance is substantially overbroad, and that the Court of Appeals did not err in holding it facially invalid. Ill The city has also urged us not to reach the merits of Hill’s constitutional challenge, but rather to abstain for reasons related to those underlying our decision in Railroad Common v. Pullman Co., 312 U. S. 496 (1941). In its view, there are certain limiting constructions readily available to the state courts that would eliminate the ordinance’s overbreadth.16 Abstention is, of course, the exception and not the rule, Colorado River Water Conservation Dist. v. United States, 424 U. S. 800, 813 (1976), and we have been particularly reluctant to abstain in cases involving facial challenges based on the First Amendent.17 We have held that “abstention ... is inappropriate for cases [where] . . . statutes are justifiably attacked on their face as abridging free expression.” Dombrowski v. Pfister, 380 U. S. 479, 489-490 (1965). “In such case[s] to force the plaintiff who has commenced a federal ac 16 The city did not raise the abstention issue until after it had lost on the merits before the panel of the Court of Appeals. After rehearing en banc, neither the majority nor the dissent addressed abstention. The city’s tardy decision to urge abstention is remarkable given its acquiescence for more than three years to federal adjudication of the merits and its insistence before the District Court and the panel that the ordinance was both unambiguous and constitutional on its face. These circumstances undercut the force of the city’s argument, but do not bar us from considering it. Wisconsin v. Constantineau, 400 U. S. 433 (1971); Railroad Comm’n v. Pullman Co., 312 U. S. 496 (1941). 17 See Dombrowski v. Pfister, 380 U. S. 479, 486-487 (1965); Procunier v. Martinez, 416 U. S. 396, 404 (1974); Baggett v. Bullitt, 377 U. S. 360, 378-379 (1964); NAACP v. Button, 371 U. S. 415, 433 (1963); but cf. Babbitt v. Farm Workers, 442 U. S. 289 (1979). 468 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. tion to suffer the delay of state-court proceedings might itself effect the impermissible chilling of the very constitutional right he seeks to protect.” Zwickler v. Koota, 389 U. S. 241, 252 (1967). Even if this case did not involve a facial challenge under the First Amendment, we would find abstention inappropriate. In cases involving a facial challenge to a statute, the pivotal question in determining whether abstention is appropriate is whether the statute is “fairly subject to an interpretation which will render unnecessary or substantially modify the federal constitutional question.” Harman v. Forssenius, 380 U. S. 528, 534-535 (1965); see also Hawaii Housing Authority v. Midkiff, 467 U. S. 229, 236 (1984) (same). If the statute is not obviously susceptible of a limiting construction, then even if the statute has “never [been] interpreted by a state tribunal ... it is the duty of the federal court to exercise its properly invoked jurisdiction.” Harman, supra, at 535; see, e. g., Wisconsin v. Constan-tineau, 400 U. S. 433, 439 (1971) (“Where there is no ambiguity in the state statute, the federal court should not abstain but should proceed to decide the federal constitutional claim”); Zwickler v. Koota, supra, at 250-251, and n. 14 (citing cases). This ordinance is not susceptible to a limiting construction because, as both courts below agreed, its language is plain and its meaning unambiguous. Its constitutionality cannot “turn upon a choice between one or several alternative meanings.” Baggett v. Bullitt, 377 U. S. 360, 378 (1964); cf. Babbitt v. Farm Workers, 442 U. S. 289, 308 (1979). Nor can the ordinance be limited by severing discrete unconstitutional subsections from the rest. For example, it cannot be limited to “core criminal conduct” such as physical assaults or fighting words because those applications are pre-empted by state law. See supra, at 460-461, and n. 10. The enforceable portion of this ordinance is a general prohibition of speech that “simply has no core” of constitutionally unprotected expression to which it might be limited. Smith v. HOUSTON v. HILL 469 451 Opinion of the Court Goguen, 415 U. S., at 578 (emphasis deleted). The city’s proposed constructions are insufficient,18 and it is doubtful that even “a remarkable job of plastic surgery upon the face of the ordinance” could save it. Shuttlesworth v. Birmingham, 394 U. S. 147, 153 (1969). In sum, “[s]ince ‘the naked question, uncomplicated by [ambiguous language], is whether the Act on its face is unconstitutional,’ Wisconsin v. Cons-tantineau, 400 U. S. 433, 439 (1971), abstention from federal jurisdiction is not required.” Hawaii Housing Authority, supra, at 237. The city relies heavily on its claim that the state courts have not had an opportunity to construe the statute. Even if true, that factor would not in itself be controlling. As stated above, when a statute is not ambiguous, there is no need to abstain even if state courts have never interpreted the statute. Harman, supra, at 534. For example, we have declined to abstain from deciding a facial challenge to a state statute when the suit was filed in federal court just four days after the statute took effect. Brockett v. Spokane Arcades, Inc., 472 U. S. 491 (1985). But in any event, the city’s claim that state courts have not had an opportunity to construe the statute is misleading. Only the state appellate courts appear to have lacked this opportunity. It is undisputed that Houston’s Municipal Courts, which have been courts of 18 The city suggests that the statute would be constitutional if construed to apply only to (1) intentional interruptions by (2) “physical, rather than verbal, acts” during (3) an officer’s attempts to make “arrests and detentions.” Brief for Appellant 30-31. These proposals are either at odds with the ordinance’s plain meaning, or do not sufficiently limit its scope. First, speech does not necessarily lose its constitutional protection because the speaker intends it to interrupt an officer, nor would an intent requirement cabin the excessive discretion the ordinance provides to officers. Second, given the pre-emption of the first part of the statute, discussed Wra, limiting the ordinance to “physical acts” would be equivalent to invalidating it on its face. Third, there is no reasonable way to read the Plain language of the ordinance as limited to arrests and detentions; even if there were, such a limitation would not significantly limit its scope. 470 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. record in Texas since 1976, have had numerous opportunities to narrow the scope of the ordinance.19 There is no evidence that they have done so.20 In fact, the city’s primary position throughout this litigation has been “to insis[t] on the validity of the ordinance as literally read.” 789 F. 2d, at 1107. We have long recognized that trial court interpretations, such as those given in jury instructions, constitute “a ruling on a question of state law that is as binding on us as though the precise words had been written into the ordinance.” Ter-miniello, 337 U. S., at 4. Thus, where municipal courts have regularly applied an unambiguous statute, there is certainly no need for a federal court to abstain until state appellate courts have an opportunity to construe it. The possibility of certification does not change our analysis.21 The certification procedure is useful in reducing the substantial burdens of cost and delay that abstention places on litigants. Where there is an uncertain question of state law that would affect the resolution of the federal claim, and where delay and expense are the chief drawbacks to abstention, the availability of certification becomes an important factor in deciding whether to abstain. E. g., Bellotti v. Baird, 428 U. S. 132 (1976). Nevertheless, even where we have recognized the importance of certification in deciding whether to abstain, we have been careful to note that the 19 The ordinance has been in force, in substantially the same language, for over 30 years. 789 F. 2d, at 1111. The Houston police arrest on average 1,000 persons per year under the ordinance. Brief for Appellee 14,35 (citing Record). 20 Indeed, Hill introduced evidence in the District Court that Houston’s Municipal Courts have declined to employ limiting constructions in jury instructions. Brief for Appellee 35 (citing Record 104-105, plaintiff s Exhibits 3, 4, 5). 21 Under Texas law, either this Court or a United States court of appeals may certify a question of Texas criminal law “which may be determinative of the cause then pending and as to which it appears to the certifying court that there is no controlling precedent in the decisions of the Court of Criminal Appeals.” Tex. Rule App. Proc. 214. HOUSTON v. HILL 471 451 Opinion of the Court availability of certification is not in itself sufficient to render abstention appropriate. Id., at 151. It would be manifestly inappropriate to certify a question in a case where, as here, there is no uncertain question of state law whose resolution might affect the pending federal claim. As we have demonstrated, supra, at 468-469, this ordinance is neither ambiguous nor obviously susceptible of a limiting construction.22 A federal court may not properly ask a state court if it would care in effect to rewrite a statute.23 We therefore see no need in this case to abstain pending certification. IV Today’s decision reflects the constitutional requirement that, in the face of verbal challenges to police action, officers and municipalities must respond with restraint. We are “Justice Powell argues that the unsettled question of the effect on this ordinance of § 6.02(b) of the Texas Penal Code, which requires “a culpable mental state” as an element of any offense, creates sufficient ambiguity to require certification. He suggests that the Texas Court of Criminal Appeals might limit convictions under the ordinance to cases in which there was a finding of “inten[t] to interfere with the officer’s performance of his duties” justifies certification, and argues that such a limit would “narrow the focus of the constitutional question” before us. Post, at 474. As Justice Powell implicitly concedes, however, there is no possibility that such an intent requirement would eliminate the excessive discretion the ordinance affords to the police in choosing whom to arrest; even with such a requirement, the ordinance would remain unconstitutionally overbroad. Moreover, the meaning and application of such an intent requirement is not self-evident, and could raise independent questions of vagueness or of overbreadth. This is therefore a case where certification “would not only hold little hope of eliminating the issue of [overbreadth] but also would very likely pose other constitutional issues for decision, a result not serving the abstention- [or certification-] justifying end of avoiding constitutional adjudication.” Baggett v. Bullitt, 377 U. S., at 378. It would also be inappropriate for a federal court to certify the entire constitutional challenge to the state court, of course, for certified questions should be confined to uncertain questions of state law. See 17 C. Wright, A. Miller, & E. Cooper, Federal Practice and Procedure § 4248, pp. 529-530 (1978). 472 OCTOBER TERM, 1986 Scalia, J., concurring in judgment 482 U. S. mindful that the preservation of liberty depends in part upon the maintenance of social order. Cf. Terminiello n. Chicago, supra, at 37 (dissenting opinion). But the First Amendment recognizes, wisely we think, that a certain amount of expressive disorder not only is inevitable in a society committed to individual freedom, but must itself be protected if that freedom would survive. We therefore affirm the judgment of the Court of Appeals. It is so ordered. Justice Blackmun, concurring. I join the Court’s opinion and its judgment except that I do not agree with any implication—if one exists—see ante, at 461-462, that Gooding v. Wilson, 405 U. S. 518 (1972), and Lewis v. City of New Orleans, 415 U. S. 130 (1974), are good law in the context of their facts, or that they lend any real support to the judgment under review in this case. I dissented in Gooding and Lewis, see 405 U. S., at 534, and 415 U. S., at 136, in the conviction that the legislation there under consideration was related to “fighting words,” within the teaching and reach of Chaplinsky n. New Hampshire, 315 U. S. 568 (1942). I am still of that view, and I therefore disassociate myself from any possible suggestion that those cases are controlling authority here. The Houston ordinance before us, however, as is evident from its very language, and as the Court demonstrates, ante, at 462-463, 465, is far more broad and more offensive to First Amendment values and is susceptible of regular application to protected expression. Justice Scalia, concurring in the judgment. For the reasons stated by Justice Powell in Part II of his opinion, I agree that abstention would not be appropriate in this case. Because I do not believe that the Houston ordinance is reasonably susceptible of a limiting construction that would avoid the constitutional question posed in this case, I agree with the Court that certification would also be inappropriate. On the merits, I agree with the views expressed by HOUSTON v. HILL 473 451 Opinion of Powell, J. Justice Powell in Part III of his opinion. I therefore concur in the judgment and joins Parts II and III of Justice Powell’s opinion. Justice Powell, with whom Justice O’Connor joins, and with whom The Chief Justice joins as to Parts I and II, and Justice Scalia joins as to Parts II and III, concurring in the judgment in part and dissenting in part. The city of Houston has made it unlawful “for any person to ... in any manner oppose, molest, abuse or interrupt any policeman in the execution of his duty.” Code of Ordinances, City of Houston, Texas § 34-11(a) (1984). The Court today concludes that this ordinance violates the First and Fourteenth Amendments of the Constitution. In my view, the Court should not have reached the merits of the constitutional claims, but instead should have certified a question to the Texas Court of Criminal Appeals. I also disagree with the Court’s reasons for declining to abstain under the principle of Railroad Comm’n v. Pullman Co., 312 U. S. 496 (1941). Finally, although I agree that the ordinance as interpreted by the Court violates the Fourteenth Amendment, I write separately because I cannot join the Court’s reasoning. I This case involves a challenge to an ordinance designed to prevent interference with police officers in the performance of their duties. Constitutional analysis should not proceed until we determine the precise meaning of the ordinance in question. But this problem does not detain the Court, because it concludes that interpretation of the ordinance presents “no uncertain question of state law.” Ante, at 471. On the contrary, I think there is a serious question as to the meaning of the ordinance. The challenged ordinance does not contain an explicit intent requirement. Both parties acknowledge, however, that the Texas Penal Code requires imputation of some culpability requirement. See Brief for Appellant 28-30; Brief for Ap 474 OCTOBER TERM, 1986 Opinion of Powell, J. 482 U. S. pellee 31. Texas Penal Code Ann. § 6.02(b) (1974) provides: “If the definition of an offense does not prescribe a culpable mental state, a culpable mental state is nevertheless required unless the definition plainly dispenses with any mental element.”1 The nature of this imputed mental state has a direct effect on the constitutional issue presented by this case. The Court apparently assumes that the requisite intent can be provided by a person’s intent to utter words that constitute an interruption. But it would be plausible for the Texas Court of Criminal Appeals to construe the intent requirement differently. For example, that court could conclude that conviction under the ordinance requires proof that the person not only intended to speak, but also intended to interfere with the officer’s performance of his duties. This interpretation would change the constitutional questions in two ways: it would narrow substantially the scope of the ordinance, and possibly resolve the overbreadth question; it also would make the language of the ordinance more precise, and possibly satisfy the concern as to vagueness. At the least, such an interpretation would narrow the focus of the constitutional question and obviate the need for the Court’s broad statements regarding First Amendment protections of speech directed at police officers. It is not this Court’s role, however, to place an interpretive gloss on the words the Houston City Council has chosen. The ordinance is not a federal law, and we do not have the power “ ‘authoritatively to construe’ ” it. Gooding v. Wilson, 405 U. S. 518, 520 (1972) (quoting United States v. Thirty-seven Photographs, 402 U. S. 363, 369 (1971)). But we are not without means of obtaining an authoritative construction. Last year the Texas voters amended the Texas Constitution to provide that the “court of criminal ap- :At least one Texas appellate court has concluded that this section applies to municipal ordinances. See Pollard n. State, 687 S. W. 2d 373, 374 (Tex. App. 1985) (pet. ref’d, Pollard n. State, No. 05-83-01161 Cr. (Jan. 29, 1986)). HOUSTON v. HILL 475 451 Opinion of Powell, J. peals [has] jurisdiction to answer questions of state law certified from a federal appellate court.” Tex. Const., Art. 5, §3-c. See Tex. Rule App. Proc. 214 (implementing this aspect of the constitutional provision). As Justice O’Connor explained recently, “[s]peculation by a federal court about the meaning of a state statute in the absence of prior state court adjudication is particularly gratuitous when . . . the state courts stand willing to address questions of state law on certification from a federal court.” Brockett v. Spokane Arcades, Inc., 472 U. S. 491, 510 (1985) (concurring). The Court repeatedly has emphasized the appropriateness of certification in cases presenting uncertain questions of state law. In such cases, certification can “‘save time, energy, and resources and hel[p] build a cooperative judicial federalism.’” Bellotti v. Baird, 428 U. S. 132, 150-151 (1976) (quoting Lehman Brothers n. Schein, 416 U. S. 386, 391 (1974)).2 In my view, the ambiguity of the ordinance, coupled with the seriousness of invalidating a state law, requires that we ascertain what the ordinance means before we address appellee’s constitutional claims. I therefore would vacate the judgment below and remand with instructions to certify the case to the Texas Court of Criminal Appeals to allow it to interpret the intent requirement of this ordinance. Accordingly, I dissent. The Court concludes, however, that the case properly is before us, and so I address the remaining issues presented.3 2 This case demonstrates two advantages of certification over the more traditional Pullman abstention procedure. First, certification saves time by sending the question directly to the court that is empowered to provide an authoritative construction of the statute. Second, certification obviates the procedural difficulties that may hinder efforts to obtain declaratory judgments from state trial courts. See infra, at 476-477. 3Cf. Brown Shoe Co. v. United States, 370 U. S. 294, 374 (1962) (Harlan, J., dissenting in part and concurring in part); Gillespie v. United ^ates Steel Corp., 379 U. S. 148, 170 (1964) (Stewart, J., concurring). See also Longshoremen v. Davis, 476 U. S. 380, 403 (1986) (Rehnquist, J*, concurring in part and concurring in judgment). 476 OCTOBER TERM, 1986 Opinion of Powell, J. 482 U. S. II Pullman abstention generally is appropriate when determination of an unsettled question of state law by a state court could avoid the need for decision of a substantial question of federal constitutional law. Although I agree with the Court that Pullman abstention is inappropriate in this case, I write separately because my reasons are somewhat different from those expressed by the Court.4 Pullman abstention is inappropriate unless the state courts “provid[e] the parties with adequate means to adjudicate the controverted state law issue.” Field, Abstention in Constitutional Cases: The Scope of the Pullman Abstention Doctrine, 122 U. Pa. L. Rev. 1071, 1144 (1974). See 17 C. Wright, A. Miller, & E. Cooper, Federal Practice and Procedure §4242, p. 468 (1978). Cf. Railroad Comm’n v. Pullman Co., 312 U. S., at 501 (abstaining because the “law of Texas appears to furnish easy and ample means for determin 4 The Court concludes that Pullman abstention is inappropriate for two reasons. First, it suggests that this Court should be “particularly reluctant to abstain in cases involving facial challenges based on the First Amendment.” Ante, at 467. The Court supports this conclusion with a citation to Dombrowski v. Pfister, 380 U. S. 479 (1965). I see nothing in that case that supports such a broad principle. The Dombrowski Court declined to abstain because “the interpretation ultimately put on the [challenged state] statutes by the state courts is irrelevant,” id., at 490, and because “no readily apparent construction suggest[ed] itself as a vehicle” for curing the constitutional problem with the statute, id., at 491. Both of these rationales are straightforward applications of the general rule that Pullman abstention is appropriate only when determination of an uncertain question of state law would obviate the need for the federal court to decide a substantial question of federal constitutional law. The Court’s second reason for not abstaining is that it believes the statute is not “ ‘fairly subject to an interpretation which will. . . substantially modify the federal constitutional question.’” Ante, at 468 (quoting Harman v. Forssenius, 380 U. S. 528, 534-535 (1965)). See supra, at 473-474, for my disagreement with this view. HOUSTON v. HILL 477 451 Opinion of Powell, J. ing the Commission’s authority”).5 It is not clear that Texas law affords a remedy by which Hill could obtain a state court interpretation of the ordinance. The only apparent means of securing such a ruling would be through an action for a declaratory judgment. See Tex. Civ. Prac. & Rem. Code Ann. §37.001 et seq. (1986) (authorizing courts to grant declaratory judgments). But Texas law treats declaratory judgment actions as civil cases. Thus, they are appealable to the Texas Supreme Court rather than the Texas Court of Criminal Appeals. See, e. g., United Services Life Ins. Co. v. Delaney, 396 S. W. 2d 855 (Tex. 1965). Moreover, because the Texas Court of Criminal Appeals has exclusive appellate jurisdiction to decide questions of Texas criminal law, see Tex. Const., Art. V, §5, the Texas Supreme Court has held, with narrow exceptions, that injunctive or declaratory relief against criminal statutes is not available in civil cases. See Texas Liquor Control Board v. Canyon Creek Land Corp., 456 S. W. 2d 891, 894-896 (Tex. 1970). Thus, it is quite unlikely that a declaratory or injunctive action would bring Hill any determination of the meaning of the ordinance—either from a trial or an appellate court. In short, the only sure ways for the ordinance to be interpreted are by certification, see supra, at 473-475, and by appeals of criminal convictions under the ordinance. Neither of these routes provides Hill a means to obtain relief sufficient to justify Pullman abstention. Aside from the barriers created by Texas procedure, the late stage at which the city of Houston raised this issue weighs heavily against abstention. Houston first suggested that abstention was appropriate after the Court of Appeals published its panel opinion invalidating the ordinance. As I note that the adequacy of state procedures is examined much more strictly in cases seeking Pullman abstention than in cases seeking Younger abstention. Compare, e. g., Pennzoil Co. v. Texaco Inc., 481 U. S. 1, 14-17 (1987). 478 OCTOBER TERM, 1986 Opinion of Powell, J. 482 U. S. we have noted in a similar case, “[t]his proposal comes nearly three years after the filing of the complaint and would produce delay attributable to abstention that the Court in recent years has sought to minimize.” See Mayor of Philadelphia v. Education Equality League, 415 U. S. 605, 628 (1974). In sum, the late presentation of this claim, coupled with the doubts as to whether relief could be secured under Texas law, convinces me that Pullman abstention is inappropriate here. Ill I agree with the Court’s conclusion that the ordinance violates the Fourteenth Amendment, but do not join the Court’s reasoning. A The Court finds that the ordinance “deals not with core criminal conduct, but with speech.” Ante, at 460. This view of the ordinance draws a distinction where none exists. The terms of the ordinance—“oppose, molest, abuse or interrupt any policeman in the execution of his duty”—include general words that can apply as fully to conduct as to speech. It is in this respect that Lewis n. City of New Orleans, 415 U. S. 130 (1974), is clearly distinguishable. In that case the New Orleans ordinance made it a breach of the peace for: “‘any person wantonly to curse or revile or to use obscene or opprobrious language toward or with reference to any member of the city police while in the actual performance of his duty.’” Id., at 132 (quoting New Orleans Ordinance 828 M. C. S. §49-7). On its face, the New Orleans ordinance criminalizes only the use of language. Justice Brennan, speaking for the Court in Lewis, explicitly noted this, stating that the ordinance “punishe[d] only spoken words. ” Id., at 134. By contrast, the ordinance presented in this case could be applied to activity that involves no element of speech or communication. For example, the ordinance evidently would punish individ HOUSTON v. HILL 479 451 Opinion of Powell, J. uals who—without saying a single word—obstructed an officer’s access to the scene of an ongoing public disturbance, or indeed the scene of a crime. Accordingly, I cannot agree with the Court that this ordinance punishes only speech. I do agree that the ordinance can be applied to speech in some cases. And I also agree that the First Amendment protects a good deal of speech that may be directed at police officers. On occasion this may include verbal criticism, but I question the implication of the Court’s opinion that the First Amendment generally protects verbal “challenge[s] directed at police officers,” ante, at 461. A “challenge” often takes the form of opposition or interruption of performance of duty.6 In many situations, speech of this type directed at police officers will be functionally indistinguishable from conduct that the First Amendment clearly does not protect. For example, I have no doubt that a municipality constitutionally may punish an individual who chooses to stand near a police officer and persistently attempt to engage the officer in conversation while the officer is directing traffic at a busy intersection. Similarly, an individual, by contentious and abusive speech, could interrupt an officer’s investigation of possible criminal conduct. A person observing an officer pursuing a person suspected of a felony could run beside him in a public street shouting at the officer. Similar tactics could interrupt a policeman lawfully attempting to interrogate persons believed to be witnesses to a crime. 6 The first definition of “challenge” in the 1980 edition of the American Heritage Dictionary is “[a] call to engage in a contest or fight.” The Court implies that municipalities can punish an attempt to interfere with police officers only if it “physically obstructfs] the officer’s investigation,” ante, at 463, n. 11, or if it constitutes “fighting words” within the meaning of Chaplinsky v. New Hampshire, 315 U. S. 568 (1942), see ante, at 464, n. 12. This implication troubles me because, as I have indicated in the text supra this page, there can be many situations where a State—in the public interest—should have the right to punish speech directed at police officers that does not fall within either of these exceptions. 480 OCTOBER TERM, 1986 Opinion of Powell, J. 482 U. S. In sum, the Court’s opinion appears to reflect a failure to apprehend that this ordinance—however it may be construed—is intended primarily to further the public’s interest in law enforcement. To be sure, there is a fine line between legitimate criticism of police and the type of criticism that interferes with the very purpose of having police officers. But the Court unfortunately seems to ignore this fine line and to extend First Amendment protection to any type of verbal molestation or interruption of an officer in the performance of his duty. B Despite the concerns expressed above, I nevertheless agree that the ambiguous terms of this ordinance “confe[r] on police a virtually unrestrained power to arrest and charge persons with a violation. ... The opportunity for abuse, especially where a statute has received a virtually open-ended interpretation, is self-evident.” Lewis v. City of New Orleans, supra, at 135-136 (Powell, J., concurring in result). No Texas court has placed a limiting construction on the ordinance. Also, it is clear that Houston has made no effort to curtail the wide discretion of police officers under the present ordinance. The record contains a sampling of complaints filed under the ordinance in 1981 and 1982. People have been charged with such crimes as “Failure to remain silent and stationary,” “Remaining,” “Refusing to remain silent,” and “Talking.” 789 F. 2d 1103, 1113-1114 (CA5 1986) (en banc). Although some of these incidents may have involved unprotected conduct, the vagueness of these charges suggests that, with respect to this ordinance, Houston officials have not been acting with proper sensitivity to the constitutional rights of their citizens. When government protects society’s interests in a manner that restricts some speech the law must be framed more precisely than the ordinance before us. Accordingly, I agree with the Court that the Houston ordinance is unconstitutional. HOUSTON v. HILL 481 451 Rehnquist, C. J., dissenting It is difficult, of course, specifically to frame an ordinance that applies in “areas of human conduct where, by the nature of the problems presented, legislatures simply cannot establish standards with great precision. Control of the broad range of disorderly conduct that may inhibit a policeman in the performance of his official duties may be one such area, requiring as it does an on-the-spot assessment of the need to keep order.” Smith v. Goguen, 415 U. S. 566, 581 (1974). In view of the difficulty of drafting precise language that never restrains speech and yet serves the public interest, the attempts of States and municipalities to draft laws of this type should be accorded some leeway. I am convinced, however, that the Houston ordinance is too vague to comport with the First and Fourteenth Amendments. As I explained supra, at 473-474, it should be possible for the present ordinance to be reframed in a way that would limit the present broad discretion of officers and at the same time protect substantially the city’s legitimate interests. For example, the ordinance could make clear that it applies to speech only if the purpose of the speech were to interfere with the performance by a police officer of his lawful duties. In this situation, the difficulties of drafting precisely should not justify upholding this ordinance. IV Although I believe that the proper course is for the Court to vacate the judgment of the Court of Appeals, I “bo[w] to the Court’s decision that the case is properly before us,” Brown Shoe Co. v. United States, 370 U. S. 294, 374 (1962) (Harlan, J., concurring in part and dissenting in part), and concur in the judgment of affirmance. Chief Justice Rehnquist, dissenting. I join Parts I and II of Justice Powell’s opinion concur-rmg in the judgment in part and dissenting in part. I do not 482 OCTOBER TERM, 1986 Rehnquist, C. J., dissenting 482 U. S. agree, however, that the Houston ordinance, in the absence of an authoritative construction by the Texas courts, is unconstitutional. See Lewis v. City of New Orleans, 415 U. S. 130, 136 (1974) (Blackmun, J., dissenting). I therefore dissent from the Court’s affirmance of the judgment of the Court of Appeals. PERRY v. THOMAS 483 Syllabus PERRY et al. v. THOMAS APPEAL FROM THE COURT OF APPEAL OF CALIFORNIA, SECOND APPELLATE DISTRICT No. 86-566. Argued April 28, 1987—Decided June 15, 1987 Appellee brought suit in California Superior Court against his former employer and appellants, two of its employees, alleging breach of contract and related causes of action arising from a dispute over commissions on securities sales. After appellee refused to arbitrate, appellants filed a petition to compel arbitration under §§ 2 and 4 of the Federal Arbitration Act, which respectively provide that contractual arbitration provisions are valid and enforceable and mandate their judicial enforcement. The demand for arbitration was based on a provision in a form appellee executed in connection with his employment application, whereby he agreed to arbitrate any dispute with his employer. Appellee opposed arbitration on the ground that his suit was authorized by California Labor Code §229, which provides that wage collection actions may be maintained without regard to the existence of any private agreement to arbitrate. The court refused to compel arbitration, characterizing as “controlling authority” Merrill Lynch, Pierce, Fenner & Smith, Inc. v. Ware, 414 U. S. 117, which upheld § 229 in the face of a Supremacy Clause preemption challenge premised on an arbitration requirement in a New York Stock Exchange rule, which was promulgated pursuant to § 6 of the Securities Exchange Act of 1934 (1934 Act). The State Court of Appeals affirmed. Both lower courts refused to consider appellee’s argument that appellants lacked “standing” to enforce the arbitration agreement since they were not parties to it. Held: 1. Under the Supremacy Clause, §2 of the Federal Arbitration Act pre-empts § 229 of the California Labor Code. In enacting § 2, Congress declared a national policy favoring arbitration and withdrew the States’ power to require a judicial forum for the resolution of claims that contracting parties agreed to resolve by arbitration. Ware is distinguishable on the ground that the language and policies of the 1934 Act and the regulations promulgated thereunder evidenced no clear federal intent to require arbitration. The oblique reference to the Federal Arbitration Act in footnote 15 of Ware cannot fairly be read as a definitive holding that that Act does not pre-empt § 229, since the footnote was concerned wrth federally created rights and did not address the issue of federal preemption of state-created rights. Pp. 489-491. 484 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. 2. Appellee’s contention that resolving in appellants’ favor the question of their “standing” to enforce the agreement to arbitrate is a prerequisite under Article III of the Constitution to their maintenance of this appeal is rejected. Appellee’s “standing” argument—which this Court does not reach because the lower courts did not address it—simply presents the straightforward contract interpretation issue whether the arbitration provision inures to appellants’ benefit and may be construed to cover the present dispute. That issue may be resolved on remand, and its status as an alternative ground for denying arbitration does not prevent this Court from reviewing the lower courts’ holdings on the preemption question. P. 492. Reversed and remanded. Marshall, J., delivered the opinion of the Court, in which Rehnquist, C. J., and Brennan, White, Blackmun, Powell, and Scalia, JJ., joined. Stevens, J., post, p. 493, and O’Connor, J., post, p. 494, filed dissenting opinions. Peter Brown Dolan argued the cause for appellants. With him on the briefs was Maren E. Nelson. Bruce Gelber argued the cause and filed a brief for appellee. Justice Marshall delivered the opinion of the Court. In this appeal we decide whether §2 of the Federal Arbitration Act, 9 U. S. C. §1 et seq., which mandates enforcement of arbitration agreements, pre-empts § 229 of the California Labor Code, which provides that actions for the collection of wages may be maintained “without regard to the existence of any private agreement to arbitrate.” Cal. Lab. Code Ann. § 229 (West 1971). I Appellee, Kenneth Morgan Thomas, brought this action in California Superior Court against his former employer, Kidder, Peabody & Co. (Kidder, Peabody), and two of its employees, appellants Barclay Perry and James Johnston. His complaint arose from a dispute over commissions on the sale of securities. Thomas alleged breach of contract, conversion, civil conspiracy to commit conversion, and breach of PERRY v. THOMAS 485 483 Opinion of the Court fiduciary duty, for which he sought compensatory and punitive damages. After Thomas refused to submit the dispute to arbitration, the defendants sought to stay further proceedings in the Superior Court. Perry and Johnston filed a petition in the Superior Court to compel arbitration; Kidder, Peabody invoked diversity jurisdiction and filed a similar petition in Federal District Court. Both petitions sought arbitration under the authority of §§2 and 4 of the Federal Arbitration Act.1 The demands for arbitration were based on a provision found in a Uniform Application for Securities Industry Registration form, which Thomas completed and executed in connection with his application for employment with Kidder, Peabody. That provision states: “I agree to arbitrate any dispute, claim or controversy that may arise between me and my firm, or a customer, or any other person, that is required to be arbitrated under the rules, constitutions or by-laws of the organizations with which I register . . . .” App. 33a. Rule 347 of the New York Stock Exchange, Inc. (1975), with which Thomas registered, provides that “[a]ny controversy between a registered representative and any member or member organization arising out of the employment or termination of employment of such registered representative by and with such member or member organization shall be settled by arbitration, at the instance of any such party . . . .” App. 34a. 1 Section 2 provides, in relevant part: “A written provision in ... a contract evidencing a transaction involving commerce to settle by arbitration a controversy thereafter arising out of such contract or transaction, or the refusal to perform the whole or any part thereof, . . . shall be valid, irrevocable and enforceable, save upon such grounds as exist at law or in equity for the revocation of any contract.” 9 U. S. C. §2. Section 4 mandates judicial enforcement of arbitration agreements where a party has failed, neglected, or refused to arbitrate. 9 U. S. C. § 4. 486 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. Kidder, Peabody sought arbitration as a member organization of the New York Stock Exchange (NYSE). Perry and Johnston relied on Thomas’ allegation that they had acted in the course and scope of their employment and argued that, as agents and employees of Kidder, Peabody, they were beneficiaries of the arbitration agreement. Thomas opposed both petitions on the ground that § 229 of the California Labor Code authorized him to maintain an action for wages, defined to include commissions,2 despite the existence of an agreement to arbitrate. He relied principally on this Court’s decision in Merrill Lynch, Pierce, Fenner & Smith, Inc. v. Ware, 414 U. S. 117 (1973), which had also considered the validity of § 229 in the face of a pre-emption challenge under the Supremacy Clause, U. S. Const., Art. VI, cl. 2. Thomas maintained that the decision in Ware stood for the proposition that the State’s interest in protecting wage earners outweighs the federal interest in uniform dispute resolution. The Superior Court denied appellants’ petition to compel arbitration.3 Thomas n. Kidder Peabody & Co., Civ. Action No. C529105 (Los Angeles County, Apr. 23, 1985) (reprinted at App. 128a-129a). The court characterized Ware as “controlling authority” which held that, “in accordance with California Labor Code Section 229, actions to collect wages may be pursued without regard to private arbitration agreements.” Id., at 129a. It further concluded that since Thomas’ claims for conversion, civil conspiracy, and breach of fiduciary duty were ancillary to his claim for breach of 2 Section 200(a) of the California Labor Code defines “wages” to include amounts earned on a “commission basis.” Cal. Lab. Code Ann. § 200(a) (West 1971). The California Superior Court and the California Court of Appeal held below that the commissions at issue in this case fall within the statutory definition. App. 128a, 140a. 3 The Federal District Court gave this ruling preclusive effect and entered a final order dismissing Kidder, Peabody’s petition in the parallel proceeding. Kidder, Peabody & Co. v. Thomas, Civ. Action No. 85-1257RJK (CD Cal., Sept. 29, 1986) (reprinted at App. 245a); id., at 235a. PERRY v. THOMAS 487 483 Opinion of the Court contract and differed only in terms of the remedies sought, they should also be tried and not severed for arbitration. Id., at 128a-129a. The Superior Court did not address Thomas’ contention that Perry and Johnston were “not parties” to the arbitration agreement, id., at 78a, and therefore lacked a contractual basis for asserting the right to arbitrate, an argument Thomas characterizes as one of “standing.”4 Before the California Court of Appeal, appellants argued that Ware resolved only the narrow issue whether § 229 was pre-empted by Rule 347’s provision for arbitration, given the promulgation of that Rule by the NYSE pursuant to § 6 of the Securities Exchange Act of 1934 (1934 Act), 48 Stat. 885, as amended, 15 U. S. C. § 78f, and the authority of the Securities and Exchange Commission (SEC) to review and modify the NYSE Rules pursuant to § 19 of the 1934 Act, 15 U. S. C. §78s.5 See 414 U. S., at 135. It was appellants’ contention that, despite an indirect reference to the Federal Arbitration 4 Having concluded that this Court’s decision in Merrill Lynch, Pierce, Fenner & Smith, Inc. v. Ware, 414 U. S. 117 (1973), was dispositive, the California Superior Court also did not address Thomas’ alternative argument that the arbitration agreement in this case constitutes an unconscionable, unenforceable contract of adhesion because “(a) the selection of arbitrators is made by the New York Stock Exchange and is presumptively biased in favor of management; and (b) the denial of meaningful. . . discovery is unduly oppressive and frustrates an employee’s claim for relief.” App. 74a. 5 The Court of Appeal rejected appellants’ contention that amendments to the 1934 Act since this Court’s decision in Ware removed the theoretical underpinnings of that decision by expanding the scope of the SEC’s authority under § 19 to review and modify NYSE rules. See 15 U. S. C. § 78s(c). Appellants continue to make this argument, in their appeal before this Court, as an alternative basis for distinguishing Ware. Brief for Appellants 17-20 (citing S. Rep. No. 94-75, pp. 22-38 (1975); Dray er v. Krasner, 572 F. 2d 348, 356-359 (CA2), cert, denied, 436 U. S. 948 (1978)). However, because we rest our decision exclusively on the Federal Arbitration Act, we decline to consider the pre-emptive effect of the amended 1934 Act as it relates to Thomas’ agreement to be bound by NYSE Rule 347. 488 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. Act in footnote 15 of the Ware opinion, the pre-emptive effect of § 2 of the Act was not at issue in that case. In an unpublished opinion, the Court of Appeal affirmed. Thomas v. Perry, 2d Civ. No. B014485 (2d Dist., Div. 5, Apr. 10, 1986) (reprinted at App. 139a-142a). It read Ware’s single reference to the Federal Arbitration Act to imply that the Court had refused to hold § 229 pre-empted by that Act and the litigants’ agreement to arbitrate disputes pursuant to Rule 347. Thus, the Court of Appeal held that a claim for unpaid wages brought under § 229 was not subject to compulsory arbitration, notwithstanding the existence of an arbitration agreement. App. 140a-141a. Like the Superior Court, the Court of Appeal also rejected appellants’ argument, based on this Court’s decision in Dean Witter Reynolds Inc. v. Byrd, 470 U. S. 213 (1985), that the ancillary claims for conversion, civil conspiracy, and breach of fiduciary duty were severable from the breach-of-contract claim and should be arbitrated. App. 142a. Finally, the Court of Appeal refused to consider Thomas’ argument that Perry and Johnston lacked “standing” to enforce the arbitration agreement. The court concluded that Thomas had raised this argument for the first time on appeal.6 Id., at 140a, n. 1. 6 Objecting to appellants’ request for a formal Statement of Decision from the Superior Court following summary denial of their motion to compel, Thomas argued that appellants had “no standing” to seek an order compelling arbitration. App. 120a. Perry and Johnston replied that their “standing” to seek arbitration inhered in their status as agents and employees of Kidder, Peabody, and as beneficiaries of the agreement between Kidder, Peabody and Thomas. Id., at 124a. In response, Thomas simply argued that Perry and Johnston had submitted no supporting evidence to show they had acted as agents for Kidder, Peabody. Id., at 132a. The Superior Court did not amend a Proposed Statement of Decision, see id., at 128a-129a, to address these arguments, and it was formally adopted as the Statement of Decision from which Perry and Johnston appealed. Id., at 135a. Having based its decision “squarely on Ware,” the Court of Appeal also declined to reach Thomas’ alternative ground for supporting the Superior Court’s decision not to compel arbitration: his contention that the arbi- PERRY v. THOMAS 489 483 Opinion of the Court The California Supreme Court denied appellants’ petition for review. Id., at 144a. We noted probable jurisdiction,7 479 U. S. 982 (1986), and now reverse. II “Section 2 is a congressional declaration of a liberal federal policy favoring arbitration agreements, notwithstanding any state substantive or procedural policies to the contrary. The effect of the section is to create a body of federal substantive law of arbitrability, applicable to any arbitration agreement within the coverage of the Act.” Moses H. Cone Memorial Hospital v. Mercury Construction Corp., 460 U. S. 1, 24 (1983). Enacted pursuant to the Commerce Clause, U. S. Const., Art. I, §8, cl. 3, this body of substantive law is enforceable in both state and federal courts. Southland Corp. v. Keating, 465 U. S. 1, 11-12 (1984) (§2 held to pre-empt a provision of the California Franchise Investment Law that California courts had interpreted to require judicial consideration of claims arising under that law). As we stated in Keating, “[i]n enacting §2 of the federal Act, Congress declared a national policy favoring arbitration and withdrew the power of the states to require a judicial forum for the resolution of claims which the contracting parties agreed to resolve by arbitration.” Id., at 10. “Congress intended to foreclose state legislative attempts to undercut the enforceability of arbitration agreements.” Id., at 16 (footnote omitted). Section 2, therefore, embodies a clear federal policy of requiring arbitration unless the agreement to arbitrate is not part of a contract evidencing interstate commerce or is revocable “upon such grounds as exist at law or in equity for the revocation of any contract.” 9 U. S. C. §2. “We see nothing in the Act indicating that the broad principle of en-tration provision constitutes an unconscionable, unenforceable contract of adhesion. Id., at 141a, n. 3; see n. 4, supra. 7 Jurisdiction over this appeal is provided by 28 U. S. C. § 1257(2). See Southland Corp. n. Keating, 465 U. S. 1, 6-8 (1984). 490 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. forceability is subject to any additional limitations under state law.” Keating, supra, at 11. In Ware, which also involved a dispute between a securities broker and his former employer, we rejected a Supremacy Clause challenge to § 229 premised in part on the contention that, because the 1934 Act had empowered the NYSE to promulgate rules and had given the SEC authority to review and modify these rules, a private agreement to be bound by the arbitration provisions of NYSE Rule 347 was enforceable as a matter of federal substantive law, and pre-empted state laws requiring resolution of the dispute in court. But the federal substantive law invoked in Ware emanated from a specific federal regulatory statute governing the securities industry—the 1934 Act. We examined the language and policies of the 1934 Act and found “no Commission rule or regulation that specifie[d] arbitration as the favored means of resolving employer-employee disputes,” 414 U. S., at 135, or that revealed a necessity for “nationwide uniformity of an exchange’s housekeeping affairs.” Id., at 136. The fact that NYSE Rule 347 was outside the scope of the SEC’s authority of review militated against finding a clear federal intent to require arbitration. Id., at 135-136. Absent such a finding, we could not conclude that enforcement of California’s §229 would interfere with the federal regulatory scheme. Id., at 139-140. By contrast, the present appeal addresses the pre-emptive effect of the Federal Arbitration Act, a statute that embodies Congress’ intent to provide for the enforcement of arbitration agreements within the full reach of the Commerce Clause. Its general applicability reflects that “[t]he preeminent concern of Congress in passing the Act was to enforce private agreements into which parties had entered . . . .” Byrd, 470 U. S., at 221. We have accordingly held that these agreements must be “rigorously enforce[d].” Ibid.; see Shear-son/American Express Inc. v. McMahon, ante, at 226; Mitsubishi Motors Corp. v. Soler Chrysler-Plymouth, Inc., 473 PERRY v. THOMAS 491 483 Opinion of the Court U. S. 614, 625-626 (1985). This clear federal policy places § 2 of the Act in unmistakable conflict with California’s § 229 requirement that litigants be provided a judicial forum for resolving wage disputes. Therefore, under the Supremacy Clause, the state statute must give way. The oblique reference to the Federal Arbitration Act in footnote 15 of the Ware decision, 414 U. S., at 135, cannot fairly be read as a definitive holding to the contrary. There, the Court noted a number of decisions as having “endorsed the suitability of arbitration to resolve federally created rights.” Ibid, (emphasis added). Footnote 15 did not address the issue of federal pre-emption of state-created rights. Rather, the import of the footnote was that the reasoning— and perhaps result—in Ware might have been different if the 1934 Act “itself ha[d] provided for arbitration.” Ibid.3 8 First among the decisions cited in footnote 15 was Wilko v. Swan, 346 U. S. 427 (1953), in which the Court resolved a conflict between the Federal Securities Act of 1933 and the Federal Arbitration Act by holding that the policies of the former prevailed and that an arbitration agreement, for which enforcement was sought under the latter, was invalid. No federal pre-emption question was presented. Only the unexplained citation to Prima Paint Corp. v. Flood & Conklin Mfg. Co., 388 U. S. 395 (1967), could be construed as a reference to principles of federal pre-emption. However, that case provides no support for Thomas’ position. It arose as a diversity action in which one party to a contract containing an arbitration clause asserted a right under state law to judicial resolution of his claim of fraud in the inducement of the contract. The Court held that, while § 2 of the Federal Arbitration Act authorized judicial determination of a claim that the arbitration clause itself had been procured through fraud, a court could not decide whether fraud had induced the making and performance of the contract generally since this claim fell within the broad scope of the agreement to arbitrate. Id., at 404. The Court dismissed the argument that the asserted right to judicial resolution adhered in state substantive law which a federal court sitting in diversity was bound to follow under the rule of Erie R. Co. v. Tompkins, 304 U. S. 64 (1938). It reasoned instead that Congress had enacted the substantive provisions of the Federal Arbitration Act pursuant, in part, to its constitutional power to regulate interstate commerce, 388 U. S., 492 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. Our holding that §2 of the Federal Arbitration Act preempts § 229 of the California Labor Code obviates any need to consider whether our decision in Byrd, supra, at 221, would have required severance of Thomas’ ancillary claims for conversion, civil conspiracy, and breach of fiduciary duty from his breach-of-contract claim. We likewise decline to reach Thomas’ contention that Perry and Johnston lack “standing” to enforce the agreement to arbitrate any of these claims, since the courts below did not address this alternative argument for refusing to compel arbitration. However, we do reject Thomas’ contention that resolving these questions in appellants’ favor is a prerequisite to their having standing under Article III of the Constitution to maintain the present appeal before this Court. As we perceive it, Thomas’ “standing” argument simply presents a straightforward issue of contract interpretation: whether the arbitration provision inures to the benefit of appellants and may be construed, in light of the circumstances surrounding the litigants’ agreement, to cover the dispute that has arisen between them. This issue may be resolved on remand; its status as an alternative ground for denying arbitration does not prevent us from reviewing the ground exclusively relied upon by the courts below.9 at 404-405, a distinction which endows these provisions with pre-emptive force under the Supremacy Clause. 9 We also decline to address Thomas’ claim that the arbitration agreement in this case constitutes an unconscionable, unenforceable contract of adhesion. This issue was not decided below, see nn. 4 and 6, supra, and may likewise be considered on remand. We note, however, the choice-of-law issue that arises when defenses such as Thomas’ so-called “standing” and unconscionability arguments are asserted. In instances such as these, the text of § 2 provides the touchstone for choosing between state-law principles and the principles of federal common law envisioned by the passage of that statute: An agreement to arbitrate is valid, irrevocable, and enforceable, as a matter of federal law, see Moses H. Cone Memorial Hospital v. Mercury Construction Corp., 460 U. S. 1, 24 (1983), “save upon such grounds as exist at law or in equity for the revocation of any contract.” 9 U. S. C. § 2 (emphasis PERRY v. THOMAS 493 483 Stevens, J., dissenting III The judgment of the California Court of Appeal is reversed, and the case is remanded for further proceedings not inconsistent with this opinion. It is so ordered. Justice Stevens, dissenting. Despite the striking similarity between this case and Merrill Lynch, Pierce, Fenner & Smith, Inc. v. Ware, 414 U. S. 117 (1973), the Court correctly concludes that the precise question now presented was not decided in Ware. Even though the Arbitration Act had been on the books for almost 50 years in 1973, apparently neither the Court nor the litigants even considered the possibility that the Act had preempted state-created rights. It is only in the last few years that the Court has effectively rewritten the statute to give it a pre-emptive scope that Congress certainly did not intend. See Southland Corp. v. Keating, 465 U. S. 1, 18-21 (1984) (Stevens, J., concurring in part and dissenting in part). The dicta in some of these recent cases are admittedly broad enough to cover this case, see ante, at 489-491, but since none of our prior holdings is on point, the doctrine of stare decisis is not controlling. Cf. Shears on/American Express Inc. v. McMahon, ante, at 268-269 (Stevens, J., concurring in part and dissenting in part). Accordingly, because I share added). Thus state law, whether of legislative or judicial origin, is applicable if that law arose to govern issues concerning the validity, revocabil-ity, and enforceability of contracts generally. A state-law principle that takes its meaning precisely from the fact that a contract to arbitrate is at issue does not comport with this requirement of § 2. See Prima Paint, supra, at 404; Southland Corp. v. Keating, 465 U. S., at 16-17, n. 11. A court may not, then, in assessing the rights of litigants to enforce an arbitration agreement, construe that agreement in a manner different from that in which it otherwise construes nonarbitration agreements under state law. Nor may a court rely on the uniqueness of an agreement to arbitrate as a basis for a state-law holding that enforcement would be unconscionable, for this would enable the court to effect what we hold today the state legislature cannot. 494 OCTOBER TERM, 1986 O’Connor, J., dissenting 482 U. S. Justice O’Connor’s opinion that the States’ power to except certain categories of disputes from arbitration should be preserved unless Congress decides otherwise, I would affirm the judgment of the California Court of Appeal. Justice O’Connor, dissenting. The Court today holds that § 2 of the Federal Arbitration Act (Act), 9 U. S. C. §1 et seq., requires the arbitration of appellee’s claim for wages despite clear state policy to the contrary. This Court held in Southland Corp. v. Keating, 465 U. S. 1 (1984), that the Act applies to state court as well as federal court proceedings. Because I continue to believe that this holding was “unfaithful to congressional intent, unnecessary, and in light of the [Act’s] antecedents and the intervening contraction of federal power, inexplicable,” id., at 36 (O’Connor, J., dissenting), I respectfully dissent. Even if I were not to adhere to my position that the Act is inapplicable to state court proceedings, however, I would still dissent. We have held that Congress can limit or preclude a waiver of a judicial forum, and that Congress’ intent to do so will be deduced from a statute’s text or legislative history, or “from an inherent conflict between arbitration and the statute’s underlying purposes.” Shear son/American Express Inc. v. McMahon, ante, at 227. As Justice Stevens has observed, the Court has not explained why state legislatures should not also be able to limit or preclude waiver of a judicial forum: “We should not refuse to exercise independent judgment concerning the conditions under which an arbitration agreement, generally enforceable under the Act, can be held invalid as contrary to public policy simply because the source of the substantive law to which the arbitration agreement attaches is a State rather than the Federal Government. I find no evidence that Congress intended such a double standard to apply, and I would not lightly impute such an intent to the 1925 Congress PERRY v. THOMAS 495 483 O’Connor, J., dissenting which enacted the Arbitration Act.” Southland Corp. v. Keating, supra, at 21. Under the standards we most recently applied in Shear-son/American Express Inc. n. McMahon, ante, p. 220, there can be little doubt that the California Legislature intended to preclude waiver of a judicial forum; it is clear, moreover, that this intent reflects an important state policy. Section 229 of the California Labor Code specifically provides that actions for the collection of wages may be maintained in the state courts “without regard to the existence of any private agreement to arbitrate.” Cal. Lab. Code Ann. §229 (West 1971). The California Legislature thereby intended “to protect the worker from the exploitative employer who would demand that a prospective employee sign away in advance his right to resort to the judicial system for redress of an employment grievance,” and § 229 has “manifested itself as an important state policy through interpretation by the California courts.” Merrill Lynch, Pierce, Fenner & Smith v. Ware, 414 U. S. 117, 131, 132-133 (1973). In my view, therefore, even if the Act applies to state court proceedings, California’s policy choice to preclude waivers of a judicial forum for wage claims is entitled to respect. Accordingly, I would affirm the judgment of the California Court of Appeal. 496 OCTOBER TERM, 1986 Syllabus 482 U. S. BOOTH v. MARYLAND CERTIORARI TO THE COURT OF APPEALS OF MARYLAND No. 86-5020. Argued March 24, 1987—Decided June 15, 1987 Having found petitioner guilty of two counts of first-degree murder and related crimes, the jury sentenced him to death after considering a presentence report prepared by the State of Maryland. The report included a victim impact statement (VIS), as required by state statute. The VIS was based on interviews with the family of the two victims, and it provided the jury with two types of information. First, it described the severe emotional impact of the crimes on the family, and the personal characteristics of the victims. Second, it set forth the family members’ opinions and characterizations of the crimes and of petitioner. The state trial court denied petitioner’s motion to suppress the VIS, rejecting the argument that this information was irrelevant, unduly inflammatory, and therefore violative of the Eighth Amendment. The Maryland Court of Appeals affirmed petitioner’s conviction and sentence, finding that the VIS did not inject an arbitrary factor into the sentencing decision. The court concluded that a VIS serves an important interest by informing the sentencer of the full measure of harm caused by the crime. Held: The introduction of a VIS at the sentencing phase of a capital murder trial violates the Eighth Amendment, and therefore the Maryland statute is invalid to the extent it requires consideration of this information. Such information is irrelevant to a capital sentencing decision, and its admission creates a constitutionally unacceptable risk that the jury may impose the death penalty in an arbitrary and capricious manner. Pp. 503-509. (a) The State’s contention that the presence or absence of emotional distress of the victims’ family and the victims’ personal characteristics are proper sentencing considerations in a capital case is rejected. In such a case, the sentencing jury must focus on the background and record of the accused and the particular circumstances of the crime. The VIS information in question may be wholly unrelated to the blameworthiness of a particular defendant, and may cause the sentencing decision to turn on irrelevant factors such as the degree to which the victim’s family is willing and able to articulate its grief, or the relative worth of the victim’s character. Thus, the evidence in question could improperly divert the jury’s attention away from the defendant. Moreover, it would be difficult, if not impossible, to provide a fair opportunity to BOOTH v. MARYLAND 497 496 Opinion of the Court rebut such evidence without shifting the focus of the sentencing hearing away from the defendant. Pp. 503-507. (b) The admission of the family members’ emotionally charged opinions and characterizations of the crimes could serve no other purpose than to inflame the jury and divert it from deciding the case on the relevant evidence concerning the crime and the defendant. Such admission is therefore inconsistent with the reasoned decisionmaking required in capital cases. Pp. 508-509. 306 Md. 172, 507 A. 2d 1098, vacated in part and remanded. Powell, J., delivered the opinion of the Court, in which Brennan, Marshall, Blackmun, and Stevens, JJ., joined. White, J., filed a dissenting opinion, in which Rehnquist, C. J., and O’Connor and Scalia, JJ., joined, post, p. 515. Scalia, J., filed a dissenting opinion, in which Rehnquist, C. J., and White and O’Connor, JJ., joined, post, p. 519. George E. Bums, Jr., argued the cause for petitioner. With him on the brief were Alan H. Murrell and Julia Doyle Bernhardt. Charles O. Monk II, Deputy Attorney General of Maryland, argued the cause for respondent. With him on the brief were J. Joseph Curran, Jr., Attorney General, and Valerie V. Cloutier, Assistant Attorney General.* Justice Powell delivered the opinion of the Court. The question presented is whether the Constitution prohibits a jury from considering a “victim impact statement” during the sentencing phase of a capital murder trial. I In 1983, Irvin Bronstein, 78, and his wife Rose, 75, were robbed and murdered in their West Baltimore home. The murderers, John Booth and Willie Reid, entered the victims’ * Julius L. Chambers, James M. Nabrit III, John Charles Boger, Vivian Berger, and Anthony G. Amsterdam filed a brief for the NAACP Legal Defense and Educational Fund, Inc., as amicus curiae urging reversal. Louis J. DiTrani filed a brief for the Stephanie Roper Foundation, Inc., as amicus curiae urging affirmance. 498 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. home for the apparent purpose of stealing money to buy heroin. Booth, a neighbor of the Bronsteins, knew that the elderly couple could identify him. The victims were bound and gagged, and then stabbed repeatedly in the chest with a kitchen knife. The bodies were discovered two days later by the Bronsteins’ son. A jury found Booth guilty of two counts of first-degree murder, two counts of robbery, and conspiracy to commit robbery.1 The prosecution requested the death penalty, and Booth elected to have his sentence determined by the jury instead of the judge. See Md. Ann. Code, Art. 27, § 413(b) (1982). Before the sentencing phase began, the State Division of Parole and Probation (DPP) compiled a presentence report that described Booth’s background, education and employment history, and criminal record. Under a Maryland statute, the presentence report in all felony cases2 also must include a victim impact statement (VIS), describing the effect of the crime on the victim and his family. Md. Ann. Code, Art. 41, §4-609(c) (1986). Specifically, the report shall: “(i) Identify the victim of the offense; “(ii) Itemize any economic loss suffered by the victim as a result of the offense; 1 Booth’s accomplice, Willie Reid, was convicted and sentenced to death as a principal in the first degree to the murder of Mrs. Bronstein. His conviction was affirmed and his sentence is currently under review. See Reid n. State, 305 Md. 9, 501 A. 2d 436 (1985). 2 When the statute was enacted it was unclear whether a VIS was admissible in a capital case. See § 4-609(c)(2)(i) (1986) (VIS required if victim suffered injury, whereas for a misdemeanor, VIS required if victim suffers injury or death); Lodowski v. State, 302 Md. 691, 761, 490 A. 2d 1228, 1264 (1985) (Cole, J., concurring), vacated on other grounds, 475 U. S. 1078 (1986). In 1983, the Maryland General Assembly amended the VIS provision to provide that: “In any case in which the death penalty is requested ... a presentence investigation, including a victim impact statement, shall be completed by the Division of Parole and Probation, and shall be considered by the court or jury before whom the separate sentencing proceeding is conducted . . . .” §4-609(d) (1986). BOOTH v. MARYLAND 499 496 Opinion of the Court “(iii) Identify any physical injury suffered by the victim as a result of the offense along with its seriousness and permanence; “(iv) Describe any change in the victim’s personal welfare or familial relationships as a result of the offense; “(v) Identity any request for psychological services initiated by the victim or the victim’s family as a result of the offense; and “(vi) Contain any other information related to the impact of the offense upon the victim or the victim’s family that the trial court requires.” § 4-609(c)(3). Although the VIS is compiled by the DPP, the information is supplied by the victim or the victim’s family. See §§ 4— 609(c)(4), (d). The VIS may be read to the jury during the sentencing phase, or the family members may be called to testify as to the information. The VIS in Booth’s case was based on interviews with the Bronsteins’ son, daughter, son-in-law, and granddaughter. Many of their comments emphasized the victims’ outstanding personal qualities, and noted how deeply the Bronsteins would be missed.3 Other parts of the VIS described the emotional and personal problems the family members have faced as a result of the crimes. The son, for example, said 8 The VIS stated: “[T]he victims’ son reports that his parents had been married for fifty-three years and enjoyed a very close relationship, spending each day together. He states that his father had worked hard all his life and had been retired for eight years. He describes his mother as a woman who was young at heart and never seemed like an old lady. She taught herself to play bridge when she was in her seventies. The victims’ son relates that his parents were amazing people who attended the senior citizens’ center and made many devout friends.” App. 59. “As described by their family members, the Bronsteins were loving parents and grandparents whose family was most important to them. Their funeral was the largest in the history of the Levinson Funeral Home and the family received over one thousand sympathy cards, some from total strangers.” Id., at 63. 500 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. that he suffers from lack of sleep and depression, and is “fearful for the first time in his life.” App. 61. He said that in his opinion, his parents were “butchered like animals.” Ibid. The daughter said she also suffers from lack of sleep, and that since the murders she has become withdrawn and distrustful. She stated that she can no longer watch violent movies or look at kitchen knives without being reminded of the murders. The daughter concluded that she could not forgive the murderer, and that such a person could “[n]ever be rehabilitated.” Id., at 62. Finally, the granddaughter described how the deaths had ruined the wedding of another close family member that took place a few days after the bodies were discovered. Both the ceremony and the reception were sad affairs, and instead of leaving for her honeymoon, the bride attended the victims’ funeral. The VIS also noted that the granddaughter had received counseling for several months after the incident, but eventually had stopped because she concluded that “no one could help her.” Id., at 63. The DPP official who conducted the interviews concluded the VIS by writing: “It became increasingly apparent to the writer as she talked to the family members that the murder of Mr. and Mrs. Bronstein is still such a shocking, painful, and devastating memory to them that it permeates every aspect of their daily lives. It is doubtful that they will ever be able to fully recover from this tragedy and not be haunted by the memory of the brutal manner in which their loved ones were murdered and taken from them.” Id., at 63-64.4 Defense counsel moved to suppress the VIS on the ground that this information was both irrelevant and unduly inflammatory, and that therefore its use in a capital case violated 4 The complete VIS is reprinted in the Appendix to this opinion. BOOTH v. MARYLAND 501 496 Opinion of the Court the Eighth Amendment of the Federal Constitution.5 The Maryland trial court denied the motion, ruling that the jury was entitled to consider “any and all evidence which would bear on the [sentencing decision].” Id., at 6. Booth’s lawyer then requested that the prosecutor simply read the VIS to the jury rather than call the family members to testify before the jury. Defense counsel was concerned that the use of live witnesses would increase the inflammatory effect of the information. The prosecutor agreed to this arrangement. The jury sentenced Booth to death for the murder of Mr. Bronstein and to life imprisonment for the murder of Mrs. Bronstein. On automatic appeal, the Maryland Court of Appeals affirmed the conviction and the sentences. 306 Md. 172, 507 A. 2d 1098 (1986). The court rejected Booth’s claim that the VIS injected an arbitrary factor into the sentencing decision. The court noted that it had considered this argument in Lodowski v. State, 302 Md. 691, 490 A. 2d 1228 (1985), vacated on other grounds, 475 U. S. 1078 (1986), and concluded that a VIS serves an important interest by informing the sentencer of the full measure of harm caused by the crime. The Court of Appeals then examined the VIS in Booth’s case, and concluded that it is a “relatively straightforward and factual description of the effects of these murders on members of the Bronstein family.” 306 Md., at 223, 507 A. 2d, at 1124. It held that the death sentence had not been imposed under the influence of passion, prejudice, or other arbitrary factors. See Md. Ann. Code, Art. 27, § 414(e)(1) (1982). We granted certiorari to decide whether the Eighth Amendment prohibits a capital sentencing jury from consid- 6 The Eighth Amendment provides: “Excessive bail shall not be required, nor excessive fines imposed, nor cruel and unusual punishments inflicted.” The prohibitions of the Eighth Amendment apply to the States through the Due Process Clause of the Fourteenth Amendment. See Robinson v. California, 370 U. S. 660, 666 (1962). 502 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. ering victim impact evidence. 479 U. S. 882 (1986). We conclude that it does, and now reverse. II It is well settled that a jury’s discretion to impose the death sentence must be “suitably directed and limited so as to minimize the risk of wholly arbitrary and capricious action.” Gregg v. Georgia, 428 U. S. 153, 189 (1976) (joint opinion of Stewart, Powell, and Stevens, JJ.); California v. Ramos, 463 U. S. 992, 999 (1983). Although this Court normally will defer to a state legislature’s determination of what factors are relevant to the sentencing decision, the Constitution places some limits on this discretion. See, e. g., id., at 1000-1001. Specifically, we have said that a jury must make an “individualized determination” whether the defendant in question should be executed, based on “the character of the individual and the circumstances of the crime.” Zant n. Stephens, 462 U. S. 862, 879 (1983) (emphasis in original). See also Eddings n. Oklahoma, 455 U. S. 104, 112 (1982). And while this Court has never said that the defendant’s record, characteristics, and the circumstances of the crime are the only permissible sentencing considerations, a state statute that requires consideration of other factors must be scrutinized to ensure that the evidence has some bearing on the defendant’s “personal responsibility and moral guilt.” Enmund v. Florida, 458 U. S. 782, 801 (1982). To do otherwise would create the risk that a death sentence will be based on considerations that are “constitutionally impermissible or totally irrelevant to the sentencing process.” See Zant n. Stephens, supra, at 885. The VIS in this case provided the jury with two types of information. First, it described the personal characteristics of the victims and the emotional impact of the crimes on the family. Second, it set forth the family members’ opinions and characterizations of the crimes and the defendant. For the reasons stated below, we find that this information is BOOTH v. MARYLAND 503 496 Opinion of the Court irrelevant to a capital sentencing decision, and that its admission creates a constitutionally unacceptable risk that the jury may impose the death penalty in an arbitrary and capricious manner. A The greater part of the VIS is devoted to a description of the emotional trauma suffered by the family and the personal characteristics of the victims. The State claims that this evidence should be considered a “circumstance” of the crime because it reveals the full extent of the harm caused by Booth’s actions. In the State’s view, there is a direct, foreseeable nexus between the murders and the harm to the family, and thus it is not “arbitrary” for the jury to consider these consequences in deciding whether to impose the death penalty. Although “victim impact” is not an aggravating factor under Maryland law,6 the State claims that by knowing the extent 6 Before the jury may impose a capital sentence, it must find that at least one of the following aggravating circumstances are present: “(1) The victim was a law enforcement officer who was murdered while in the performance of his duties. “(2) The defendant committed the murder at a time when he was confined in any correctional institution. “(3) The defendant committed the murder in furtherance of an escape or an attempt to escape from or to evade the lawful custody, arrest, or detention of or by an officer or guard of a correctional institution or by a law enforcement officer. “(4) The victim was taken or attempted to be taken in the course of a kidnapping or abduction, or an attempt to kidnap or abduct. “(5) The victim was a child abducted in violation of § 2 of this article. “(6) The defendant committed the murder pursuant to an agreement or contract for remuneration or the promise of remuneration to commit the murder. “(7) The defendant engaged or employed another person to commit the murder and the murder was committed pursuant to an agreement or contract for remuneration or the promise of remuneration. “(8) At the time of the murder the defendant was under sentence of death or imprisonment for life. “(9) The defendant committed more than one offense of murder in the first degree arising out of the same incident. [Footnote 6 is continued on p. 504] 504 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. of the impact upon and the severity of the loss to the family, the jury was better able to assess the “ ‘gravity or aggravating quality’ ” of the offense. Brief for Respondent 21 (quoting Lodowski v. State, 302Md., at 741-742,490 A. 2d, at 1254). While the full range of foreseeable consequences of a defendant’s actions may be relevant in other criminal and civil contexts, we cannot agree that it is relevant in the unique circumstance of a capital sentencing hearing. In such a case, it is the function of the sentencing jury to “express the conscience of the community on the ultimate question of life or death.” Witherspoon v. Illinois, 391 U. S. 510, 519 (1968). When carrying out this task the jury is required to focus on the defendant as a “uniquely individual human bein[g].” Woodson v. North Carolina, 428 U. S. 280, 304 (1976) (plurality opinion of Stewart, Powell, and Stevens, JJ.). The focus of a VIS, however, is not on the defendant, but on the character and reputation of the victim and the effect on his family. These factors may be wholly unrelated to the blameworthiness of a particular defendant. As our cases have shown, the defendant often will not know the victim, and therefore will have no knowledge about the existence or characteristics of the victim’s family. Moreover, defendants rarely select their victims based on whether the murder will have an effect on anyone other than the person murdered.7 “(10) The defendant committed the murder while committing or attempting to commit a robbery, arson, or rape, or sexual offense in the first degree.” See Md. Ann. Code, Art. 27, § 413(d) (1982 and Supp. 1986). Because the impact of the crime on the victim is not a statutorily defined aggravating circumstance, it would not be sufficient, standing alone, to support a capital sentence. § 413(f). 7 As one state court has noted: “We think it obvious that a defendant’s level of culpability depends not on fortuitous circumstances such as the composition of his victim’s family, but on circumstances over which he has control. A defendant may choose, or decline, to premeditate, to act callously, to attack a vulnerable victim, to commit a crime while on probation, or to amass a record of offenses. . . . In contrast, the fact that a victim’s family is irredeemably bereaved can be BOOTH v. MARYLAND 505 496 Opinion of the Court Allowing the jury to rely on a VIS therefore could result in imposing the death sentence because of factors about which the defendant was unaware, and that were irrelevant to the decision to kill. This evidence thus could divert the jury’s attention away from the defendant’s background and record, and the circumstances of the crime. It is true that in certain cases some of the information contained in a VIS will have been known to the defendant before he committed the offense. As we have recognized, a defendant’s degree of knowledge of the probable consequences of his actions may increase his moral culpability in a constitutionally significant manner. See Tison v. Arizona, 481 U. S. 137, 157-158 (1987). We nevertheless find that because of the nature of the information contained in a VIS, it creates an impermissible risk that the capital sentencing decision will be made in an arbitrary manner. As evidenced by the full text of the VIS in this case, see Appendix to this opinion, the family members were articulate and persuasive in expressing their grief and the extent of their loss. But in some cases the victim will not leave behind a family, or the family members may be less articulate in describing their feelings even though their sense of loss is equally severe. The fact that the imposition of the death sentence may turn on such distinctions illustrates the danger of allowing juries to consider this information. Certainly the degree to which a family is willing and able to express its grief is irrelevant to the decision whether a defendant, who may merit the death penalty, should live or die. See 306 Md., at 233, 507 A. 2d, at 1129 (Cole, J., concurring in part and dissenting in part) (concluding that it is arbitrary to make capital sentencing decisions based on a VIS, “which attributable to no act of will of the defendant other than his commission of homicide in the first place. Such bereavement is relevant to damages in a civil action, but it has no relationship to the proper purposes of sentencing in a criminal case.” People v. Levitt, 156 Cal. App. 3d 500, 516-517, 203 Cal. Rptr. 276, 287-288 (1984). 506 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. vary greatly from case to case depending upon the ability of the family member to express his grief”). Nor is there any justification for permitting such a decision to turn on the perception that the victim was a sterling member of the community rather than someone of questionable character.8 This type of information does not provide a “principled way to distinguish [cases] in which the death penalty was imposed, from the many cases in which it was not.” Godfrey v. Georgia, 446 U. S. 420, 433 (1980) (opinion of Stewart, J.). See also Skipper n. South Carolina, 476 U. S. 1, 14-15 (1986) (Powell, J., concurring in judgment). We also note that it would be difficult—if not impossible— to provide a fair opportunity to rebut such evidence without shifting the focus of the sentencing hearing away from the defendant. A threshold problem is that victim impact information is not easily susceptible to rebuttal. Presumably the defendant would have the right to cross-examine the declarants, but he rarely would be able to show that the family members have exaggerated the degree of sleeplessness, depression, or emotional trauma suffered. Moreover, if the state is permitted to introduce evidence of the victim’s personal qualities,9 it cannot be doubted that the defendant also 8 We are troubled by the implication that defendants whose victims were assets to their community are more deserving of punishment than those whose victims are perceived to be less worthy. Of course, our system of justice does not tolerate such distinctions. Cf. Furman v. Georgia, 408 U. S. 238, 242 (1972) (Douglas, J., concurring). 9 See n. 3, supra. The Maryland sentencing statute does not expressly permit evidence of the victim’s character and community status to be included in the VIS. The Maryland Court of Appeals, however, apparently has determined that the statute only establishes the minimum amount of information that must be provided. Consideration of other information in the VIS is subject to the trial judge’s discretion. See Reid n. State, 302 Md. 811, 820-821, 490 A. 2d 1289, 1294 (1985). This type of information is not unique to the VIS in Booth’s case. In Lodowski n. State, the trial court admitted a VIS based on an interview with the victim’s wife that said in part: BOOTH v. MARYLAND 507 496 Opinion of the Court must be given the chance to rebut this evidence. See Gardner v. Florida, 430 U. S. 349, 362 (1977) (opinion of Stevens, J.) (due process requires that defendant be given a chance to rebut presentence report). See also Md. Ann. Code, Art. 27, § 413(c)(v) (1982). Putting aside the strategic risks of attacking the victim’s character before the jury, in appropriate cases the defendant presumably would be permitted to put on evidence that the victim was of dubious moral character, was unpopular, or was ostracized from his family. The prospect of a “mini-trial” on the victim’s character is more than simply unappealing; it could well distract the sentencing jury from its constitutionally required task— determining whether the death penalty is appropriate in light of the background and record of the accused and the particular circumstances of the crime. We thus reject the contention that the presence or absence of emotional distress of the victim’s family, or the victim’s personal characteristics, are proper sentencing considerations in a capital case.10 “[The victim] was the perfect family person, he was totally devoted to his family. It was like a miracle to find a man like him—we had something very special. We had created a love that could withstand anything in life. We were not only husband and wife, but best friends.” 302 Md., at 766, 490 A. 2d, at 1266 (Cole, J., concurring) The court in Lodowski found that VIS evidence in general is not constitutionally proscribed, and is relevant to a capital sentencing determination. Id., at 751, 752, 490 A. 2d, at 1259. 10 Our disapproval of victim impact statements at the sentencing phase of a capital case does not mean, however, that this type of information will never be relevant in any context. Similar types of information may well be admissible because they relate directly to the circumstances of the crime. Facts about the victim and family also may be relevant in a noncapital criminal trial. Moreover, there may be times that the victim’s personal characteristics are relevant to rebut an argument offered by the defendant. See, e. g., Fed. Rule Evid. 404(a)(2) (prosecution may show peaceable nature of victim to rebut charge that victim was aggressor). The trial judge, of course, continues to have the primary responsibility for deciding when this information is sufficiently relevant to some legitimate 508 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. B The second type of information presented to the jury in the VIS was the family members’ opinions and characterizations of the crimes. The Bronsteins’ son, for example, stated that his parents were “butchered like animals,” and that he “doesn’t think anyone should be able to do something like that and get away with it.” App. 61. The VIS also noted that the Bronstein’s daughter “could never forgive anyone for killing [her parents] that way. She can’t believe that anybody could do that to someone. The victims’ daughter states that animals wouldn’t do this. [The perpetrators] didn’t have to kill because there was no one to stop them from looting. ... The murders show the viciousness of the killers’ anger. She doesn’t feel that the people who did this could ever be rehabilitated and she doesn’t want them to be able to do this again or put another family through this.” Id., at 62. One can understand the grief and anger of the family caused by the brutal murders in this case, and there is no doubt that jurors generally are aware of these feelings. But the formal presentation of this information by the State can serve no other purpose than to inflame the jury and divert it from deciding the case on the relevant evidence concerning the crime and the defendant. As we have noted, any decision to impose the death sentence must “be, and appear to be, based on reason rather than caprice or emotion.” Gardner v. Florida, supra, at 358 (opinion of Stevens, J.). The admission of these emotionally charged opinions as to what conclusions the jury should draw from the evidence clearly consideration to be admissible, and when its probative value outweighs any prejudicial effect. Cf. Fed. Rule Evid. 403. BOOTH v. MARYLAND 509 496 Appendix to opinion of the Court is inconsistent with the reasoned decisionmaking we require in capital cases.11 Ill We conclude that the introduction of a VIS at the sentencing phase of a capital murder trial violates the Eighth Amendment, and therefore the Maryland statute is invalid to the extent it requires consideration of this information.12 The decision of the Maryland Court of Appeals is vacated to the extent that it affirmed the capital sentence. The case is remanded for further proceedings not inconsistent with this opinion. It is so ordered. APPENDIX TO OPINION OF THE COURT ‘VICTIM IMPACT STATEMENT [The Victim Impact Statement in this case was prepared by the Maryland Division of Parole and Probation. See n. 2, supra.] 11 The same problem is presented by the VIS summary written by the DPP that might be viewed by the jury as representing the views of the State. As noted supra, at 500, the writer concluded that the crimes had a “shocking, painful, and devast[at]ing” effect on the family, and that “[i]t is doubtful that they will ever be able to fully recover.” App. 63-64. See Appendix to this opinion. 12 We note, however, that our decision today is guided by the fact death is a “punishment different from all other sanctions,” see Woodson v. North Carolina, 428 U. S. 280, 303-304, 305 (1976) (plurality opinion of Stewart, Powell, and Stevens, JJ.), and that therefore the considerations that inform the sentencing decision may be different from those that might be relevant to other liability or punishment determinations. At least 36 States permit the use of victim impact statements in some contexts, reflecting a legislative judgment that the effect of the crime on victims should have a place in the criminal justice system. See National Organization for Victim Assistance, Victim Rights and Services: A Legislative Directory 32-33 (1985) (chart); McLeod, Victim Participation at Sentencing, 22 Crim. L. Bull. 501, 507, and n. 22 (1986). Congress also has provided for victim participation in federal criminal cases. See Fed. Rule Crim. Proc. 32(c)(2)(C). We imply no opinion as to the use of these statements in noncapital cases. 510 OCTOBER TERM, 1986 Appendix to opinion of the Court 482 U. S. “Mr. and Mrs. Bronstein’s son, daughter, son-in-law, and granddaughter were interviewed for purposes of the Victim Impact Statement. There are also four other grandchildren in the family. The victims’ son reports that his parents had been married for fifty-three years and enjoyed a very close relationship, spending each day together. He states that his father had worked hard all his life and had been retired for eight years. He describes his mother as a woman who was young at heart and never seemed like an old lady. She taught herself to play bridge when she was in her seventies. The victims’ son relates that his parents were amazing people who attended the senior citizens’ center and made many devout friends. He indicates that he was very close to his parents, and that he talked to them every day. The victims’ daughter also spent lots of time with them. “The victims’ son saw his parents alive for the last time on May 18th. They were having their lawn manicured and were excited by the onset of spring. He called them on the phone that evening and received no answer. He had made arrangements to pick Mr. Bronstein up on May 20th. They were both to be ushers in a granddaughter’s wedding and were going to pick up their tuxedos. When he arrived at the house on May 20th he noticed that his parents’ car wasn’t there. A neighbor told him that he hadn’t seen the car in several days and he knew something was wrong. He went to his parents’ house and found them murdered. He called his sister crying and told her to come right over because something terrible had happened and their parents were both dead. “The victims’ daughter recalls that when she arrived at her parents’ house, there were police officers and television crews everywhere. She felt numb and cold. She was not allowed to go into the house and so she went to a neighbor’s home. There were people and reporters everywhere and all she could feel was cold. She called her older daughter and told her what had happened. She told her daughter to get BOOTH v. MARYLAND 511 496 Appendix to opinion of the Court her husband and then tell her younger daughter what had happened. The younger daughter was to be married two days later. “The victims’ granddaughter reports that just before she received the call from her mother she had telephoned her grandparents and received no answer. After her mother told her what happened she turned on the television and heard the news reports about it. The victims’ son reports that his children first learned about their grandparents death from the television reports. “Since the Jewish religion dictates that birth and marriage are more important than death, the granddaughter’s wedding had to proceed on May 22nd. She had been looking forward to it eagerly, but it was a sad occasion with people crying. The reception, which normally would have lasted for hours, was very brief. The next day, instead of going on her honeymoon, she attended her grandparents’ funerals. The victims’ son, who was an usher at the wedding, cannot remember being there or coming and going from his parents’ funeral the next day. The victims’ granddaughter, on the other hand, vividly remembers every detail of the days following her grandparents’ death. Perhaps she described the impact of the tragedy most eloquently when she stated that it was a completely devastating and life altering experience. “The victims’ son states that he can only think of his parents in the context of how he found them that day, and he can feel their fear and horror. It was 4:00 p.m. when he discovered their bodies and this stands out in his mind. He is always aware of when 4:00 p.m. comes each day, even when he is not near a clock. He also wakes up at 4:00 a.m. each morning. The victims’ son states that he suffers from lack of sleep. He is unable to drive on the streets that pass near his parents’ home. He also avoids driving past his father’s favorite restaurant, the supermarket where his parents shopped, etc. He is constantly reminded of his parents. He sees his father coming out of synagogues, sees his parents’ 512 OCTOBER TERM, 1986 Appendix to opinion of the Court 482 U. S. car, and feels very sad whenever he sees old people. The victims’ son feels that his parents were not killed, but were butchered like animals. He doesn’t think anyone should be able to do something like that and get away with it. He is very angry and wishes he could sleep and not feel so depressed all the time. He is fearful for the first time in his life, putting all the lights on and checking the locks frequently. His children are scared for him and concerned for his health. They phone him several times a day. At the same time he takes a fearful approach to the whereabouts of his children. He also calls his sister every day. He states that he is frightened by his own reaction of what he would do if someone hurt him or a family member. He doesn’t know if he’ll ever be the same again. “The victims’ daughter and her husband didn’t eat dinner for three days following the discovery of Mr. and Mrs. Bronstein’s bodies. They cried together every day for four months and she still cries every day. She states that she doesn’t sleep through a single night and thinks a part of her died too when her parents were killed. She reports that she doesn’t find much joy in anything and her powers of concentration aren’t good. She feels as if her brain is on overload. The victims’ daughter relates that she had to clean out her parents’ house and it took several weeks. She saw the bloody carpet, knowing that her parents had been there, and she felt like getting down on the rug and holding her mother. She wonders how this could have happened to her family because they’re just ordinary people. The victims’ daughter reports that she had become noticeably withdrawn and depressed at work and is now making an effort to be more outgoing. She notes that she is so emotionally tired because she doesn’t sleep at night, that she has a tendency to fall asleep when she attends social events such as dinner parties or the symphony. The victims’ daughter states that wherever she goes she sees and hears her parents. This happens every day. She cannot look at kitchen knives without being re BOOTH v. MARYLAND 513 496 Appendix to opinion of the Court minded of the murders and she is never away from it. She states that she can’t watch movies with bodies or stabbings in it. She can’t tolerate any reminder of violence. The victims’ daughter relates that she used to be very trusting, but is not any longer. When the doorbell rings she tells her husband not to answer it. She is very suspicious of people and was never that way before. “The victims’ daughter attended the defendant’s trial and that of the co-defendant because she felt someone should be there to represent her parents. She had never been told the exact details of her parents’ death and had to listen to the medical examiner’s report. After a certain point, her mind blocked out and she stopped hearing. She states that her parents were stabbed repeatedly with viciousness and she could never forgive anyone for killing them that way. She can’t believe that anybody could do that to someone. The victims’ daughter states that animals wouldn’t do this. They didn’t have to kill because there was no one to stop them from looting. Her father would have given them anything. The murders show the viciousness of the killers’ anger. She doesn’t feel that the people who did this could ever be rehabilitated and she doesn’t want them to be able to do this again or put another family through this. She feels that the lives of her family members will never be the same again. “The victims’ granddaughter states that unless you experience something like this you can’t understand how it feels. You are in a state of shock for several months and then a terrible depression sets in. You are so angry and feel such rage. She states that she only dwells on the image of their death when thinking of her grandparents. For a time she would become hysterical whenever she saw dead animals on the road. She is not able to drive near her grandparents’ house and will never be able to go into their neighborhood again. The victims’ granddaughter also has a tendency to turn on all the lights in her house. She goes into a panic if 514 OCTOBER TERM, 1986 Appendix to opinion of the Court 482 U. S. her husband is late coming home from work. She used to be an avid reader of murder mysteries, but will never be able to read them again. She has to turn off the radio or T. V. when reports of violence come on because they hit too close to home. When she gets a newspaper she reads the comics and throws the rest away. She states that it is the small everyday things that haunt her constantly and always will. She saw a counselor for several months but stopped because she felt that no one could help her. “The victims’ granddaughter states that the whole thing has been very hard on her sister too. Her wedding anniversary will always be bittersweet and tainted by the memory of what happened to her grandparents. This year on her anniversary she and her husband quietly went out of town. The victims’ granddaughter finds that she is unable to look at her sister’s wedding pictures. She also has a picture of her grandparents, but had to put it away because it was too painful to look at it. “The victims’ family members note that the trials of the suspects charged with these offenses have been delayed for over a year and the postponements have been very hard on the family emotionally. The victims’ son notes that he keeps seeing news reports about his parents’ murder which show their house and the police removing their bodies. This is a constant reminder to him. The family wants the whole thing to be over with and they would like to see swift and just punishment. “As described by their family members, the Bronsteins were loving parents and grandparents whose family was most important to them. Their funeral was the largest in the history of the Levinson Funeral Home and the family received over one thousand sympathy cards, some from total strangers. They attempted to answer each card personally. The family states that Mr. and Mrs. Bronstein were extremely good people who wouldn’t hurt a fly. Because of their loss, a terrible void has been put into their lives and BOOTH v. MARYLAND 515 496 White, J., dissenting every day is still a strain just to get through. It became increasingly apparent to the writer as she talked to the family members that the murder of Mr. and Mrs. Bronstein is still such a shocking, painful, and devastating memory to them that it permeates every aspect of their daily lives. It is doubtful that they will ever be able to fully recover from this tragedy and not be haunted by the memory of the brutal manner in which their loved ones were murdered and taken from them.” App. 59-64. Justice White, with whom The Chief Justice, Justice O’Connor, and Justice Scalia join, dissenting. “[T]he decision that capital punishment may be the appropriate sanction in extreme cases is an expression of the community’s belief that certain crimes are themselves so grievous an affront to humanity that the only adequate response may be the penalty of death.” Gregg v. Georgia, 428 U. S. 153, 184 (1976) (joint opinion of Stewart, Powell, and Stevens, JJ.). The affront to humanity of a brutal murder such as petitioner committed is not limited to its impact on the victim or victims; a victim’s community is also injured, and in particular the victim’s family suffers shock and grief of a kind difficult even to imagine for those who have not shared a similar loss. Maryland’s legislature has decided that the jury should have the testimony of the victim’s family in order to assist it in weighing the degree of harm that the defendant has caused and the corresponding degree of punishment that should be inflicted. This judgment is entitled to particular deference; determinations of appropriate sentencing considerations are “‘peculiarly questions of legislative policy,’” id., at 176 (quoting Gore v. United States, 357 U. S. 386, 393 (1958)), and the Court should recognize that ‘“[i]n a democratic society legislatures, not courts, are constituted to respond to the will and consequently the moral values of the people,’” 428 U. S., at 175 (quoting Furman v. Georgia, 408 U. S. 238, 383 (1972) (Burger, C. J., dissenting)). I cannot agree that there was anything “cruel or unusual” or other 516 OCTOBER TERM, 1986 White, J., dissenting 482 U. S. wise unconstitutional about the legislature’s decision to use victim impact statements in capital sentencing hearings. The Court’s judgment is based on the premises that the harm that a murderer causes a victim’s family does not in general reflect on his blameworthiness, and that only evidence going to blameworthiness is relevant to the capital sentencing decision. Many if not most jurors, however, will look less favorably on a capital defendant when they appreciate the full extent of the the harm he caused, including the harm to the victim’s family. There is nothing aberrant in a juror’s inclination to hold a murderer accountable not only for his internal disposition in committing the crime but also for the full extent of the harm he caused; many if not most persons would also agree, for example, that someone who drove his car recklessly through a stoplight and unintentionally killed a pedestrian merits significantly more punishment than someone who drove his car recklessly through the same stoplight at a time when no pedestrian was there to be hit. I am confident that the Court would not overturn a sentence for reckless homicide by automobile merely because the punishment exceeded the maximum sentence for reckless driving; and I would hope that the Court would not overturn the sentence in such a case if a judge mentioned, as relevant to his sentencing decision, the fact that the victim was a mother or father. But if punishment can be enhanced in noncapital cases on the basis of the harm caused, irrespective of the offender’s specific intention to cause such harm,1 I fail to see Congress considers the effect of crime on its victims a relevant sentencing consideration. Thus, presentence reports prepared pursuant to Federal Rule of Criminal Procedure 32(c)(2) must include “information concerning any harm, including financial, social, psychological, and physical harm, done to or loss suffered by any victim of the offense . . . .” This Court’s cases also indicate that the harm caused by an offense may be the basis for punishment even if the offender lacked the specific intent to commit that harm. See, e. g., United States v. Feola, 420 U. S. 671 (1975) (conviction under 18 U. S. C. § 111 for assaulting a federal officer does not require proof that the defendant knew the victim’s status). BOOTH v. MARYLAND 517 496 White, J., dissenting why the same approach is unconstitutional in death cases. If anything, I would think that victim impact statements are particularly appropriate evidence in capital sentencing hearings: the State has a legitimate interest in counteracting the mitigating evidence which the defendant is entitled to put in, see, e. g., Eddings v. Oklahoma, 455 U. S. 104 (1982), by reminding the sentencer that just as the murderer should be considered as an individual, so too the victim is an individual whose death represents a unique loss to society and in particular to his family. The Court is “troubled by the implication that defendants whose victims were assets to their community are more deserving of punishment than those whose victims are perceived to be less worthy,” and declares that “our system of justice does not tolerate such distinctions.” Ante, at 506, n. 8. It is no doubt true that the State may not encourage the sentencer to rely on a factor such as the victim’s race in determining whether the death penalty is appropriate. Cf. McCleskey v. Kemp, 481 U. S. 279 (1987). But I fail to see why the State cannot, if it chooses, include as a sentencing consideration the particularized harm that an individual’s murder causes to the rest of society2 and in particular to his family. To the extent that the Court is concerned that sentencing juries might be moved by victim impact statements to rely on impermissible factors such as the race of the victim, there is no showing that the statements in this case encouraged this, nor should we lightly presume such misconduct on the jury’s part. Cf. McCleskey v. Kemp, supra. The Court’s reliance on the alleged arbitrariness that can result from the differing ability of victims’ families to articu 21 doubt that the Court means to suggest that there is any constitutional impediment, for example, to authorizing the death sentence for the assassination of the President or Vice President, see 18 U. S. C. §§ 1751, HU, a Congressman, Cabinet official, Supreme Court Justice, or the head of an executive department, 18 U. S. C. §351, or the murder of a policeman on active duty, see Md. Ann. Code, Art. 27, § 413(d)(1) (1982). 518 OCTOBER TERM, 1986 White, J., dissenting 482 U. S. late their sense of loss is a makeweight consideration: No two prosecutors have exactly the same ability to present their arguments to the jury; no two witnesses have exactly the same ability to communicate the facts; but there is no requirement in capital cases that the evidence and argument be reduced to the lowest common denominator. The supposed problems arising from a defendant’s rebuttal of victim impact statements are speculative and unconnected to the facts of this case. No doubt a capital defendant must be allowed to introduce relevant evidence in rebuttal to a victim impact statement, but Maryland has in no wise limited the right of defendants in this regard. Petitioner introduced no such rebuttal evidence, probably because he considered, wisely, that it was not in his best interest to do so.3 At bottom, the Court’s view seems to be that it is somehow unfair to confront a defendant with an account of the loss his deliberate act has caused the victim’s family and society. I do not share that view, but even if I did I would be unwilling to impose it on States that see matters differently. The Court’s concern that the grief and anger of a victim’s family will “inflame the jury,” ante, at 508, is based in large part on its view that the loss which such survivors suffer is irrelevant to the issue of punishment—a view with which I have already expressed my disagreement. To the extent that the Court determines that in this case it was inappropriate to allow the victims’ family to express their opinions on, for example, whether petitioner could be rehabilitated, that is obviously not an inherent fault in all victim impact statements and no reason to declare the practice of admitting 8 The possibility that the jury would be distracted by rebuttal evidence is purely hypothetical, since petitioner introduced no such evidence. It is also unclear how distracting (as opposed to offending) the jury would disadvantage the defendant, and why, if there were some disadvantage to the defendant in pressing too hard a rebuttal to a victim impact statement, he should be heard to complain of the consequences of his tactical decisions. BOOTH v. MARYLAND 519 496 Scalia, J., dissenting such statements at capital sentencing hearings per se unconstitutional. I respectfully dissent. Justice Scalia, with whom The Chief Justice, Justice White, and Justice O’Connor join, dissenting. The Court holds that because death is a “ ‘punishment different from all other sanctions,”’ ante, at 509, n. 12 (quoting Woodson v. North Carolina, 428 U. S. 280, 303-304 (1976) (plurality opinion of Stewart, Powell, and Stevens, JJ.)), considerations not relevant to “the defendant’s ‘personal responsibility and moral guilt’ ” cannot be taken into account in deciding whether a defendant who is eligible for the death penalty should receive it, ante, at 502 (quoting Enmund v, Florida, 458 U. S. 782, 801 (1982)). It seems to me, however—and, I think, to most of mankind—that the amount of harm one causes does bear upon the extent of his “personal responsibility.” We may take away the license of a driver who goes 60 miles an hour on a residential street; but we will put him in jail for manslaughter if, though his moral guilt is no greater, he is unlucky enough to kill someone during the escapade. Nor, despite what the Court says today, do we depart from this principle where capital punishment is concerned. The Court’s opinion does not explain why a defendant’s eligibility for the death sentence can (and always does) turn upon considerations not relevant to his moral guilt. If a bank robber aims his gun at a guard, pulls the trigger, and kills his target, he may be put to death. If the gun unexpectedly misfires, he may not. His moral guilt in both cases is identical, but his responsibility in the former is greater. Less than two months ago, we held that two brothers who planned and assisted in their father’s escape from prison could be sentenced to death because in the course of the escape their father and an accomplice murdered a married couple and two children. Tison n. Arizona, 481 U. S. 137 (1987). Had their father allowed the victims to live, the brothers could not be put to death; but because he decided to kill, the brothers may. 520 OCTOBER TERM, 1986 Scalia, J., dissenting 482 U. S. The difference between life and death for these two defendants was thus a matter “wholly unrelated to the[ir] blameworthiness.” Ante, at 504. But it was related to their personal responsibility, i. e., to the degree of harm that they had caused. In sum, the principle upon which the Court’s opinion rests—that the imposition of capital punishment is to be determined solely on the basis of moral guilt—does not exist, neither in the text of the Constitution, nor in the historic practices of our society, nor even in the opinions of this Court. Recent years have seen an outpouring of popular concern for what has come to be known as “victims’ rights”—a phrase that describes what its proponents feel is the failure of courts of justice to take into account in their sentencing decisions not only the factors mitigating the defendant’s moral guilt, but also the amount of harm he has caused to innocent members of society. Many citizens have found one-sided and hence unjust the criminal trial in which a parade of witnesses comes forth to testify to the pressures beyond normal human experience that drove the defendant to commit his crime, with no one to lay before the sentencing authority the full reality of human suffering the defendant has produced—which (and not moral guilt alone) is one of the reasons society deems his act worthy of the prescribed penalty. Perhaps these sentiments do not sufficiently temper justice with mercy, but that is a question to be decided through the democratic processes of a free people, and not by the decrees of this Court. There is nothing in the Constitution that dictates the answer, no more in the field of capital punishment than elsewhere. To require, as we have, that all mitigating factors which render capital punishment a harsh penalty in the particular case be placed before the sentencing authority, while simultaneously requiring, as we do today, that evidence of much of the human suffering the defendant has inflicted be suppressed, is in effect to prescribe a debate on the appropriateness of the capital penalty with one side muted. If that pen- BOOTH v. MARYLAND 521 496 Scalia, J., dissenting alty is constitutional, as we have repeatedly said it is, it seems to me not remotely unconstitutional to permit both the pros and the cons in the particular case to be heard. 522 OCTOBER TERM, 1986 Syllabus 482 U. S. SOCIÉTÉ NATIONALE INDUSTRIELLE AÉROSPATIALE et AL. v. UNITED STATES DISTRICT COURT FOR THE SOUTHERN DISTRICT OF IOWA CERTIORARI TO THE UNITED STATES COURT OF APPEALS FOR THE EIGHTH CIRCUIT No. 85-1695. Argued January 14, 1987—Decided June 15, 1987 The United States, France, and 15 other countries have acceded to the Hague Evidence Convention, which prescribes procedures by which a judicial authority in one contracting state may request evidence located in another. Plaintiffs brought suits (later consolidated) in Federal District Court for personal injuries resulting from the crash of an aircraft built and sold by petitioners, two corporations owned by France. Petitioners answered the complaints without questioning the court’s jurisdiction, and engaged in initial discovery without objection. However, when plaintiffs served subsequent discovery requests under the Federal Rules of Civil Procedure, petitioners filed a motion for a-protective order, alleging that the Convention dictated the exclusive procedures that must be followed since petitioners are French and the discovery sought could only be had in France. A Magistrate denied the motion, and the Court of Appeals denied petitioners’ mandamus petition, holding, inter alia, that when a district court has jurisdiction over a foreign litigant, the Convention does not apply even though the information sought may be physically located within the territory of a foreign signatory to the Convention. Held: 1. The Convention does not provide exclusive or mandatory procedures for obtaining documents and information located in a foreign signatory’s territory. The Convention’s plain language, as well as the history of its proposal and ratification by the United States, unambiguously supports the conclusion that it was intended to establish optional procedures for obtaining evidence abroad. Its preamble speaks in nonmandatory terms, specifying its purpose to “facilitate” discovery and to “improve mutual judicial co-operation.” Similarly, its text uses permissive language, and does not expressly modify the law of contracting states or require them to use the specified procedures or change their own procedures. The Convention does not deprive the District Court of its jurisdiction to order, under the Federal Rules, a foreign national party to produce evidence physically located within a signatory nation. Pp. 529-540. SOCIÉTÉ NAT. IND. AÉRO. v. U. S. DIST. COURT 523 522 Syllabus 2. The Court of Appeals erred in concluding that the Convention “does not apply” to discovery sought from a foreign litigant that is subject to an American court’s jurisdiction. Although they are not mandatory, the Convention’s procedures are available whenever they will facilitate the gathering of evidence, and “apply” in the sense that they are one method of seeking evidence that a court may elect to employ. Pp. 540-541. 3. International comity does not require in all instances that American litigants first resort to Convention procedures before initiating discovery under the Federal Rules. In many situations, Convention procedures would be unduly time consuming and expensive, and less likely to produce needed evidence than direct use of the Federal Rules. The concept of comity requires in this context a more particularized analysis of the respective interests of the foreign and requesting nations than a blanket “first resort” rule would generate. Thus, the determination whether to resort to the Convention requires prior scrutiny in each case of the particular facts, sovereign interests, and likelihood that such resort will prove effective. Pp. 541-546. 782 F. 2d 120, vacated and remanded. Stevens, J., delivered the opinion of the Court, in which Rehnquist, C. J., and White, Powell, and Scalia, JJ., joined. Blackmun, J., filed an opinion concurring in part and dissenting in part, in which Brennan, Marshall, and O’Connor, JJ., joined, post, p. 547. John W. Ford argued the cause for petitioners. With him on the briefs were Stephen C. Johnson, Lawrence N. Minch, and William L. Robinson. Jeffrey P. Minear argued the cause for the United States et al. as amici curiae. With him on the brief were Solicitor General Fried, Assistant Attorney General Willard, Deputy Solicitor General Lauber, Deputy Assistant Attorney General Spears, David Epstein, Abraham D. Sofaer, and Daniel L. Goelzer. Richard H. Doyle IV argued the cause for respondent. With him on the brief were Verne Lawyer, Roland D. Peddicord, and Thomas C. Farr* *Briefs of amici curiae urging reversal were filed for the Government of the United Kingdom of Great Britain and Northern Ireland by Douglas E. Rosenthal, Willard K. Tom, and Bruno A. Ristau; for the Republic of France by George J. Grumbach, Jr.; for the Federal Republic of Germany 524 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. Justice Stevens delivered the opinion of the Court. The United States, the Republic of France, and 15 other Nations have acceded to the Hague Convention on the Taking of Evidence Abroad in Civil or Commercial Matters, opened for signature, Mar. 18, 1970, 23 U. S. T. 2555, T. I. A. S. No. 7444? This Convention—sometimes referred to as the “Hague Convention” or the “Evidence Convention”—prescribes certain procedures by which a judicial authority in one contracting state may request evidence located in another contracting state. The question presented in this case concerns the extent to which a federal district court must employ the procedures set forth in the Convention when litigants seek answers to interrogatories, the production of documents, and admissions from a French adversary over whom the court has personal jurisdiction. I The two petitioners are corporations owned by the Republic of France.2 They are engaged in the business of design- by Peter Heidenberger; for the Government of Switzerland by Robert E. Herzstein; for Anschuetz & Co., GmbH, et al. by James S. Campbell, David Westin, Carol F. Lee, Neal D. Hobson, Andrew N. Vollmer, and Gerhard Nagorny; and for the Motor Vehicle Manufacturers Association of the United States, Inc., et al. by Michael Hoenig, Herbert Rubin, William H. Crabtree, and Edward P. Good. Paul A. Nalty, Derek A. Walker, and Kenneth J. Servay filed a brief for Compania Gijonesa de Navigacion, S. A., as amicus curiae urging affirmance. Richard L. Mattiaccio and David A. Botwinik filed a brief for the Italy-America Chamber of Commerce, Inc., as amicus curiae. JThe Hague Convention entered into force between the United States and France on October 6, 1974. The Convention is also in force in Barbados, Cyprus, Czechoslovakia, Denmark, Finland, the Federal Republic of Germany, Israel, Italy, Luxembourg, the Netherlands, Norway, Portugal, Singapore, Sweden, and the United Kingdom. Office of the Legal Adviser, United States Dept, of State, Treaties in Force 261-262 (1986). 2 Petitioner Société Nationale Industrielle Aérospatiale is wholly owned by the Government of France. Petitioner Société de Construction d’Avions SOCIÉTÉ NAT. IND. AÉRO. v. U. S. DIST. COURT 525 522 Opinion of the Court ing, manufacturing, and marketing aircraft. One of their planes, the “Rallye,” was allegedly advertised in American aviation publications as “the World’s safest and most economical STOL plane.”3 On August 19, 1980, a Rallye crashed in Iowa, injuring the pilot and a passenger. Dennis Jones, John George, and Rosa George brought separate suits based upon this accident in the United States District Court for the Southern District of Iowa, alleging that petitioners had manufactured and sold a defective plane and that they were guilty of negligence and breach of warranty. Petitioners answered the complaints, apparently without questioning the jurisdiction of the District Court. With the parties’ consent, the cases were consolidated and referred to a Magistrate. See 28 U. S. C. § 636(c)(1). Initial discovery was conducted by both sides pursuant to the Federal Rules of Civil Procedure without objection.4 When plaintiffs5 served a second request for the production of documents pursuant to Rule 34, a set of interrogatories pursuant to Rule 33, and requests for admission pursuant to Rule 36, however, petitioners filed a motion for a protective order. App. 27-37. The motion alleged that because petitioners are “French corporations, and the discovery sought de Tourisme is a wholly owned subsidiary of Société Nationale Industrielle Aérospatiale. 8 App. 22, 24. The term “STOL,” an acronym for “short takeoff and landing,” “refers to a fixed-wing aircraft that either takes off or lands with only a short horizontal run of the aircraft.” Douglas n. United States, 206 Ct. Cl. 96, 99, 510 F. 2d 364, 365, cert, denied, 423 U. S. 825 (1975). 4 Plaintiffs made certain requests for the production of documents pur- suant to Rule 34(b) and for admissions pursuant to Rule 36. App. 19-23. Apparently the petitioners responded to those requests without objection, at least insofar as they called for material or information that was located in the United States. App. to Pet. for Cert. 12a. In turn, petitioners deposed witnesses and parties pursuant to Rule 26, and served interrogatories pursuant to Rule 33 and a request for the production of documents pursuant to Rule 34. App. 13. Plaintiffs complied with those requests. 6 Although the District Court is the nominal respondent in this mandamus proceeding, plaintiffs are the real respondent parties in interest. 526 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. can only be found in a foreign state, namely France,” the Hague Convention dictated the exclusive procedures that must be followed for pretrial discovery. App. 2. In addition, the motion stated that under French penal law, the petitioners could not respond to discovery requests that did not comply with the Convention. Ibid.6 The Magistrate denied the motion insofar as it related to answering interrogatories, producing documents, and making admissions.7 After reviewing the relevant cases, the Magistrate explained: “To permit the Hague Evidence Convention to override the Federal Rules of Civil Procedure would frustrate the courts’ interests, which particularly arise in products li- 6 Article 1A of the French “blocking statute,” French Penal Code Law No. 80-538, provides: “Subject to treaties or international agreements and applicable laws and regulations, it is prohibited for any party to request, seek or disclose, in writing, orally or otherwise, economic, commercial, industrial, financial or technical documents or information leading to the constitution of evidence with a view to foreign judicial or administrative proceedings or in connection therewith. “Art. 1er bis. — Sous réserve des traités ou accords internationaux et des lois et règlements en vigueur, il est interdit à toute personne de demander, de rechercher ou de communiquer, par écrit, oralement ou sous toute autre forme, des documents ou renseignements d’ordre économique, commercial, industriel, financier ou technique tendant à la constitution de preuves en vue de procédures judiciaires ou administratives étrangères ou dans le cadre de celles-ci.” Article 2 provides: “The parties mentioned in [Article IA] shall forthwith inform the competent minister if they receive any request concerning such disclosures. “Art. 2. Les personnes visées aux articles 1er et 1er bis sont tenues d’informer sans délai le ministre compétent lorsqu’elles se trouvent saisies de toute demande concernant de telles communications.” App. to Pet. for Cert. 47a-50a. 7Id., at 25a. The Magistrate stated, however, that if oral depositions were to be taken in France, he would require compliance with the Hague Evidence Convention. Ibid. SOCIÉTÉ NAT. IND. AÉRO. v. U. S. DIST. COURT 527 522 Opinion of the Court ability cases, in protecting United States citizens from harmful products and in compensating them for injuries arising from use of such products.” App. to Pet. for Cert. 25a. The Magistrate made two responses to petitioners’ argument that they could not comply with the discovery requests without violating French penal law. Noting that the law was originally “ ‘inspired to impede enforcement of United States antitrust laws,”’8 and that it did not appear to have been strictly enforced in France, he first questioned whether it would be construed to apply to the pretrial discovery requests at issue.9 Id., at 22a-24a. Second, he balanced the interests in the “protection of United States citizens from harmful foreign products and compensation for injuries caused by such products” against France’s interest in protecting its citizens “from intrusive foreign discovery procedures.” The Magistrate concluded that the former interests were stronger, particularly because compliance with the requested discovery will “not have to take place in France” and will not be greatly intrusive or abusive. Id., at 23a-25a. Petitioners sought a writ of mandamus from the Court of Appeals for the Eighth Circuit under Federal Rule of Appellate Procedure 21(a). Although immediate appellate review of an interlocutory discovery order is not ordinarily available, see Kerr v. United States District Court, 426 U. S. 394, 8 His quotation was from Toms, The French Response to Extraterritorial Application of United States Antitrust Laws, 15 Int’l Law. 585, 586 (1981). 9 He relied on a passage in the Toms article stating that “the legislative history [of the Law] shows only that the Law was adopted to protect French interests from abusive foreign discovery procedures and excessive assertions of extraterritorial jurisdiction. Nowhere is there an indication that the Law was to impede litigation preparations by French companies, either for their own defense or to institute lawsuits abroad to protect their interests, and arguably such applications were unintended.” App. to Pet. for Cert. 22a-23a (citing Toms, supra, at 598). 528 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. 402-403 (1976), the Court of Appeals considered that the novelty and the importance of the question presented, and the likelihood of its recurrence, made consideration of the merits of the petition appropriate. 782 F. 2d 120 (1986). It then held that “when the district court has jurisdiction over a foreign litigant the Hague Convention does not apply to the production of evidence in that litigant’s possession, even though the documents and information sought may physically be located within the territory of a foreign signatory to the Convention.” Id., at 124. The Court of Appeals disagreed with petitioners’ argument that this construction would render the entire Hague Convention “meaningless,” noting that it would still serve the purpose of providing an improved procedure for obtaining evidence from nonparties. Id., at 125. The court also rejected petitioners’ contention that considerations of international comity required plaintiffs to resort to Hague Convention procedures as an initial matter (“first use”), and correspondingly to invoke the federal discovery rules only if the treaty procedures turned out to be futile. The Court of Appeals believed that the potential overruling of foreign tribunals’ denial of discovery would do more to defeat than to promote international comity. Id., at 125-126. Finally, the Court of Appeals concluded that objections based on the French penal statute should be considered in two stages: first, whether the discovery order was proper even though compliance may require petitioners to violate French law; and second, what sanctions, if any, should be imposed if petitioners are unable to comply. The Court of Appeals held that the Magistrate properly answered the first question and that it was premature to address the second.10 The court 10 “The record before this court does not indicate whether the Petitioners have notified the appropriate French Minister of the requested discovery in accordance with Article 2 of the French Blocking Statute, or whether the Petitioners have attempted to secure a waiver of prosecution from the French government. Because the Petitioners are corporations owned by SOCIÉTÉ NAT. IND. AÉRO. v. U. S. DIST. COURT 529 522 Opinion of the Court therefore denied the petition for mandamus. We granted certiorari. 476 U. S. 1168 (1986). II In the District Court and the Court of Appeals, petitioners contended that the Hague Evidence Convention “provides the exclusive and mandatory procedures for obtaining documents and information located within the territory of a foreign signatory.” 782 F. 2d, at 124.11 We are satisfied that the Court of Appeals correctly rejected this extreme position. We believe it is foreclosed by the plain language of the Convention. Before discussing the text of the Convention, however, we briefly review its history. The Hague Conference on Private International Law, an association of sovereign states, has been conducting periodic sessions since 1893. S. Exec. Doc. A, 92d Cong., 2d Sess., p. v (1972) (S. Exec. Doc. A). The United States participated in those sessions as an observer in 1956 and 1960, and as a member beginning in 1964 pursuant to congressional authorization.12 In that year Congress amended the Judicial Code to grant foreign litigants, without any requirement of reciprocity, special assistance in obtaining evidence in the the Republic of France, they stand in a most advantageous position to receive such a waiver. However, these issues will only be relevant should the Petitioners fail to comply with the magistrate’s discovery order, and we need not presently address them.” 782 F. 2d, at 127. 11 The Republic of France likewise takes the following position in this case: “THE HAGUE CONVENTION IS THE EXCLUSIVE MEANS OF DISCOVERY IN TRANSNATIONAL LITIGATION AMONG THE CONVENTION’S SIGNATORIES UNLESS THE SOVEREIGN ON WHOSE TERRITORY DISCOVERY IS TO OCCUR CHOOSES OTHERWISE.” Brief for Republic of France as Amicus Curiae 4. 12 See S. Exec. Doc. A, p. V; Pub. L. 88-244, 77 Stat. 775 (1963). 530 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. United States.13 In 1965 the Hague Conference adopted a Convention on the Service Abroad of Judicial and Extrajudicial Documents in Civil or Commercial Matters (Service Convention), 20 U. S. T. 361, T. I. A. S. No. 6638, to which the Senate gave its advice and consent in 1967. The favorable response to the Service Convention, coupled with the longstanding interest of American lawyers in improving procedures for obtaining evidence abroad, motivated the United States to take the initiative in proposing that an evidence convention be adopted. Statement of Carl F. Salans, Deputy Legal Adviser, Department of State, Convention on Taking of Evidence Abroad, S. Exec. Rep. No. 92-25, p. 3 (1972). The Conference organized a special commission to prepare the draft convention, and the draft was approved without a dissenting vote on October 26, 1968. S. Exec. Doc. A, p. v. It was signed on behalf of the United States in 1970 and ratified by a unanimous vote of the Senate in 1972.14 The Convention’s purpose was to establish a system for obtaining evidence located abroad that would be “tolerable” to the state executing the request and would produce evidence “utilizable” in the requesting state. Amram, Explanatory Report on the Convention on the Taking of Evi- 13 As the Rapporteur for the session of the Hague Conference which produced the Hague Evidence Convention stated: “In 1964 Rule 28(b) of the Federal Rules of Civil Procedure and 28 U. S. C. §§ 1781 and 1782 were amended to offer to foreign countries and litigants, without a requirement of reciprocity, wide judicial assistance on a unilateral basis for the obtaining of evidence in the United States. The amendments named the Department of State as a conduit for the receipt and transmission of letters of request. They authorized the use in the federal courts of evidence taken abroad in civil law countries, even if its form did not comply with the conventional formalities of our normal rules of evidence. No country in the world has a more open and enlightened policy.” Amram, The Proposed Convention on the Taking of Evidence Abroad, 55 A. B. A. J. 651 (1969). 14118 Cong. Rec. 20623 (1972). SOCIÉTÉ NAT. IND. AÉRO. v. U. S. DIST. COURT 531 522 Opinion of the Court dence Abroad in Civil or Commercial Matters, in S. Exec. Doc. A, p. 11. In his letter of transmittal recommending ratification of the Convention, the President noted that it was “supported by such national legal organizations as the American Bar Association, the Judicial Conference of the United States, the National Conference of Commissions on Uniform State Laws, and by a number of State, local, and specialized bar associations.” S. Exec. Doc. A, p. in. There is no evidence of any opposition to the Convention in any of those organizations. The Convention was fairly summarized in the Secretary of State’s letter of submittal to the President: “The willingness of the Conference to proceed promptly with work on the evidence convention is perhaps attributable in large measure to the difficulties encountered by courts and lawyers in obtaining evidence abroad from countries with markedly different legal systems. Some countries have insisted on the exclusive use of the complicated, dilatory and expensive system of letters rogatory or letters of request. Other countries have refused adequate judicial assistance because of the absence of a treaty or convention regulating the matter. The substantial increase in litigation with foreign aspects arising, in part, from the unparalleled expansion of international trade and travel in recent decades had intensified the need for an effective international agreement to set up a model system to bridge differences between the common law and civil law approaches to the taking of evidence abroad. “Civil law countries tend to concentrate on commissions rogatoires, while common law countries take testimony on notice, by stipulation and through commissions to consuls or commissioners. Letters of request for judicial assistance from courts abroad in securing needed evidence have been the exception, rather than the rule. The civil law technique results normally in a résumé of 532 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. the evidence, prepared by the executing judge and signed by the witness, while the common law technique results normally in a verbatim transcript of the witness’s testimony certified by the reporter. “Failure by either the requesting state or the state of execution fully to take into account the differences of approach to the taking of evidence abroad under the two systems and the absence of agreed standards applicable to letters of request have frequently caused difficulties for courts and litigants. To minimize such difficulties in the future, the enclosed convention, which consists of a preamble and forty-two articles, is designed to: “1. Make the employment of letters of request a principal means of obtaining evidence abroad; “2. Improve the means of securing evidence abroad by increasing the powers of consuls and by introducing in the civil law world, on a limited basis, the concept of the commissioner; “3. Provide means for securing evidence in the form needed by the court where the action is pending; and “4. Preserve all more favorable and less restrictive practices arising from internal law, internal rules of procedure and bilateral or multilateral conventions. “What the convention does is to provide a set of minimum standards with which contracting states agree to comply. Further, through articles 27, 28 and 32, it provides a flexible framework within which any future liberalizing changes in policy and tradition in any country with respect to international judicial cooperation may be translated into effective change in international procedures. At the same time it recognizes and preserves procedures of every country which now or hereafter may provide international cooperation in the taking of evidence on more liberal and less restrictive bases, whether this is effected by supplementary agreements or by municipal law and practice.” Id., p. vi. SOCIETY NAT. IND. AERO. v. U. S. DIST. COURT 533 522 Opinion of the Court III In arguing their entitlement to a protective order, petitioners correctly assert that both the discovery rules set forth in the Federal Rules of Civil Procedure and the Hague Convention are the law of the United States. Brief for Petitioners 31. This observation, however, does not dispose of the question before us; we must analyze the interaction between these two bodies of federal law. Initially, we note that at least four different interpretations of the relationship between the federal discovery rules and the Hague Convention are possible. Two of these interpretations assume that the Hague Convention by its terms dictates the extent to which it supplants normal discovery rules. First, the Hague Convention might be read as requiring its use to the exclusion of any other discovery procedures whenever evidence located abroad is sought for use in an American court. Second, the Hague Convention might be interpreted to require first, but not exclusive, use of its procedures. Two other interpretations assume that international comity, rather than the obligations created by the treaty, should guide judicial resort to the Hague Convention. Third, then, the Convention might be viewed as establishing a supplemental set of discovery procedures, strictly optional under treaty law, to which concerns of comity nevertheless require first resort by American courts in all cases. Fourth, the treaty may be viewed as an undertaking among sovereigns to facilitate discovery to which an American court should resort when it deems that course of action appropriate, after considering the situations of the parties before it as well as the interests of the concerned foreign state. In interpreting an international treaty, we are mindful that it is “in the nature of a contract between nations,” Trans World Airlines, Inc. v. Franklin Mint Corp., 466 U. S. 243, 253 (1984), to which “[g]eneral rules of construction apply.” Id., at 262. See Ware v. Hylton, 3 Dall. 199, 240-241 (1796) 534 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. (opinion of Chase, J.). We therefore begin “with the text of the treaty and the context in which the written words are used.” Air France n. Saks, 470 U. S. 392, 397 (1985). The treaty’s history, “‘the negotiations, and the practical construction adopted by the parties’ ” may also be relevant. Id., at 396 (quoting Choctaw Nation of Indians v. United States, 318 U. S. 423, 431-432 (1943)). We reject the first two of the possible interpretations as inconsistent with the language and negotiating history of the Hague Convention. The preamble of the Convention specifies its purpose “to facilitate the transmission and execution of Letters of Request” and to “improve mutual judicial cooperation in civil or commercial matters.” 23 U. S. T., at 2557, T. I. A. S. No. 7444. The preamble does not speak in mandatory terms which would purport to describe the procedures for all permissible transnational discovery and exclude all other existing practices.15 The text of the Evidence Convention itself does not modify the law of any contracting state, require any contracting state to use the Convention procedures, either in requesting evidence or in responding to such requests, or compel any contracting state to change its own evidence-gathering procedures.16 15 The Hague Conference on Private International Law’s omission of mandatory language in the preamble is particularly significant in light of the same body’s use of mandatory language in the preamble to the Hague Service Convention, 20 U. S. T. 361, T. I. A. S. No. 6638. Article 1 of the Service Convention provides: “The present Convention shall apply in all cases, in civil or commercial matters, where there is occasion to transmit a judicial or extrajudicial document for service abroad.” Id., at 362, T. I. A. S. No. 6638. As noted, supra, at 530, the Service Convention was drafted before the Evidence Convention, and its language provided a model exclusivity provision that the drafters of the Evidence Convention could easily have followed had they been so inclined. Given this background, the drafters’ election to use permissive language instead is strong evidence of their intent. 16 At the time the Convention was drafted, Federal Rule of Civil Procedure 28(b) clearly authorized the taking of evidence on notice either in ac- SOCIÉTÉ NAT. IND. AÉRO. v. U. S. DIST. COURT 535 522 Opinion of the Court The Convention contains three chapters. Chapter I, entitled “Letters of Requests,” and chapter II, entitled “Taking of Evidence by Diplomatic Officers, Consular Agents and Commissioners,” both use permissive rather than mandatory language. Thus, Article 1 provides that a judicial authority in one contracting state “may” forward a letter of request to the competent authority in another contracting state for the purpose of obtaining evidence.17 Similarly, Articles 15, 16, and 17 provide that diplomatic officers, consular agents, and commissioners “may . . . without compulsion,” take evidence under certain conditions.18 The absence of any command that a contracting state must use Convention procedures when they are not needed is conspicuous.19 cordance with the laws of the foreign country or in pursuance of the law of the United States. 17 The first paragraph of Article 1 reads as follows: “In civil or commercial matters a judicial authority of a Contracting State may, in accordance with the provisions of the law of that State, request the competent authority of another Contracting State, by means of a Letter of Request, to obtain evidence, or to perform some other judicial act.” 23 U. S. T., at 2557, T. I. A. S. 7444. 18 Thus, Article 17 provides: “In a civil or commercial matter, a person duly appointed as a commissioner for the purpose may, without compulsion, take evidence in the territory of a Contracting State in aid of proceedings commenced in the courts of another Contracting State if— “(a) a competent authority designated by the State where the evidence is to be taken has given its permission either generally or in the particular case; and “(b) he complies with the conditions which the competent authority has specified in the permission. “A Contracting State may declare that evidence may be taken under this Article without its prior permission.” Id., at 2565, T. I. A. S. 7444. 19 Our conclusion is confirmed by the position of the Executive Branch and the Securities and Exchange Commission, which interpret the “language, history, and purposes” of the Hague Convention as indicating “that it was not intended to prescribe the exclusive means by which American plaintiffs might obtain foreign evidence.” Brief for United States as Amicus Curiae 9 (citation omitted). “[T]he meaning attributed to treaty pro 536 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. Two of the Articles in chapter III, entitled “General Clauses,” buttress our conclusion that the Convention was intended as a permissive supplement, not a pre-emptive replacement, for other means of obtaining evidence located abroad.20 Article 23 expressly authorizes a contracting state to declare that it will not execute any letter of request in aid of pretrial discovery of documents in a common-law country.21 Surely, if the Convention had been intended to replace completely the broad discovery powers that the common-law courts in the United States previously exercised over foreign litigants subject to their jurisdiction, it would have been most anomalous for the common-law contracting parties to agree to visions by the Government agencies charged with their negotiation and enforcement is entitled to great weight.” Sumitomo Shoji America, Inc. v. Avagliano, 457 U. S. 176, 184-185 (1982); see also O’Connor v. United States, 479 U. S. 27, 33 (1986). As a member of the United States delegation to the Hague Conference concluded: “[The Convention] makes no major changes in United States procedure and requires no major changes in United States legislation or rules. On the other front, it will give the United States courts and litigants abroad enormous aid by providing an international agreement for the taking of testimony, the absence of which has created barriers to our courts and litigants.” Amram, Explanatory Report on the Convention on the Taking of Evidence Abroad in Civil or Commercial Matters, S. Exec. Doc. A, at pp. 1, 3. 20 In addition to the Eighth Circuit, other Courts of Appeals and the West Virginia Supreme Court have held that the Convention cannot be viewed as the exclusive means of securing discovery transnationally. See Société Nationale Industrielle Aérospatiale v. United States District Court, 788 F. 2d 1408, 1410 (CA9 1986); In re Messerschmitt Bolkow Blohm GmbH, 757 F. 2d 729, 731 (CA5 1985), cert, vacated, 476 U. S. 1168 (1986); In re Anschuetz & Co., GmbH, 754 F. 2d 602, 606-615, and n. 7 (CA5 1985), cert, pending, No. 85-98; Gebr. Eickhoff Maschinenfabrik und Eisengieberei mbH v. Starcher,-------W. Va.-------,----, 328 S. E. 2d 492, 497-501 (1985). 21 Article 23 provides: “A Contracting State may at the time of signature, ratification or accession, declare that it will not execute Letters of Request issued for the purpose of obtaining pre-trial discovery of documents as known in Common Law countries.” 23 U. S. T., at 2568, T. I. A. S. 7444. SOCIÉTÉ NAT. IND. AÉRO. v. U. S. DIST. COURT 537 522 Opinion of the Court Article 23, which enables a contracting party to revoke its consent to the treaty’s procedures for pretrial discovery.22 In the absence of explicit textual support, we are unable to accept the hypothesis that the common-law contracting states abjured recourse to all pre-existing discovery procedures at the same time that they accepted the possibility that a contracting party could unilaterally abrogate even the Convention’s procedures.23 Moreover, Article 27 plainly states that “Thirteen of the seventeen signatory states have made declarations under Article 23 of the Convention that restrict pretrial discovery of documents. See 7 Martindale-Hubbell Law Directory (pt. VII) 15-19 (1986). ““The great object of an international agreement is to define the common ground between sovereign nations. Given the gulfs of language, culture, and values that separate nations, it is essential in international agreements for the parties to make explicit their common ground on the most rudimentary of matters.” Trans World Airlines, Inc. v. Franklin Mint Corp., 466 U. S. 243, 262 (1984) (Stevens, J., dissenting). The utter absence in the Hague Convention of an exclusivity provision has an obvious explanation: The contracting states did not agree that its procedures were to be exclusive. The words of the treaty delineate the extent of their agreement; without prejudice to their existing rights and practices, they bound themselves to comply with any request for judicial assistance that did comply with the treaty’s procedures. See Carter, Obtaining Foreign Discovery and Evidence for Use in Litigation in the United States: Existing Rules and Procedures, 13 Int’l Law. 5, 11, n. 14 (1979) (common-law nations and civil-law jurisdictions have separate traditions of bilateral judicial cooperation; the Evidence Convention “attempts to bridge” the two traditions.) The separate opinion reasons that the Convention procedures are not optional because unless other signatory states “had expected the Convention to provide the normal channels for discovery, [they] would have had no incentive to agree to its terms.” Post, at 550. We find the treaty language that the parties have agreed upon and ratified a surer indication of their intentions than the separate opinion’s hypothesis about the expectations of the parties. Both comity and concern for the separation of powers counsel the utmost restraint in attributing motives to sovereign states which have bargained as equals. Indeed, Justice Blackmun notes that “the Convention represents a political determination—one that, consistent with the principle of separation of powers, courts should not attempt to second guess.” Post, at 552. Moreover, it is important to remember that the 538 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. the Convention does not prevent a contracting state from using more liberal methods of rendering evidence than those authorized by the Convention.24 Thus, the text of the Evidence Convention, as well as the history of its proposal and ratification by the United States, unambiguously supports the conclusion that it was intended to establish optional procedures that would facilitate the taking of evidence abroad. See Amram, The Proposed Convention on the Taking of Evidence Abroad, 55 A. B. A. J. 651, 655 (1969); President’s Letter of Transmittal, Sen. Exec. Doc. A, p. in. evidence-gathering procedures implemented by the Convention would still provide benefits to the signatory states even if the United States were not a party. 24 Article 27 provides: “The provisions of the present Convention shall not prevent a Contracting State from— “(a) declaring that Letters of Request may be transmitted to its judicial authorities through channels other than those provided for in Article 2; “(b) permitting, by internal law or practice, any act provided for in this Convention to be performed upon less restrictive conditions; “(c) permitting, by internal law or practice, methods of taking evidence other than those provided for in this Convention.” 23 U. S. T., at 2569, T. I. A. S. 7444. Thus, for example, the United Kingdom permits foreign litigants, by a letter of request, to “apply directly to the appropriate courts in the United Kingdom for judicial assistance” or to seek information directly from parties in the United Kingdom “if, as in this case, the court of origin exercises jurisdiction consistent with accepted norms of international law.” Brief for the Government of the United Kingdom and Northern Ireland as Amicus Curiae 6 (footnote omitted). On its face, the term “Contracting State” comprehends both the requesting state and the receiving state. Even if Article 27 is read to apply only to receiving states, see, e. g., Gebr. Eick-hoff Maschinenfabrik und Eisengieberei mbH v. Starcher,---W. Va., at ----, 328 S. E. 2d, at 499-500, n. 11 (rejecting argument that Article 27 authorizes more liberal discovery procedures by requesting as well as executing states), the treaty’s internal failure to authorize more liberal procedures for obtaining evidence would carry no pre-emptive meaning. We are unpersuaded that Article 27 supports a “negative inference” that would curtail the pre-existing authority of a state to obtain evidence in accord with its normal procedures. SOCIÉTÉ NAT. IND. AÉRO. v. U. S. DIST. COURT 539 522 Opinion of the Court An interpretation of the Hague Convention as the exclusive means for obtaining evidence located abroad would effectively subject every American court hearing a case involving a national of a contracting state to the internal laws of that state. Interrogatories and document requests are staples of international commercial litigation, no less than of other suits, yet a rule of exclusivity would subordinate the court’s supervision of even the most routine of these pretrial proceedings to the actions or, equally, to the inactions of foreign judicial authorities. As the Court of Appeals for the Fifth Circuit observed in In re Anschuetz & Co., GmbH, 754 F. 2d 602, 612 (1985), cert, pending, No. 85-98: “It seems patently obvious that if the Convention were interpreted as preempting interrogatories and document requests, the Convention would really be much more than an agreement on taking evidence abroad. Instead, the Convention would amount to a major regulation of the overall conduct of litigation between nationals of different signatory states, raising a significant possibility of very serious interference with the jurisdiction of United States courts. “While it is conceivable that the United States could enter into a treaty giving other signatories control over litigation instituted and pursued in American courts, a treaty intended to bring about such a curtailment of the rights given to all litigants by the federal rules would surely state its intention clearly and precisely identify crucial terms.” The Hague Convention, however, contains no such plain statement of a pre-emptive intent. We conclude accordingly that the Hague Convention did not deprive the District Court of the jurisdiction it otherwise possessed to order a foreign 540 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. national party before it to produce evidence physically located within a signatory nation.25 IV While the Hague Convention does not divest the District Court of jurisdiction to order discovery under the Federal Rules of Civil Procedure, the optional character of the Convention procedures sheds light on one aspect of the Court of Appeals’ opinion that we consider erroneous. That court concluded that the Convention simply “does not apply” to discovery sought from a foreign litigant that is subject to the jurisdiction of an American court. 782 F. 2d, at 124. Plaintiffs argue that this conclusion is supported by two considerations. First, the Federal Rules of Civil Procedure provide 28 The opposite conclusion of exclusivity would create three unacceptable asymmetries. First, within any lawsuit between a national of the United States and a national of another contracting party, the foreign party could obtain discovery under the Federal Rules of Civil Procedure, while the domestic party would be required to resort first to the procedures of the Hague Convention. This imbalance would run counter to the fundamental maxim of discovery that “[m]utual knowledge of all the relevant facts gathered by both parties is essential to proper litigation.” Hickman v. Taylor, 329 U. S. 495, 507 (1947). Second, a rule of exclusivity would enable a company which is a citizen of another contracting state, to compete with a domestic company on uneven terms, since the foreign company would be subject to less extensive discovery procedures in the event that both companies were sued in an American court. Petitioners made a voluntary decision to market their products in the United States. They are entitled to compete on equal terms with other companies operating in this market. But since the District Court unquestionably has personal jurisdiction over petitioners, they are subject to the same legal constraints, including the burdens associated with American judicial procedures, as their American competitors. A general rule according foreign nationals a preferred position in pretrial proceedings in our courts would conflict with the principle of equal opportunity that governs the market they elected to enter. Third, since a rule of first use of the Hague Convention would apply to cases in which a foreign party is a national of a contracting state, but not to cases in which a foreign party is a national of any other foreign state, the rule would confer an unwarranted advantage on some domestic litigants over others similarly situated. SOCIÉTÉ NAT. IND. AÉRO. v. U. S. DIST. COURT 541 522 Opinion of the Court ample means for obtaining discovery from parties who are subject to the court’s jurisdiction, while before the Convention was ratified it was often extremely difficult, if not impossible, to obtain evidence from nonparty witnesses abroad. Plaintiffs contend that it is appropriate to construe the Convention as applying only in the area in which improvement was badly needed. Second, when a litigant is subject to the jurisdiction of the district court, arguably the evidence it is required to produce is not “abroad” within the meaning of the Convention, even though it is in fact located in a foreign country at the time of the discovery request and even though it will have to be gathered or otherwise prepared abroad. See In re Anschuetz & Co., GmbH, 754 F. 2d, at 611; In re Messerschmitt Bolkow Blohm GmbH, 757 F. 2d 729, 731 (CA5 1985), cert, vacated, 476 U. S. 1168 (1986); Daimler-Benz Aktiengesellschaft v. United States District Court, 805 F. 2d 340, 341-342 (CAIO 1986). Nevertheless, the text of the Convention draws no distinction between evidence obtained from third parties and that obtained from the litigants themselves; nor does it purport to draw any sharp line between evidence that is “abroad” and evidence that is within the control of a party subject to the jurisdiction of the requesting court. Thus, it appears clear to us that the optional Convention procedures are available whenever they will facilitate the gathering of evidence by the means authorized in the Convention. Although these procedures are not mandatory, the Hague Convention does “apply” to the production of evidence in a litigant’s possession in the sense that it is one method of seeking evidence that a court may elect to employ. See Briefs of Amici Curiae for the United States and the SEC 9-10, the Federal Republic of Germany 5-6, the Republic of France 8-12, and the Government of the United Kingdom and Northern Ireland 8. V Petitioners contend that even if the Hague Convention’s procedures are not mandatory, this Court should adopt a rule 542 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. requiring that American litigants first resort to those procedures before initiating any discovery pursuant to the normal methods of the Federal Rules of Civil Procedure. See, e. g., Laker Airways, Ltd. v. Pan American World Airways, 103 F. R. D. 42 (DC 1984); Philadelphia Gear Corp. n. American Pfauter Corp., 100 F. R. D. 58 (ED Pa. 1983). The Court of Appeals rejected this argument because it was convinced that an American court’s order ultimately requiring discovery that a foreign court had refused under Convention procedures would constitute “the greatest insult” to the sovereignty of that tribunal. 782 F. 2d, at 125-126. We disagree with the Court of Appeals’ view. It is well known that the scope of American discovery is often significantly broader than is permitted in other jurisdictions, and we are satisfied that foreign tribunals will recognize that the final decision on the evidence to be used in litigation conducted in American courts must be made by those courts. We therefore do not believe that an American court should refuse to make use of Convention procedures because of a concern that it may ultimately find it necessary to order the production of evidence that a foreign tribunal permitted a party to withhold. Nevertheless, we cannot accept petitioners’ invitation to announce a new rule of law that would require first resort to Convention procedures whenever discovery is sought from a foreign litigant. Assuming, without deciding, that we have the lawmaking power to do so, we are convinced that such a general rule would be unwise. In many situations the Letter of Request procedure authorized by the Convention would be unduly time consuming and expensive, as well as less certain to produce needed evidence than direct use of the Federal Rules.26 A rule of first resort in all cases would 26 We observe, however, that in other instances a litigant’s first use of the Hague Convention procedures can be expected to yield more evidence abroad more promptly than use of the normal procedures governing pretrial civil discovery. In those instances, the calculations of the litigant will naturally lead to a first-use strategy. SOCIÉTÉ NAT. IND. AÉRO. v. U. S. DIST. COURT 543 522 Opinion of the Court therefore be inconsistent with the overriding interest in the “just, speedy, and inexpensive determination” of litigation in our courts. See Fed. Rule Civ. Proc. 1. Petitioners argue that a rule of first resort is necessary to accord respect to the sovereignty of states in which evidence is located. It is true that the process of obtaining evidence in a civil-law jurisdiction is normally conducted by a judicial officer rather than by private attorneys. Petitioners contend that if performed on French soil, for example, by an unauthorized person, such evidence-gathering might violate the “judicial sovereignty” of the host nation. Because it is only through the Convention that civil-law nations have given their consent to evidence-gathering activities within their borders, petitioners argue, we have a duty to employ those procedures whenever they are available. Brief for Petitioners 27-28. We find that argument unpersuasive. If such a duty were to be inferred from the adoption of the Convention itself, we believe it would have been described in the text of that document. Moreover, the concept of international comity27 requires in this context a more particularized analysis of 27 Comity refers to the spirit of cooperation in which a domestic tribunal approaches the resolution of cases touching the laws and interests of other sovereign states. This Court referred to the doctrine of comity among nations in Emory v. Grenough, 3 Dall. 369, 370, n. (1797) (dismissing appeal from judgment for failure to plead diversity of citizenship, but setting forth an extract from a treatise by Ulrich Huber (1636-1694), a Dutch jurist): “ ‘By the courtesy of nations, whatever laws are carried into execution, within the limits of any government, are considered as having the same effect every where, so far as they do not occasion a prejudice to the rights of the other governments, or their citizens. “ ‘[N]othing would be more inconvenient in the promiscuous intercourse and practice of mankind, than that what was valid by the laws of one place, should be rendered of no effect elsewhere, by a diversity of law. . . Ibid, (quoting 2 U. Huber, Praelectiones Juris Romani et hodiemi, bk. 1, fit. 3, pp. 26-31 (C. Thomas, L. Menke, & G. Gebauer eds. 1725)). See also Hilton v. Guyot, 159 U. S. 113, 163-164 (1895): “ ‘Comity,’ in the legal sense, is neither a matter of absolute obligation, on the one hand, nor of mere courtesy and good will, upon the other. But 544 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. the respective interests of the foreign nation and the requesting nation than petitioners’ proposed general rule would generate.28 We therefore decline to hold as a blanket matter that comity requires resort to Hague Evidence Convention procedures without prior scrutiny in each case of the particular facts, sovereign interests, and likelihood that resort to those procedures will prove effective.29 it is the recognition which one nation allows within its territory to the legislative, executive or judicial acts of another nation, having due regard both to international duty and convenience, and to the rights of its own citizens or of other persons who are under the protection of its laws.” 28 The nature of the concerns that guide a comity analysis is suggested by the Restatement of Foreign Relations Law of the United States (Revised) §437(l)(c) (Tent. Draft No. 7, 1986) (approved May 14, 1986) (Restatement). While we recognize that § 437 of the Restatement may not represent a consensus of international views on the scope of the district court’s power to order foreign discovery in the face of objections by foreign states, these factors are relevant to any comity analysis: “(1) the importance to the . . . litigation of the documents or other information requested; “(2) the degree of specificity of the request; “(3) whether the information originated in the United States; “(4) the availability of alternative means of securing the information; and “(5) the extent to which noncompliance with the request would undermine important interests of the United States, or compliance with the request would undermine important interests of the state where the information is located.” Ibid. 29 The French “blocking statute,” n. 6, supra, does not alter our conclusion. It is well settled that such statutes do not deprive an American court of the power to order a party subject to its jurisdiction to produce evidence even though the act of production may violate that statute. See Société Internationale Pour Participations Industrielles et Commerciales, S. A. v. Rogers, 357 U. S. 197, 204-206 (1958). Nor can the enactment of such a statute by a foreign nation require American courts to engraft a rule of first resort onto the Hague Convention, or otherwise to provide the nationals of such a country with a preferred status in our courts. It is clear that American courts are not required to adhere blindly to the directives of such a statute. Indeed, the language of the statute, if taken literally, would appear to represent an extraordinary exercise of legislative jurisdiction by the Republic of France over a United States district judge, forbidding him or her to order any discovery from a party of French nationality, even simple requests for admissions or interrogatories that the party SOCIÉTÉ NAT. IND. AÉRO. v. U. S. DIST. COURT 545 522 Opinion of the Court Some discovery procedures are much more “intrusive” than others. In this case, for example, an interrogatory asking petitioners to identify the pilots who flew flight tests in the Rallye before it was certified for flight by the Federal Aviation Administration, or a request to admit that petitioners authorized certain advertising in a particular magazine, is certainly less intrusive than a request to produce all of the “design specifications, line drawings and engineering plans and all engineering change orders and plans and all drawings concerning the leading edge slats for the Rallye type aircraft manufactured by the Defendants.” App. 29. Even if a court might be persuaded that a particular document request was too burdensome or too “intrusive” to be granted in full, with or without an appropriate protective order, it might well refuse to insist upon the use of Convention procedures could respond to on the basis of personal knowledge. It would be particularly incongruous to recognize such a preference for corporations that are wholly owned by the enacting nation. Extraterritorial assertions of jurisdiction are not one-sided. While the District Court’s discovery orders arguably have some impact in France, the French blocking statute asserts similar authority over acts to take place in this country. The lesson of comity is that neither the discovery order nor the blocking statute can have the same omnipresent effect that it would have in a world of only one sovereign. The blocking statute thus is relevant to the court’s particularized comity analysis only to the extent that its terms and its enforcement identify the nature of the sovereign interests in nondisclosure of specific kinds of material. The American Law Institute has summarized this interplay of blocking statutes and discovery orders: “[W]hen a state has jurisdiction to prescribe and its courts have jurisdiction to adjudicate, adjudication should (subject to generally applicable rules of evidence) take place on the basis of the best information available. . . . [Blocking] statutes that frustrate this goal need not be given the same deference by courts of the United States as substantive rules of law at variance with the law of the United States.” See Restatement §437, Reporter’s Note 5, pp. 41, 42. “On the other hand, the degree of friction created by discovery requests . . . and the differing perceptions of the acceptability of American-style discovery under national and international law, suggest some efforts to moderate the application abroad of U. S. procedural techniques, consistent with the overall principle of reasonableness in the exercise of jurisdiction.” Id., at 42. 546 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. before requiring responses to simple interrogatories or requests for admissions. The exact line between reasonableness and unreasonableness in each case must be drawn by the trial court, based on its knowledge of the case and of the claims and interests of the parties and the governments whose statutes and policies they invoke. American courts, in supervising pretrial proceedings, should exercise special vigilance to protect foreign litigants from the danger that unnecessary, or unduly burdensome, discovery may place them in a disadvantageous position. Judicial supervision of discovery should always seek to minimize its costs and inconvenience and to prevent improper uses of discovery requests. When it is necessary to seek evidence abroad, however, the district court must supervise pretrial proceedings particularly closely to prevent discovery abuses. For example, the additional cost of transportation of documents or witnesses to or from foreign locations may increase the danger that discovery may be sought for the improper purpose of motivating settlement, rather than finding relevant and probative evidence. Objections to “abusive” discovery that foreign litigants advance should therefore receive the most careful consideration. In addition, we have long recognized the demands of comity in suits involving foreign states, either as parties or as sovereigns with a coordinate interest in the litigation. See Hilton v. Guyot, 159 U. S. 113 (1895). American courts should therefore take care to demonstrate due respect for any special problem confronted by the foreign litigant on account of its nationality or the location of its operations, and for any sovereign interest expressed by a foreign state. We do not articulate specific rules to guide this delicate task of adjudication.30 30 Under the Hague Convention, a letter of request must specify “the evidence to be obtained or other judicial act to be performed,” Art. 3, and must be in the language of the executing authority or be accompanied by a translation into that language. Art. 4, 23 U. S. T., at 2558-2559, T. I. A. S. 7444. Although the discovery request must be specific, the SOCIÉTÉ NAT. IND. AÉRO. v. U. S. DIST. COURT 547 522 Opinion of Blackmun, J. VI In the case before us, the Magistrate and the Court of Appeals correctly refused to grant the broad protective order that petitioners requested. The Court of Appeals erred, however, in stating that the Evidence Convention does not apply to the pending discovery demands. This holding may be read as indicating that the Convention procedures are not even an option that is open to the District Court. It must be recalled, however, that the Convention’s specification of duties in executing states creates corresponding rights in requesting states; holding that the Convention does not apply in this situation would deprive domestic litigants of access to evidence through treaty procedures to which the contracting states have assented. Moreover, such a rule would deny the foreign litigant a full and fair opportunity to demonstrate appropriate reasons for employing Convention procedures in the first instance, for some aspects of the discovery process. Accordingly, the judgment of the Court of Appeals is vacated, and the case is remanded for further proceedings consistent with this opinion. It is so ordered. Justice Blackmun, with whom Justice Brennan, Justice Marshall, and Justice O’Connor join, concurring in part and dissenting in part. Some might well regard the Court’s decision in this case as an affront to the nations that have joined the United States in ratifying the Hague Convention on the Taking of Evidence party seeking discovery may find it difficult or impossible to determine in advance what evidence is within the control of the party urging resort to the Convention and which parts of that evidence may qualify for international judicial assistance under the Convention. This information, however, is presumably within the control of the producing party from which discovery is sought. The district court may therefore require, in appropriate situations, that this party bear the burden of providing translations and detailed descriptions of relevant documents that are needed to assure prompt and complete production pursuant to the terms of the Convention. 548 OCTOBER TERM, 1986 Opinion of Blackmun, J. 482 U. S. Abroad in Civil or Commercial Matters, opened for signature, Mar. 18, 1970, 23 U. S. T. 2555, T. I. A. S. No. 7444. The Court ignores the importance of the Convention by relegating it to an “optional” status, without acknowledging the significant achievement in accommodating divergent interests that the Convention represents. Experience to date indicates that there is a large risk that the case-by-case comity analysis now to be permitted by the Court will be performed inadequately and that the somewhat unfamiliar procedures of the Convention will be invoked infrequently. I fear the Court’s decision means that courts will resort unnecessarily to issuing discovery orders under the Federal Rules of Civil Procedure in a raw exercise of their jurisdictional power to the detriment of the United States’ national and international interests. The Court’s view of this country’s international obligations is particularly unfortunate in a world in which regular commercial and legal channels loom ever more crucial. I do agree with the Court’s repudiation of the positions at both extremes of the spectrum with regard to the use of the Convention. Its rejection of the view that the Convention is not “applicable” at all to this case is surely correct: the Convention clearly applies to litigants as well as to third parties, and to requests for evidence located abroad, no matter where that evidence is actually “produced.” The Court also correctly rejects the far opposite position that the Convention provides the exclusive means for discovery involving signatory countries. I dissent, however, because I cannot endorse the Court’s case-by-case inquiry for determining whether to use Convention procedures and its failure to provide lower courts with any meaningful guidance for carrying out that inquiry. In my view, the Convention provides effective discovery procedures that largely eliminate the conflicts between United States and foreign law on evidence gathering. I therefore would apply a general presumption that, in most cases, courts should resort first to the Con ven- SOCIÉTÉ NAT. IND. AÉRO. v. U. S. DIST. COURT 549 522 Opinion of Blackmun, J. tion procedures.1 An individualized analysis of the circumstances of a particular case is appropriate only when it appears that it would be futile to employ the Convention or when its procedures prove to be unhelpful. I Even though the Convention does not expressly require discovery of materials in foreign countries to proceed exclusively according to its procedures, it cannot be viewed as merely advisory. The Convention was drafted at the request and with the enthusiastic participation of the United States, which sought to broaden the techniques available for the taking of evidence abroad. The differences between discovery practices in the United States and those in other countries are significant, and “[n]o aspect of the extension of the American legal system beyond the territorial frontier of the United States has given rise to so much friction as the request for documents associated with investigation and litigation in the United States.” Restatement of Foreign Relations Law of the United States (Revised) §437, Reporters’ Note 1, p. 35 (Tent. Draft No. 7, Apr. 10, 1986). Of par 1 Many courts that have examined the issue have adopted a rule of first resort to the Convention. See, e. g., Philadelphia Gear Corp. v. American Pfauter Corp. 100 F. R. D. 58, 61 (ED Pa. 1983) (“avenue of first resort for plaintiff [is] the Hague Convention”); Gebr. Eickhoff Maschinen-fabrik und Eisengieberei mbH v. Starcher,---W. Va. -----,----, 328 S. E. 2d 492, 504-506 (1985) (“principle of international comity dictates first resort to [Convention] procedures”); Vincent v. Ateliers de la Moto-bécane, S. A., 193 N. J. Super. 716, 723, 475 A. 2d 686, 690 (App. Div. 1984) (litigant should first attempt to comply with Convention); Th. Goldschmidt A. G. v. Smith, 676 S. W. 2d 443, 445 (Tex. App. 1984) (Convention procedures not mandatory but are “avenue of first resort”); Pierburg GmbH & Co. KG v. Superior Court, 137 Cal. App. 3d 238, 247, 186 Cal. Rptr. 876, 882-883 (1982) (plaintiffs must attempt to comply with the Convention); Volkswagenwerk Aktiengesellschaft v. Superior Court, 123 Cal. App. 3d 840, 857-859, 176 Cal. Rptr. 874, 885-886 (1981) (“Hague Convention establishes not a fixed rule but rather a minimum measure of international cooperation”). 550 OCTOBER TERM, 1986 Opinion of Blackmun, J. 482 U. S. ticular import is the fact that discovery conducted by the parties, as is common in the United States, is alien to the legal systems of civil-law nations, which typically regard evidence gathering as a judicial function. The Convention furthers important United States interests by providing channels for discovery abroad that would not be available otherwise. In general, it establishes “methods to reconcile the differing legal philosophies of the Civil Law, Common Law and other systems with respect to the taking of evidence.” Rapport de la Commission spéciale, 4 Conférence de La Haye de droit international privé: Actes et documents de la Onzième session 55 (1970) (Actes et documents). It serves the interests of both requesting and receiving countries by advancing the following goals: “[T]he techniques for the taking of evidence must be ‘utilizable’ in the eyes of the State where the lawsuit is pending and must also be ‘tolerable’ in the eyes of the State where the evidence is to be taken.” Id., at 56. The Convention also serves the long-term interests of the United States in helping to further and to maintain the climate of cooperation and goodwill necessary to the functioning of the international legal and commercial systems. It is not at all satisfactory to view the Convention as nothing more than an optional supplement to the Federal Rules of Civil Procedure, useful as a means to “facilitate discovery” when a court “deems that course of action appropriate.” Ante, at 533. Unless they had expected the Convention to provide the normal channels for discovery, other parties to the Convention would have had no incentive to agree to its terms. The civil-law nations committed themselves to employ more effective procedures for gathering evidence within their borders, even to the extent of requiring some commonlaw practices alien to their systems. At the time of the Convention’s enactment, the liberal American policy, which allowed foreigners to collect evidence with ease in the United States, see ante at 529-530, and n. 13, was in place and, be- SOCIÉTÉ NAT. IND. AÉRO. v. U. S. DIST. COURT 551 522 Opinion of Blackmun, J. cause it was not conditioned on reciprocity, there was little likelihood that the policy would change as a result of treaty negotiations. As a result, the primary benefit the other signatory nations would have expected in return for their concessions was that the United States would respect their territorial sovereignty by using the Convention procedures.2 II By viewing the Convention as merely optional and leaving the decision whether to apply it to the court in each individual case, the majority ignores the policies established by the political branches when they negotiated and ratified the treaty. The result will be a duplicative analysis for which courts are not well designed. The discovery process usually concerns discrete interests that a court is well equipped to accommodate—the interests of the parties before the court coupled with the interest of the judicial system in resolving the conflict on the basis of the best available information. When a lawsuit requires discovery of materials located in a foreign nation, however, foreign legal systems and foreign interests 2 Article 27 of the Convention, see ante, at 538, n. 24, is not to the contrary. The only logical interpretation of this Article is that a state receiving a discovery request may permit less restrictive procedures than those designated in the Convention. The majority finds plausible a reading that authorizes both a requesting and a receiving state to use methods outside the Convention. Ibid. If this were the case, Article 27(c), which allows a state to permit methods of taking evidence that are not provided in the Convention, would make the rest of the Convention wholly superfluous. If a requesting state could dictate the methods for taking evidence in another state, there would be no need for the detailed procedures provided by the Convention. Moreover, the United States delegation’s explanatory report on the Convention describes Article 27 as “designed to preserve existing internal law and practice in a Contracting State which provides broader, more generous and less restrictive rules of international cooperation in the taking of evidence for the benefit of foreign courts and litigants.” S. Exec. Doc. A, 92d Cong., 2d Sess., 39 (1972). Article 27 authorizes the use of alternative methods for gathering evidence “if the internal law or practice of the State of execution so permits.” Id., at 39-40 (emphasis added). 552 OCTOBER TERM, 1986 Opinion of Blackmun, J. 482 U. S. are implicated as well. The presence of these interests creates a tension between the broad discretion our courts normally exercise in managing pretrial discovery and the discretion usually allotted to the Executive in foreign matters. It is the Executive that normally decides when a course of action is important enough to risk affronting a foreign nation or placing a strain on foreign commerce. It is the Executive, as well, that is best equipped to determine how to accommodate foreign interests along with our own.3 Unlike the courts, “diplomatic and executive channels are, by definition, designed to exchange, negotiate, and reconcile the problems which accompany the realization of national interests within the sphere of international association.” Laker Airways, Ltd. v. Sabena, Belgian World Airlines, 235 U. S. App. D. C. 207, 253, 731 F. 2d 909, 955 (1984). The Convention embodies the result of the best efforts of the Executive Branch, in negotiating the treaty, and the Legislative Branch, in ratifying it, to balance competing national interests. As such, the Convention represents a political determination-one that, consistent with the principle of separation of powers, courts should not attempt to second-guess. Not only is the question of foreign discovery more appropriately considered by the Executive and Congress, but in addition, courts are generally ill equipped to assume the role of balancing the interests of foreign nations with that of our own. Although transnational litigation is increasing, relatively few judges are experienced in the area and the procedures of foreign legal systems are often poorly understood. Wilkey, Transnational Adjudication: A View from the Bench, 18 Int’l Lawyer 541, 543 (1984); Ristau, Overview of Interna- 3 Our Government’s interests themselves are far more complicated than can be represented by the limited parties before a court. The United States is increasingly concerned, for example, with protecting sensitive technology for both economic and military reasons. It may not serve the country’s long-term interest to establish precedents that could allow foreign courts to compel production of the records of American corporations. SOCIÉTÉ NAT. IND. AÉRO. v. U. S. DIST. COURT 553 522 Opinion of Blackmun, J. tional Judicial Assistance, 18 Int’l Lawyer 525, 531 (1984). As this Court recently stated, it has “little competence in determining precisely when foreign nations will be offended by particular acts.” Container Corp. v. Franchise Tax Bd., 463 U. S. 159, 194 (1983). A pro-forum bias is likely to creep into the supposedly neutral balancing process4 and courts not surprisingly often will turn to the more familiar procedures established by their local rules. In addition, it simply is not reasonable to expect the Federal Government or the foreign state in which the discovery will take place to participate in every individual case in order to articulate the broader international and foreign interests that are relevant 4 One of the ways that a pro-forum bias has manifested itself is in United States courts’ preoccupation with their own power to issue discovery orders. All too often courts have regarded the Convention as some kind of threat to their jurisdiction and have rejected use of the treaty procedures. See, e. g., In re Anschuetz & Co., GmbH, 754 F. 2d 602, 606, 612 (CA5 1985), cert, pending, No. 85-98. It is well established that a court has the power to impose discovery under the Federal Rules of Civil Procedure when it has personal jurisdiction over the foreign party. Societe Internationale Pour Participations Industrielles et Commerciales, S. A. v. Rogers, 357 U. S. 197, 204-206 (1958). But once it is determined that the Convention does not provide the exclusive means for foreign discovery, jurisdictional power is not the issue. The relevant question, instead, becomes whether a court should forgo exercise of the full extent of its power to order discovery. The Convention, which is valid United States law, provides an answer to that question by establishing a strong policy in favor of self-restraint for the purpose of furthering United States interests and minimizing international disputes. There is also a tendency on the part of courts, perhaps unrecognized, to view a dispute from a local perspective. “[D]omestic courts do not sit as internationally constituted tribunals. . . . The courts of most developed countries follow international law only to the extent it is not overridden by national law. Thus courts inherently find it difficult neutrally to balance competing foreign interests. When there is any doubt, national interests will tend to be favored over foreign interests.” Laker Airways, Ltd. n. Sabena, Belgian World Airlines, 235 U. S. App. D. C. 207, 249, 731 F. 2d 909, 951 (1984) (footnotes omitted); see also In re Uranium Antitrust Litigation, 480 F. Supp. 1138, 1148 (ND Ill. 1979). 554 OCTOBER TERM, 1986 Opinion of Blackmun, J. 482 U. S. to the decision whether to use the Convention. Indeed, the opportunities for such participation are limited.5 Exacerbating these shortcomings is the limited appellate review of interlocutory discovery decisions,6 which prevents any effective case-by-case correction of erroneous discovery decisions. Ill The principle of comity leads to more definite rules than the ad hoc approach endorsed by the majority. The Court asserts that the concept of comity requires an individualized analysis of the interests present in each particular case before a court decides whether to apply the Convention. See ante, at 543-544. There is, however, nothing inherent in the comity principle that requires case-by-case analysis. The Court frequently has relied upon a comity analysis when it has adopted general rules to cover recurring situations in areas such as choice of forum,7 maritime law,8 and sovereign 6 The Department of State in general does not transmit diplomatic notes from foreign governments to state or federal trial courts. In addition, it adheres to a policy that it does not take positions regarding, or participate in, litigation between private parties, unless required to do so by applicable law. See Oxman, The Choice Between Direct Discovery and Other Means of Obtaining Evidence Abroad: The Impact of the Hague Evidence Convention, 37 U. Miami L. Rev. 733, 748, n. 39 (1983). 6 See Kerr v. United States District Court, 426 U. S. 394, 402-405 (1976); see also Boreri v. Fiat S. P. A., 763 F. 2d 17, 20 (CAI 1985) (refusing to review on interlocutory appeal District Court order involving extraterritorial discovery). 7 See, e. g., Mitsubishi Motors Corp. v. Soler Chrysler-Plymouth, Inc., 473 U. S. 614, 630 (1985); Scherk n. Alberto-Culver Co., 417 U. S. 506, 516-519 (1974); The Bremen v. Zapata Off-Shore Co., 407 U. S. 1, 12-14 (1972). 8See, e. g., Romero v. International Terminal Operating Co., 358 U. S. 354, 382-384 (1959); Lauritzen v. Larsen, 345 U. S. 571, 577-582 (1953); Berizzi Bros. Co. v. The Pesaro, 271 U. S. 562, 575 (1926); Wilden-hus’s Case, 120 U. S. 1, 12 (1887); The Belgenland, 114 U. S. 355, 363-364 (1885); The Scotia, 14 Wall. 170, 187-188 (1872); Brown v. Duchesne, 19 How. 183, 198 (1857); The Schooner Exchange v. McFaddon, 7 Cranch 116, 137 (1812). SOCIÉTÉ NAT. IND. AÉRO. v. U. S. DIST. COURT 555 522 Opinion of Blackmun, J. immunity,9 and the Court offers no reasons for abandoning that approach here. Comity is not just a vague political concern favoring international cooperation when it is in our interest to do so. Rather it is a principle under which judicial decisions reflect the systemic value of reciprocal tolerance and goodwill. See Maier, Extraterritorial Jurisdiction at a Crossroads: An Intersection Between Public and International Law, 76 Am. J. Int’l L. 280, 281-285 (1982); J. Story, Commentaries on the Conflict of Laws §§35, 38 (8th ed. 1883).10 As in the choice-of-law analysis, which from the very beginning has been linked to international comity, the threshold question in a comity analysis is whether there is in fact a true conflict between domestic and foreign law. When there is a conflict, a court should seek a reasonable accommodation that reconciles the central concerns of both sets of laws. In doing so, it should perform a tripartite analysis that considers the foreign interests, the interests of the United States, and the mutual interests of all nations in a smoothly functioning international legal regime.11 9 See, e. g., First National City Bank v. Banco Para el Comercio Exterior de Cuba, 462 U. S. 611, 626-627 (1983) (presumption that for purposes of sovereign immunity “government instrumentalities established as juridical entities distinct and independent from their sovereign should normally be treated as such” on the basis of respect for “principles of comity between nations”). 10 Justice Story used the phrase “comity of nations” to “express the true foundation and extent of the obligation of the laws of one nation within the territories of another.” § 38. “The true foundation on which the administration of international law must rest is, that the rules which are to govern are those which arise from mutual interest and utility, from a sense of the inconveniences which would result from a contrary doctrine, and from a sort of moral necessity to do justice, in order that justice may be done to us in return.” §35. 11 Choice-of-law decisions similarly reflect the needs of the system as a whole as well as the concerns of the forums with an interest in the controversy. “Probably the most important function of choice-of-law rules is to make the interstate and international systems work well. Choice-of-law 556 OCTOBER TERM, 1986 Opinion of Blackmun, J. 482 U. S. In most cases in which a discovery request concerns a nation that has ratified the Convention there is no need to resort to comity principles; the conflicts they are designed to resolve already have been eliminated by the agreements expressed in the treaty. The analysis set forth in the Restatement (Revised) of Foreign Relations Law of the United States, see ante, at 544, n. 28, is perfectly appropriate for courts to use when no treaty has been negotiated to accommodate the different legal systems. It would also be appropriate if the Convention failed to resolve the conflict in a particular case. The Court, however, adds an additional layer of so-called comity analysis by holding that courts should determine on a case-by-case basis whether resort to the Convention is desirable. Although this analysis is unnecessary in the absence of any conflicts, it should lead courts to the use of the Convention if they recognize that the Convention already has largely accommodated all three categories of interests relevant to a comity analysis—foreign interests, domestic interests, and the interest in a well-functioning international order. A I am encouraged by the extent to which the Court emphasizes the importance of foreign interests and by its admonition to lower courts to take special care to respect those interests. See ante, at 546. Nonetheless, the Court’s view of the Convention rests on an incomplete analysis of the sovereign interests of foreign states. The Court acknowledges that evidence is normally obtained in civil-law countries by a judicial officer, ante, at 543, but it fails to recognize the significance of that practice. Under the classic view of territo-rules, among other things, should seek to further harmonious relations between states and to facilitate commercial intercourse between them. In formulating rules of choice of law, a state should have regard for the needs and policies of other states and of the community of states.” Restatement (Second) of Conflict of Laws § 6, Comment d, p. 13 (1971). SOCIÉTÉ NAT. IND. AÉRO. v. U. S. DIST. COURT 557 522 Opinion of Blackmun, J. rial sovereignty, each state has a monopoly on the exercise of governmental power within its borders and no state may perform an act in the territory of a foreign state without consent.12 As explained in the Report of United States Delegation to Eleventh Session of the Hague Conference on Private International Law, the taking of evidence in a civil-law country may constitute the performance of a public judicial act by an unauthorized foreign person: “In drafting the Convention, the doctrine of ‘judicial sovereignty’ had to be constantly borne in mind. Unlike the common-law practice, which places upon the parties to the litigation the duty of privately securing and presenting the evidence at the trial, the civil law considers obtaining of evidence a matter primarily for the courts, with the parties in the subordinate position of assisting the judicial authorities. “The act of taking evidence in a common-law country from a willing witness, without compulsion and without a breach of the peace, in aid of a foreign proceeding, is a purely private matter, in which the host country has no interest and in which its judicial authorities have normally no wish to participate. To the contrary, the same act in a civil-law country may be a public matter, and may constitute the performance of a public judicial act by an unauthorized foreign person. It may violate the 12 Chief Justice Marshall articulated the American formulation of this principle in The Schooner Exchange v. McFaddon, 7 Cranch, at 136: “The jurisdiction of the nation within its own territory is necessarily exclusive and absolute. It is susceptible of no limitation not imposed by itself. Any restriction upon it, deriving validity from an external source, would imply a diminution of its sovereignty to the extent of the restriction .... “All exceptions, therefore, to the full and complete power of a nation within its own territories, must be traced up to the consent of the nation itself. They can flow from no other legitimate source.” 558 OCTOBER TERM, 1986 Opinion of Blackmun, J. 482 U. S. ‘judicial sovereignty’ of the host country, unless its authorities participate or give their consent.” 8 Int’l Legal Materials 785, 806 (1969).13 Some countries also believe that the need to protect certain underlying substantive rights requires judicial control of the taking of evidence. In the Federal Republic of Germany, for example, there is a constitutional principle of proportionality, pursuant to which a judge must protect personal privacy, commercial property, and business secrets. Interference with these rights is proper only if “necessary to protect other persons’ rights in the course of civil litigation.” See Meessen, The International Law on Taking Evidence From, Not In, a Foreign State, The Anschutz and Messerschmitt opinions of the United States Court of Appeals for the Fifth Circuit (Mar. 31, 1986), as set forth in App. to Brief for Anschuetz & Co. GmbH and Messerschmitt-Boelkow-Blohm GmbH as Amici Curiae 27a-28a.14 13 Many of the nations that participated in drafting the Convention regard nonjudicial evidence taking from even a willing witness as a violation of sovereignty. A questionnaire circulated to participating governments prior to the negotiations contained the question, “Is there in your State any legal provision or any official practice, based on concepts of sovereignty or public policy, preventing the taking of voluntary testimony for use in a foreign court without passing through the courts of your State?” Questionnaire on the Taking of Evidence Abroad, with Annexes, Actes et documents 9, 10. Of the 20 replies, 8 Governments—Egypt, France, West Germany, Italy, Luxembourg, Norway, Switzerland, and Turkey— stated that they did have objections to unauthorized evidence taking. Réponses des Gouvernements au Questionnaire sur la réception des dépositions à l’étranger, Actes et documents 21-46; see also Oxman, 37 U. Miami L. Rev., at 764, n. 84. 14 The Federal Republic of Germany, in its diplomatic protests to the United States, has emphasized the constitutional basis of the rights violated by American discovery orders. See, e. g., Diplomatic Note, dated Apr. 8, 1986, from the Embassy of the Federal Republic of Germany. App. A to Brief for Federal Republic of Germany as Amicus Curiae 20a. SOCIÉTÉ NAT. IND. AÉRO. v. U. S. DIST. COURT 559 522 Opinion of Blackmun, J. The United States recently recognized the importance of these sovereignty principles by taking the broad position that the Convention “must be interpreted to preclude an evidence taking proceeding in the territory of a foreign state party if the Convention does not authorize it and the host country does not otherwise permit it.” Brief for United States as Amicus Curiae in Volkswagenwerk Aktiengesellschaft n. Falzon, O. T. 1983, No. 82-1888, p. 6. Now, however, it appears to take a narrower view of what constitutes an "evidence taking procedure,” merely stating that “oral depositions on foreign soil. . . are improper without the consent of the foreign nation.” Tr. of Oral Arg. 23. I am at a loss to understand why gathering documents or information in a foreign country, even if for ultimate production in the United States, is any less an imposition on sovereignty than the taking of a deposition when gathering documents also is regarded as a judicial function in a civil-law nation. Use of the Convention advances the sovereign interests of foreign nations because they have given consent to Convention procedures by ratifying them. This consent encompasses discovery techniques that would otherwise impinge on the sovereign interests of many civil-law nations. In the absence of the Convention, the informal techniques provided by Articles 15-22 of the Convention—taking evidence by a diplomatic or consular officer of the requesting state and the use of commissioners nominated by the court of the state where the action is pending—would raise sovereignty issues similar to those implicated by a direct discovery order from a foreign court. “Judicial” activities are occurring on the soil of the sovereign by agents of a foreign state.15 These voluntary discovery procedures are a great boon to United States liti 15 See Edwards, Taking of Evidence Abroad in Civil or Commercial Matters, 18 Int’l & Comp. L. Q. 618, 647 (1969). A number of countries that ratified the Convention also expressed fears that the taking of evidence by consuls or commissioners could lead to abuse. Ibid. 560 OCTOBER TERM, 1986 Opinion of Blackmun, J. 482 U. S. gants and are used far more frequently in practice than is compulsory discovery pursuant to letters of request.16 Civil-law contracting parties have also agreed to use, and even to compel, procedures for gathering evidence that are diametrically opposed to civil-law practices. The civil-law system is inquisitional rather than adversarial and the judge normally questions the witness and prepares a written summary of the evidence.17 Even in common-law countries no system of evidence-gathering resembles that of the United States.18 Under Article 9 of the Convention, however, a foreign court must grant a request to use a “special method or procedure,” which includes requests to compel attendance of “According to the French Government, the overwhelming majority of discovery requests by American litigants are “satisfied willingly. . . before consular officials and, occasionally, commissioners, and without the need for involvement by a French court or use of its coercive powers.” Brief for Republic of France as Amicus Curiae 24. Once a United States court in which an action is pending issues an order designating a diplomatic or consular official of the United States stationed in Paris to take evidence, oral examination of American parties or witnesses may proceed. If evidence is sought from French nationals or other non-Americans, or if a commissioner has been named pursuant to Article 17 of the Convention, the Civil Division of International Judicial Assistance of the Ministry of Justice must authorize the discovery. The United States Embassy will obtain authorization at no charge or a party may make the request directly to the Civil Division. Authorization is granted routinely and, when necessary, has been obtained within one to two days. Brief, at 25. 17 For example, after the filing of the initial pleadings in a German court, the judge determines what evidence should be taken and who conducts the taking of evidence at various hearings. See, e. g., Langbein, The German Advantage in Civil Procedure, 52 U. Chi. L. Rev. 823, 826-828 (1985). All these proceedings are part of the “trial,” which is not viewed as a separate proceeding distinct from the rest of the suit. Id., at 826. 18 “In most common law countries, even England, one must often look hard to find the resemblances between pre-trial discovery there and pretrial discovery in the U. S. In England, for example, although document discovery is available, depositions do not exist, interrogatories have strictly limited use, and discovery as to third parties is not generally allowed.” S. Seidel, Extraterritorial Discovery in International Litigation 24 (1984). SOCIÉTÉ NAT. IND. AÉRO. v. U. S. DIST. COURT 561 522 Opinion of Blackmun, J. witnesses abroad, to administer oaths, to produce verbatim transcripts, or to permit examination of witnesses by counsel for both parties.19 These methods for obtaining evidence, which largely eliminate conflicts between the discovery procedures of the United States and the laws of foreign systems, have the consent of the ratifying nations. The use of these methods thus furthers foreign interests because discovery can proceed without violating the sovereignty of foreign nations. B The primary interest of the United States in this context is in providing effective procedures to enable litigants to obtain evidence abroad. This was the very purpose of the United States’ participation in the treaty negotiations and, for the most part, the Convention provides those procedures. The Court asserts that the letters of request procedure authorized by the Convention in many situations will be “unduly time consuming and expensive.” Ante, at 542. The Court offers no support for this statement and until the Convention is used extensively enough for courts to develop experience with it, such statements can be nothing other than speculation.20 Conspicuously absent from the Court’s assess 19 In France, the Nouveau Code de Procédure Civile, Arts. 736-748 (76th ed. Dalloz 1984), implements the Convention by permitting examination and cross-examination of witnesses by the parties and their attorneys, Art. 740, permitting a foreign judge to attend the proceedings, Art. 741, and authorizing the preparation of a verbatim transcript of the questions and answers at the expense of the requesting authority, Arts. 739, 748. German procedures are described in Shemanski, Obtaining Evidence in the Federal Republic of Germany: The Impact of The Hague Evidence Convention on German-American Judicial Cooperation, 17 Int’l Lawyer 465, 473-474 (1983). 20 The United States recounts the time and money expended by the SEC in attempting to use the Convention’s procedures to secure documents and testimony from third-party witnesses residing in England, France, Italy, and Guernsey to enforce the federal securities laws’ insider-trading provisions. See Brief for United States and Securities and Exchange Commission as Amici Curiae 15-18. As the United States admits, however, 562 OCTOBER TERM, 1986 Opinion of Blackmun, J. 482 U. S. ment is any consideration of resort to the Convention’s less formal and less time-consuming alternatives—discovery conducted by consular officials or an appointed commissioner. Moreover, unless the costs become prohibitive, saving time and money is not such a high priority in discovery that some additional burden cannot be tolerated in the interest of international goodwill. Certainly discovery controlled by litigants under the Federal Rules of Civil Procedure is not known for placing a high premium on either speed or costeffectiveness. There is also apprehension that the Convention procedures will not prove fruitful. Experience with the Convention suggests otherwise—contracting parties have honored their obligation to execute letters of request expeditiously and to use compulsion if necessary. See, e. g., Report on the Work of the Special Commission on the Operation of the Convention of 18 March 1970 on the Taking of Evidence Abroad in Civil or Commercial Matters, 17 Int’l Legal Materials 1425, 1431, § 5 F (1978) (“[r]efusal to execute turns out to be very infre- the experience of a governmental agency bringing an enforcement suit is “atypical” and has little relevance for the use of the Convention in disputes between private parties. In fact, according to the State Department, private plaintiffs “have found resort to the Convention more successful.” Id., at 18. The SEC’s attempts to use the Convention have raised questions of first impression, whose resolution in foreign courts has led to delays in particular litigation. For example, in In re Testimony of Constandi Nasser, Trib. Admin, de Paris, 6eme section—2eme chambre, No. 51546/6 (Dec. 17, 1985), the French Ministry of Justice approved expeditiously the SEC’s letter of request for testimony of a nonparty witness. The witness then raised a collateral attack, arguing that the SEC’s requests were administrative and therefore outside the scope of the Convention, which is limited by its terms to “civil or commercial matters.” The Ministry of Justice ruled against the attack and, on review, the French Administrative Court ruled in favor of the French Government and the SEC. By then, however, the SEC was in the process of settling the underlying litigation and did not seek further action on the letter of request. See Reply Brief for Petitioners 17, and nn. 35, 36. SOCIÉTÉ NAT. IND. AÉRO. v. U. S. DIST. COURT 563 522 Opinion of Blackmun, J. quent in practice”). By and large, the concessions made by parties to the Convention not only provide United States litigants with a means for obtaining evidence, but also ensure that the evidence will be in a form admissible in court. There are, however, some situations in which there is legitimate concern that certain documents cannot be made available under Convention procedures. Thirteen nations have made official declarations pursuant to Article 23 of the Convention, which permits a contracting state to limit its obligation to produce documents in response to a letter of request. See ante, at 536, n. 21. These reservations may pose problems that would require a comity analysis in an individual case, but they are not so all-encompassing as the majority implies — they certainly do not mean that a “contracting party could unilaterally abrogate . . . the Convention’s procedures.” Ante, at 537. First, the reservations can apply only to letters of request for documents. Thus, an Article 23 reservation affects neither the most commonly used informal Convention procedures for taking of evidence by a consul or a commissioner nor formal requests for depositions or interrogatories. Second, although Article 23 refers broadly to “pretrial discovery,” the intended meaning of the term appears to have been much narrower than the normal United States usage.21 The contracting parties for the most part have mod 21 The use of the term “pre-trial” seems likely to have been the product of a lack of communication. According to the United States delegates’ report, at a meeting of the Special Commission on the Operation of the Evidence Convention held in 1978, delegates from civil-law countries revealed a “gross misunderstanding” of the meaning of “pre-trial discovery,” thinking that it is something used before the institution of a suit to search for evidence that would lead to litigation. Report of the United States Delegation, 17 Int’l Legal Materials 1417, 1421 (1978). This misunderstanding is evidenced by the explanation of a French commentator that the “pretrial discovery” exception was a reinforcement of the rule in Article 1 of the Convention that a letter of request “shall not be used to obtain evidence which is not intended for use in judicial proceedings, commenced or contemplated” and by his comment that the Article 23 exception referred to 564 OCTOBER TERM, 1986 Opinion of Blackmun, J. 482 U. S. ified the declarations made pursuant to Article 23 to limit their reach. See 7 Martindale-Hubbell Law Directory (pt. VII) 14-19 (1986).22 Indeed, the emerging view of this exception to discovery is that it applies only to “requests that lack sufficient specificity or that have not been reviewed for the collection of evidence in advance of litigation. Gouguenheim, Convention sur l’obtention des preuves à l’étranger en matière civile et commerciale, 96 Journal du Droit International 315, 319 (1969). 22 France has recently modified its declaration as follows: “The declaration made by the Republic of France pursuant to Article 23 relating to letters of request whose purpose is ‘pre-trial discovery of documents’ does not apply so long as the requested documents are limitatively enumerated in the letter of request and have a direct and clear nexus with the subject matter of the litigation.” “La déclaration faite par la République française conformément à l’article 23 relatif aux commissions rogatoires qui ont pour objet la procédure de ‘pre-trial discovery of documents’ ne s’applique pas lorsque les documents demandés sont limitativement énumérés dans la commission rogatoire et ont un lien direct et précis avec l’objet du litige.” Letter from J. B. Raimond, Minister of Foreign Affairs, France, to H. H. van den Broek, Minister of Foreign Affairs, The Netherlands (Dec. 24, 1986). The Danish declaration is more typical: “The declaration made by the Kingdom of Denmark in accordance with article 23 concerning ‘Letters of Request issued for the purpose of obtaining pre-trial discovery of documents’ shall apply to any Letter of Request which requires a person: “a) to state what documents relevant to the proceedings to which the Letter of Request relates are, or have been, in his possession, other than particular documents specified in the Letter of Request; “or “b) to produce any documents other than particular documents which are specified in the Letter of Request, and which are likely to be in his possession.” Declaration of July 23, 1980, 7 Martindale-Hubbell Law Directory (pt. VII) 15 (1986). The Federal Republic of Germany, Italy, Luxembourg, and Portugal continue to have unqualified Article 23 declarations, id., at 16-18, but the German Government has drafted new regulations that would “permit pretrial production of specified and relevant documents in response to letters of request.” Brief for Anschuetz & Co. GmbH and Messerschmitt-Boelkow-Blohm GmbH as Amici Curiae 21. SOCIÉTÉ NAT. IND. AÉRO. v. U. S. DIST. COURT 565 522 Opinion of Blackmun, J. relevancy by the requesting court.” Oxman, The Choice Between Direct Discovery and Other Means of Obtaining Evidence Abroad: The Impact of the Hague Evidence Convention, 37 U. Miami L. Rev., at 777. Thus, in practice, a reservation is not the significant obstacle to discovery under the Convention that the broad wording of Article 23 would suggest.23 In this particular case, the “French ‘blocking statute,’” see ante, at 526, n. 6, poses an additional potential barrier to obtaining discovery from France. But any conflict posed by this legislation is easily resolved by resort to the Convention’s procedures. The French statute’s prohibitions are expressly “subject to” international agreements and applicable laws and it does not affect the taking of evidence under the Convention. See Toms, The French Response to the Extraterritorial Application of United States Antitrust Laws, 15 Int’l Lawyer 585, 593-599 (1981); Heck, Federal Republic of Germany and the EEC, 18 Int’l Lawyer 793, 800 (1984). The second major United States interest is in fair and equal treatment of litigants. The Court cites several fairness concerns in support of its conclusion that the Convention is not exclusive and apparently fears that a broad endorsement of the use of the Convention would lead to the same “unacceptable asymmetries.” See ante, at 540, n. 25. Courts can protect against the first two concerns noted by the majority—that a foreign party to a lawsuit would have a discovery advantage over a domestic litigant because it could obtain the advantages of the Federal Rules of Civil Procedure, and that a foreign company would have an economic 28 An Article 23 reservation and, in fact, the Convention in general require an American court to give closer scrutiny to the evidence requested than is normal in United States discovery, but this is not inconsistent with recent amendments to the Federal Rules of Civil Procedure that provide for a more active role on the part of the trial judge as a means of limiting discovery abuse. See Fed. Rule Civ. Proc. 26(b), (f), and (g) and accompanying Advisory Committee Notes. 566 OCTOBER TERM, 1986 Opinion of Blackmun, J. 482 U. S. competitive advantage because it would be subject to less extensive discovery—by exercising their discretionary powers to control discovery in order to ensure fairness to both parties. A court may “make any order which justice requires” to limit discovery, including an order permitting discovery only on specified terms and conditions, by a particular discovery method, or with limitation in scope to certain matters. Fed. Rule Civ. Proc. 26(c). If, for instance, resort to the Convention procedures would put one party at a disadvantage, any possible unfairness could be prevented by postponing that party’s obligation to respond to discovery requests until completion of the foreign discovery. Moreover, the Court’s arguments focus on the nationality of the parties, while it is actually the locus of the evidence that is relevant to use of the Convention: a foreign litigant trying to secure evidence from a foreign branch of an American litigant might also be required to resort to the Convention. The Court’s third fairness concern is illusory. It fears that a domestic litigant suing a national of a state that is not a party to the Convention would have an advantage over a litigant suing a national of a contracting state. This statement completely ignores the very purpose of the Convention. The negotiations were proposed by the United States in order to facilitate discovery, not to hamper litigants. Dissimilar treatment of litigants similarly situated does occur, but in the manner opposite to that perceived by the Court. Those who sue nationals of noncontracting states are disadvantaged by the unavailability of the Convention procedures. This is an unavoidable inequality inherent in the benefits conferred by any treaty that is less than universally ratified. In most instances, use of the Convention will serve to advance United States interests, particularly when those interests are viewed in a context larger than the immediate interest of the litigants’ discovery. The approach I propose is not a rigid per se rule that would require first use of the Convention without regard to strong indications that no evidence SOCIÉTÉ NAT. IND. AÉRO. v. U. S. DIST. COURT 567 522 Opinion of Blackmun, J. would be forthcoming. All too often, however, courts have simply assumed that resort to the Convention would be unproductive and have embarked on speculation about foreign procedures and interpretations. See, e. g., International Society for Krishna Consciousness, Inc. n. Lee, 105 F. R. D. 435, 449-450 (SDNY 1984); Graco, Inc. v. Kremlin, Inc., 101 F. R. D. 503, 509-512 (ND Ill. 1984). When resort to the Convention would be futile, a court has no choice but to resort to a traditional comity analysis. But even then, an attempt to use the Convention will often be the best way to discover if it will be successful, particularly in the present state of general inexperience with the implementation of its procedures by the various contracting states. An attempt to use the Convention will open a dialogue with the authorities in the foreign state and in that way a United States court can obtain an authoritative answer as to the limits on what it can achieve with a discovery request in a particular contracting state. C The final component of a comity analysis is to consider if there is a course that furthers, rather than impedes, the development of an ordered international system. A functioning system for solving disputes across borders serves many values, among them predictability, fairness, ease of commercial interactions, and “stablility through satisfaction of mutual expectations.” Laker Airways, Ltd. n. Sabena, Belgian World Airlines, 235 U. S. App. D. C., at 235, 731 F. 2d, at 937. These interests are common to all nations, including the United States. Use of the Convention would help develop methods for transnational litigation by placing officials in a position to communicate directly about conflicts that arise during discovery, thus enabling them to promote a reduction in those conflicts. In a broader framework, courts that use the Convention will avoid foreign perceptions of unfairness that result when United States courts show insensitivity to the interests 568 OCTOBER TERM, 1986 Opinion of Blackmun, J. 482 U. S. safeguarded by foreign legal regimes. Because of the position of the United States, economically, politically, and militarily, many countries may be reluctant to oppose discovery orders of United States courts. Foreign acquiescence to orders that ignore the Convention, however, is likely to carry a price tag of accumulating resentment, with the predictable long-term political cost that cooperation will be withheld in other matters. Use of the Convention is a simple step to take toward avoiding that unnecessary and undesirable consequence. IV I can only hope that courts faced with discovery requests for materials in foreign countries will avoid the parochial views that too often have characterized the decisions to date. Many of the considerations that lead me to the conclusion that there should be a general presumption favoring use of the Convention should also carry force when courts analyze particular cases. The majority fails to offer guidance in this endeavor, and thus it has missed its opportunity to provide predictable and effective procedures for international litigants in United States courts. It now falls to the lower courts to recognize the needs of the international commercial system and the accommodation of those needs already endorsed by the political branches and embodied in the Convention. To the extent indicated, I respectfully dissent. AIRPORT COMM’RS v. JEWS FOR JESUS, INC. 569 Syllabus BOARD OF AIRPORT COMMISSIONERS OF THE CITY OF LOS ANGELES et al. v. JEWS FOR JESUS, INC., ET AL. CERTIORARI TO THE UNITED STATES COURT OF APPEALS FOR THE NINTH CIRCUIT No. 86-104. Argued March 3, 1987—Decided June 15, 1987 Petitioner Board of Airport Commissioners of Los Angeles adopted a resolution banning all “First Amendment activities” within the “Central Terminal Area” at Los Angeles International Airport. Respondents, a nonprofit religious corporation and a minister for that organization, filed an action in Federal District Court challenging the resolution’s constitutionality, after the minister had stopped distributing free religious literature in the airport’s Central Terminal Area when warned against doing so by an airport officer. The court held that the Central Terminal Area was a traditional public forum under federal law and that the resolution was facially unconstitutional under the Federal Constitution. The Court of Appeals affirmed. Held: The resolution violates the First Amendment. It is facially unconstitutional under the First Amendment overbreadth doctrine regardless of whether the forum involved is a public or nonpublic forum (which need not be decided here). The resolution’s facial overbreadth is substantial since it prohibits all protected expression and does not merely regulate expressive activity that might create problems such as congestion or the disruption of airport users’ activities. Under such a sweeping ban, virtually every individual who enters the airport may be found to violate the resolution by engaging in some “First Amendment activit[y].” The ban would be unconstitutional even if the airport were a nonpublic forum because no conceivable governmental interest would justify such an absolute prohibition of speech. Moreover, the resolution’s words leave no room for a narrowing, saving construction by state courts. Cf. Baggett v. Bullitt, 377 U. S. 360. The suggestion that the resolution is not substantially overbroad because it is intended to reach only expressive activity unrelated to airport-related purposes is unpersuasive. Much non-disruptive speech may not be airport related, but is still protected speech even in a nonpublic forum. Moreover, the vagueness of the suggested construction—which would result in giving airport officials the power to decide in the first instance whether a given activity is airport related—presents serious constitutional difficulty. Pp. 572-577. 785 F. 2d 791, affirmed. 570 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. O’Connor, J., delivered the opinion for a unanimous Court. White, J., filed a concurring opinion, in which Rehnquist, C. J., joined, post, p. 577. James R. Kapel argued the cause for petitioners. With him on the briefs was James H. Pearson. Jay Alan Sekulow argued the cause pro hac vice for respondents. With him on the brief were Andrew J. Ekono-mou, Barry A. Fisher, and Wendell R. Bird* Justice O’Connor delivered the opinion of the Court. The issue presented in this case is whether a resolution banning all “First Amendment activities” at Los Angeles International Airport (LAX) violates the First Amendment. I On July 13, 1983, the Board of Airport Commissioners (Board) adopted Resolution No. 13787, which provides in pertinent part: “NOW, THEREFORE, BE IT RESOLVED by the Board of Airport Commissioners that the Central Terminal Area at Los Angeles International Airport is not *Briefs of amici curiae urging reversal were filed for the city of St. Louis, Missouri, by James J. Wilson and Edward J. Hanlon; and for the Airport Operators Council International by Arthur P. Berg, Anne M. Tannenbaum, and Arnold D. Kolikoff. Briefs of amici curiae urging affirmance were filed for the American Federation of Labor and Congress of Industrial Organizations by Marsha S. Berzon and Laurence Gold; for the Christian Legal Society et al. by Michael J. Woodruff, Samuel E. Ericsson, Kimberlee W. Colby, and Forest D. Montgomery; for the Council on Religious Freedom by Lee Boothby, James M. Parker, Robert W. Nixon, and Rolland Truman; for the Rutherford Institute et al. by W. Charles Bundren, Ira W. Still III, Wendell R. Bird, Thomas W. Strahan, James J. Knicely, and Alfred J. Lindh; and for the Jesus People U. S. A. Full Gospel Ministries by Robert L. Graham. A brief of amicus curiae was filed for the International Society for Krishna Consciousness of California, Inc., by David M. Liberman. AIRPORT COMM’RS v. JEWS FOR JESUS, INC. 571 569 Opinion of the Court open for First Amendment activities by any individual and/or entity; “BE IT FURTHER RESOLVED that after the effective date of this Resolution, if any individual and/or entity seeks to engage in First Amendment activities within the Central Terminal Area at Los Angeles International Airport, said individual and/or entity shall be deemed to be acting in contravention of the stated policy of the Board of Airport Commissioners in reference to the uses permitted within the Central Terminal Area at Los Angeles International Airport; and “BE IT FURTHER RESOLVED that if any individual or entity engages in First Amendment activities within the Central Terminal Area at Los Angeles International Airport, the City Attorney of the City of Los Angeles is directed to institute appropriate litigation against such individual and/or entity to ensure compliance with this Policy statement of the Board of Airport Commissioners . . . .” App. 4a-5a. Respondent Jews for Jesus, Inc., is a nonprofit religious corporation. On July 6, 1984, Alan Howard Snyder, a minister of the Gospel for Jews for Jesus, was stopped by a Department of Airports peace officer while distributing free religious literature on a pedestrian walkway in the Central Terminal Area at LAX. The officer showed Snyder a copy of the resolution, explained that Snyder’s activities violated the resolution, and requested that Snyder leave LAX. The officer warned Snyder that the city would take legal action against him if he refused to leave as requested. Id., at 19a-20a. Snyder stopped distributing the leaflets and left the airport terminal. Id., at 20a. Jews for Jesus and Snyder then filed this action in the District Court for the Central District of California, challeng 572 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. ing the constitutionality of the resolution under both the California and Federal Constitutions. First, respondents contended that the resolution was facially unconstitutional under Art. I, §2, of the California Constitution and the First Amendment to the United States Constitution because it bans all speech in a public forum. Second, they alleged that the resolution had been applied to Jews for Jesus in a discriminatory manner. Finally, respondents urged that the resolution was unconstitutionally vague and overbroad. When the case came before the District Court for trial, the parties orally stipulated to the facts, and the District Court treated the trial briefs as cross-motions for summary judgment. The District Court held that the Central Terminal Area was a traditional public forum under federal law, and that the resolution was facially unconstitutional under the United States Constitution. The District Court declined to reach the other issues raised by Jews for Jesus, and did not address the constitutionality of the resolution under the California Constitution. The Court of Appeals for the Ninth Circuit affirmed. 785 F. 2d 791 (1986). Relying on Rosen v. Port of Portland, 641 F. 2d 1243 (CA9 1981), and Kuszynski n. Oakland, 479 F. 2d 1130 (CA9 1973), the Court of Appeals concluded that “an airport complex is a traditional public forum,” 785 F. 2d, at 795, and held that the resolution was unconstitutional on its face under the Federal Constitution. We granted certiorari, 479 U. S. 812 (1986), and now affirm, but on different grounds. II In balancing the government’s interest in limiting the use of its property against the interests of those who wish to use the property for expressive activity, the Court has identified three types of fora: the traditional public forum, the public forum created by government designation, and the nonpublic forum. Perry Ed. Assn. v. Perry Local Educators’ Assn., 460 U. S. 37, 45-46 (1983). The proper First Amendment analysis differs depending on whether the area in question AIRPORT COMM’RS v. JEWS FOR JESUS, INC. 573 569 Opinion of the Court falls in one category rather than another. In a traditional public forum or a public forum by government designation, we have held that First Amendment protections are subject to heightened scrutiny: “In these quintessential public forums, the government may not prohibit all communicative activity. For the State to enforce a content-based exclusion it must show that its regulation is necessary to serve a compelling state interest and that it is narrowly drawn to achieve that end. . . . The State may also enforce regulations of the time, place, and manner of expression which are content-neutral, are narrowly tailored to serve a significant government interest, and leave open ample alternative channels of communication.” Id., at 45. We have further held, however, that access to a nonpublic forum may be restricted by government regulation as long as the regulation “is reasonable and not an effort to suppress expression merely because officials oppose the speaker’s view.” Id., at 46. The petitioners contend that LAX is neither a traditional public forum nor a public forum by government designation, and accordingly argue that the latter standard governing access to a nonpublic forum is appropriate. The respondents, in turn, argue that LAX is a public forum subject only to reasonable time, place, or manner restrictions. Moreover, at least one commentator contends that Perry does not control a case such as this in which the respondents already have access to the airport, and therefore concludes that this case is analogous to Tinker v. Des Moines School Dist., 393 U. S. 503 (1969). See Laycock, Equal Access and Moments of Silence: The Equal Status of Religious Speech by Private Speakers, 81 Nw. U. L. Rev. 1, 48 (1986). Because we conclude that the resolution is facially unconstitutional under the the First Amendment overbreadth doctrine regardless of the proper standard, we need not decide whether LAX is indeed 574 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. a public forum, or whether the Perry standard is applicable when access to a nonpublic forum is not restricted. Under the First Amendment overbreadth doctrine, an individual whose own speech or conduct may be prohibited is permitted to challenge a statute on its face “because it also threatens others not before the court—those who desire to engage in legally protected expression but who may refrain from doing so rather than risk prosecution or undertake to have the law declared partially invalid.” Brockett v. Spokane Arcades, Inc., 472 U. S. 491, 503 (1985). A statute may be invalidated on its face, however, only if the overbreadth is “substantial.” Houston v. Hill, ante, at 458-459; New York v. Ferber, 458 U. S. 747, 769 (1982); Broadrick v. Oklahoma, 413 U. S. 601, 615 (1973). The requirement that the overbreadth be substantial arose from our recognition that application of the overbreadth doctrine is, “manifestly, strong medicine,” Broadrick n. Oklahoma, supra, at 613, and that “there must be a realistic danger that the statute itself will significantly compromise recognized First Amendment protections of parties not before the Court for it to be facially challenged on overbreadth grounds.” City Council of Los Angeles v. Taxpayers for Vincent, 466 U. S. 789, 801 (1984). On its face, the resolution at issue in this case reaches the universe of expressive activity, and, by prohibiting all protected expression, purports to create a virtual “First Amendment Free Zone” at LAX. The resolution does not merely regulate expressive activity in the Central Terminal Area that might create problems such as congestion or the disruption of the activities of those who use LAX. Instead, the resolution expansively states that LAX “is not open for First Amendment activities by any individual and/or entity,” and that “any individual and/or entity [who] seeks to engage in First Amendment activities within the Central Terminal Area. . . shall be deemed to be acting in contravention of the stated policy of the Board of Airport Commissioners.” App. 4a-5a. The resolution therefore does not merely reach the AIRPORT COMM’RS v. JEWS FOR JESUS, INC. 575 569 Opinion of the Court activity of respondents at LAX; it prohibits even talking and reading, or the wearing of campaign buttons or symbolic clothing. Under such a sweeping ban, virtually every individual who enters LAX may be found to violate the resolution by engaging in some “First Amendment activit[y].” We think it obvious that such a ban cannot be justified even if LAX were a nonpublic forum because no conceivable governmental interest would justify such an absolute prohibition of speech. Additionally, we find no apparent saving construction of the resolution. The resolution expressly applies to all “First Amendment activities,” and the words of the resolution simply leave no room for a narrowing construction. In the past the Court sometimes has used either abstention or certification when, as here, the state courts have not had the opportunity to give the statute under challenge a definite construction. See, e. g., Babbitt v. Farm Workers, 442 U. S. 289 (1979). Neither option, however, is appropriate in this case because California has no certification procedure, and the resolution is not “fairly subject to an interpretation which will render unnecessary or substantially modify the federal constitutional question.” Harmon n. Forssenius, 380 U. S. 528, 535 (1965). The difficulties in adopting a limiting construction of the resolution are not unlike those found in Baggett v. Bullitt, 377 U. S. 360 (1964). At issue in Baggett was the constitutionality of several statutes requiring loyalty oaths. The Baggett Court concluded that abstention would serve no purpose given the lack of any limiting construction, and held the statutes unconstitutional on their face under the First Amendment overbreadth doctrine. We observed that the challenged loyalty oath was not “open to one or a few interpretations, but to an indefinite number,” and concluded that “[i]t is fictional to believe that anything less than extensive adjudications, under the impact of a variety of factual situations, would bring the oath within the bounds of permissible constitutional certainty.” Id., at 378. Here too, it is 576 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. difficult to imagine that the resolution could be limited by anything less than a series of adjudications, and the chilling effect of the resolution on protected speech in the meantime would make such a case-by-case adjudication intolerable. The petitioners suggest that the resolution is not substantially overbroad because it is intended to reach only expressive activity unrelated to airport-related purposes. Such a limiting construction, however, is of little assistance in substantially reducing the overbreadth of the resolution. Much nondisruptive speech—such as the wearing of a T-shirt or button that contains a political message—may not be “airport related,” but is still protected speech even in a nonpublic forum. See Cohen n. California, 403 U. S. 15 (1971). Moreover, the vagueness of this suggested construction itself presents serious constitutional difficulty. The line between airport-related speech and nonairport-related speech is, at best, murky. The petitioners, for example, suggest that an individual who reads a newspaper or converses with a neighbor at LAX is engaged in permitted “airport-related” activity because reading or conversing permits the traveling public to “pass the time.” Reply Brief for Petitioners 12. We presume, however, that petitioners would not so categorize the activities of a member of a religious or political organization who decides to “pass the time” by distributing leaflets to fellow travelers. In essence, the result of this vague limiting construction would be to give LAX officials alone the power to decide in the first instance whether a given activity is airport related. Such a law that “confers on police a virtually unrestrained power to arrest and charge persons with a violation” of the resolution is unconstitutional because “[t]he opportunity for abuse, especially where a statute has received a virtually open-ended interpretation, is self-evident.” Lewis n. City of New Orleans, 415 U. S. 130, 135-136 (1974) (Powell, J., concurring); see also Houston v. Hill, ante, at 465; Kolender v. Lawson, 461 U. S. 352, 358 (1983). AIRPORT COMM’RS v. JEWS FOR JESUS, INC. 577 569 White, J., concurring We conclude that the resolution is substantially overbroad, and is not fairly subject to a limiting construction. Accordingly, we hold that the resolution violates the First Amendment. The judgment of the Court of Appeals is Affirmed. Justice White, with whom The Chief Justice joins, concurring. I join the Court’s opinion but suggest that it should not be taken as indicating that a majority of the Court considers the Los Angeles International Airport to be a traditional public forum. That issue was one of the questions on which we granted certiorari, and we should not have postponed it for another day. 578 OCTOBER TERM, 1986 Syllabus 482 U. S. EDWARDS, GOVERNOR OF LOUISIANA, et al. v. AGUILLARD et al. APPEAL FROM THE UNITED STATES COURT OF APPEALS FOR THE FIFTH CIRCUIT No. 85-1513. Argued December 10, 1986—Decided June 19, 1987 Louisiana’s “Creationism Act” forbids the teaching of the theory of evolution in public elementary and secondary schools unless accompanied by instruction in the theory of “creation science.” The Act does not require the teaching of either theory unless the other is taught. It defines the theories as “the scientific evidences for [creation or evolution] and inferences from those scientific evidences.” Appellees, who include Louisiana parents, teachers, and religious leaders, challenged the Act’s constitutionality in Federal District Court, seeking an injunction and declaratory relief. The District Court granted summary judgment to appellees, holding that the Act violated the Establishment Clause of the First Amendment. The Court of Appeals affirmed. Held: 1. The Act is facially invalid as violative of the Establishment Clause of the First Amendment, because it lacks a clear secular purpose. Pp. 585-594. (a) The Act does not further its stated secular purpose of “protecting academic freedom.” It does not enhance the freedom of teachers to teach what they choose and fails to further the goal of “teaching all of the evidence.” Forbidding the teaching of evolution when creation science is not also taught undermines the provision of a comprehensive scientific education. Moreover, requiring the teaching of creation science with evolution does not give schoolteachers a flexibility that they did not already possess to supplant the present science curriculum with the presentation of theories, besides evolution, about the origin of life. Furthermore, the contention that the Act furthers a “basic concept of fairness” by requiring the teaching of all of the evidence on the subject is without merit. Indeed, the Act evinces a discriminatory preference for the teaching of creation science and against the teaching of evolution by requiring that curriculum guides be developed and resource services supplied for teaching creationism but not for teaching evolution, by limiting membership on the resource services panel to “creation scientists,” and by forbidding school boards to discriminate against anyone who “chooses to be a creation-scientist” or to teach creation science, while failing to protect those who choose to teach other theories or who refuse EDWARDS v. AGUILLARD 579 578 Syllabus to teach creation science. A law intended to maximize the comprehensiveness and effectiveness of science instruction would encourage the teaching of all scientific theories about human origins. Instead, this Act has the distinctly different purpose of discrediting evolution by counterbalancing its teaching at every turn with the teaching of creationism. Pp. 586-589. (b) The Act impermissibly endorses religion by advancing the religious belief that a supernatural being created humankind. The legislative history demonstrates that the term “creation science,” as contemplated by the state legislature, embraces this religious teaching. The Act’s primary purpose was to change the public school science curriculum to provide persuasive advantage to a particular religious doctrine that rejects the factual basis of evolution in its entirety. Thus, the Act is designed either to promote the theory of creation science that embodies a particular religious tenet or to prohibit the teaching of a scientific . theory disfavored by certain religious sects. In either case, the Act violates the First Amendment. Pp. 589-594. 2. The District Court did not err in granting summary judgment upon a finding that appellants had failed to raise a genuine issue of material fact. Appellants relied on the “uncontroverted” affidavits of scientists, theologians, and an education administrator defining creation science as “origin through abrupt appearance in complex form” and alleging that such a viewpoint constitutes a true scientific theory. The District Court, in its discretion, properly concluded that the postenactment testimony of these experts concerning the possible technical meanings of the Act’s terms would not illuminate the contemporaneous purpose of the state legislature when it passed the Act. None of the persons making the affidavits produced by appellants participated in or contributed to the enactment of the law. Pp. 594-596. 765 F. 2d 1251, affirmed. Brennan, J., delivered the opinion of the Court, in which Marshall, Blackmun, Powell, and Stevens, JJ., joined, and in all but Part II of which O’Connor, J., joined. Powell, J., filed a concurring opinion, in which O’Connor, J., joined, post, p. 597. White, J., filed an opinion concurring in the judgment, post, p. 608. Scalia, J., filed a dissenting opinion, in which Rehnquist, C. J., joined, post, p. 610. Wendell R. Bird, Special Assistant Attorney General of Georgia, argued the cause for appellants. With him on the briefs were A. Morgan Brian, Jr., and Thomas T. Anderson, Special Assistant Attorneys General, Kendall L. Vick, and 580 OCTOBER TERM, 1986 Syllabus 482 U. S. Patricia Nalley Bowers, Assistant Attorney General of Louisiana. Jay Topkis argued the cause for appellees. With him on the brief was John DiGiulio, Samuel I. Rosenberg, Allen Blumstein, Gerard E. Harper, Jack D. Novik, Burt Neu-borne, Norman Dorsen, John Sexton, and Ron Wilson.* Justice Brennan delivered the opinion of the Court.! The question for decision is whether Louisiana’s “Balanced Treatment for Creation-Science and Evolution-Science in Public School Instruction” Act (Creationism Act), La. Rev. Stat. Ann. §§17:286.1-17:286.7 (West 1982), is facially in *Briefs of amici curiae urging reversal were filed for the Catholic League for Religious and Civil Rights by Steven Frederick McDowell; for the Christian Legal Society et al. by Michael J. Woodruff, Kimberlee W. Colby, Samuel E. Ericsson, and Forest D. Montgomery; and for Concerned Women for America by Michael P. Farris and Jordan W. Lorence. Briefs of amici curiae urging affirmance were filed for the State of New York et al. by Robert Abrams, Attorney General of New York, 0. Peter Sherwood, Solicitor General, Paul M. Glickman, Jane Levine, Suzanne Lynn, and Marla Tepper, Assistant Attorneys General, and Neil F. Hartigan, Attorney General of Illinois; for the American Association of University Professors et al. by Ann H. Franke, Jacqueline W. Mintz, and Sheldon E. Steinbach; for the American Federation of Teachers, AFL-CIO, by Bruce A. Miller and Stuart M. Israel; for the American Jewish Congress et al. by Marvin E. Frankel, Marc D. Stem, and Ronald A. Krauss; for Americans United for Separation of Church and State et al. by Lee Boothby, Samuel Rabinove, Richard T. Foltin, and James M. Parker; for the Anti-Defamation League of B’nai B’rith et al. by Ruti G. Teitel, Justin J. Finger, Jeffrey P. Sinensky, and Steven M. Freeman; for the National Academy of Sciences by Barry H. Garfinkel and Mark Herlihy; for the New York Committee for Public Education and Religious Liberty by Leo Pfeffer; for People for the American Way et al. by Timothy B. Dyk, A. Douglas Melamed, and Kerry W. Kircher; for the Spartacist League et al. by Rachel H. Wolkenstein; and for 72 Nobel Laureates et al. by Walter B. Slocombe. Briefs of amici curiae were filed for the Rabbinical Alliance of America et al. by John W. Whitehead and Larry L. Crain; and for Reverend Bill McLean et al. by Philip E. Kaplan. ■¡■Justice O’Connor joins all but Part II of this opinion. EDWARDS v. AGUILLARD 581 578 Opinion of the Court valid as violative of the Establishment Clause of the First Amendment. I The Creationism Act forbids the teaching of the theory of evolution in public schools unless accompanied by instruction in “creation science.” § 17:286.4A. No school is required to teach evolution or creation science. If either is taught, however, the other must also be taught. Ibid. The theories of evolution and creation science are statutorily defined as “the scientific evidences for [creation or evolution] and inferences from those scientific evidences.” §§ 17.286.3(2) and (3). Appellees, who include parents of children attending Louisiana public schools, Louisiana teachers, and religious leaders, challenged the constitutionality of the Act in District Court, seeking an injunction and declaratory relief.1 Appellants, Louisiana officials charged with implementing the Act, defended on the ground that the purpose of the Act is to protect a legitimate secular interest, namely, academic freedom.2 Appellees attacked the Act as facially invalid because ’Appellants, the Louisiana Governor, the Attorney General, the State Superintendent, the State Department of Education and the St. Tammany Parish School Board, agreed not to implement the Creationism Act pending the final outcome of this litigation. The Louisiana Board of Elementary and Secondary Education, and the Orleans Parish School Board were among the original defendants in the suit but both later realigned as plaintiffs. 2 The District Court initially stayed the action pending the resolution of a separate lawsuit brought by the Act’s legislative sponsor and others for declaratory and injunctive relief. After the separate suit was dismissed on jurisdictional grounds, Keith v. Louisiana Department of Education, 553 F. Supp. 295 (MD La. 1982), the District Court lifted its stay in this case and held that the Creationism Act violated the Louisiana Constitution. The court ruled that the State Constitution grants authority over the public school system to the Board of Elementary and Secondary Education rather than the state legislature. On appeal, the Court of Appeals certified the question to the Louisiana Supreme Court, which found the Creationism Act did not violate the State Constitution, Aguillard n. Treen, 440 So. 2d 704 (1983). The Court of Appeals then remanded the case 582 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. it violated the Establishment Clause and made a motion for summary judgment. The District Court granted the motion. Aguillard n. Treen, 634 F. Supp. 426 (ED La. 1985). The court held that there can be no valid secular reason for prohibiting the teaching of evolution, a theory historically opposed by some religious denominations. The court further concluded that “the teaching of ‘creation-science’ and ‘creationism,’ as contemplated by the statute, involves teaching ‘tailored to the principles’ of a particular religious sect or group of sects.” Id., at 427 (citing Epperson n. Arkansas, 393 U. S. 97, 106 (1968)). The District Court therefore held that the Creationism Act violated the Establishment Clause either because it prohibited the teaching of evolution or because it required the teaching of creation science with the purpose of advancing a particular religious doctrine. The Court of Appeals affirmed. 765 F. 2d 1251 (CA5 1985). The court observed that the statute’s avowed purpose of protecting academic freedom was inconsistent with requiring, upon risk of sanction, the teaching of creation science whenever evolution is taught. Id., at 1257. The court found that the Louisiana Legislature’s actual intent was “to discredit evolution by counterbalancing its teaching at every turn with the teaching of creationism, a religious belief.” Ibid. Because the Creationism Act was thus a law furthering a particular religious belief, the Court of Appeals held that the Act violated the Establishment Clause. A suggestion for rehearing en banc was denied over a dissent. 778 F. 2d 225 (CA5 1985). We noted probable jurisdiction, 476 U. S. 1103 (1986), and now affirm. II The Establishment Clause forbids the enactment of any law “respecting an establishment of religion.”3 The Court to the District Court to determine whether the Creationism Act violates the Federal Constitution. Aguillard v. Treen, 720 F. 2d 676 (CA5 1983). 3 The First Amendment states: “Congress shall make no law respecting an establishment of religion . . . .” Under the Fourteenth Amendment, EDWARDS v. AGUILLARD 583 578 Opinion of the Court has applied a three-pronged test to determine whether legislation comports with the Establishment Clause. First, the legislature must have adopted the law with a secular purpose. Second, the statute’s principal or primary effect must be one that neither advances nor inhibits religion. Third, the statute must not result in an excessive entanglement of government with religion. Lemon v. Kurtzman, 403 U. S. 602, 612-613 (1971).4 State action violates the Establishment Clause if it fails to satisfy any of these prongs. In this case, the Court must determine whether the Establishment Clause was violated in the special context of the public elementary and secondary school system. States and local school boards are generally afforded considerable discretion in operating public schools. See Bethel School Dist. No. 403 v. Fraser, 478 U. S. 675, 683 (1986); id., at 687 (Brennan, J., concurring in judgment); Tinker v. Des Moines Independent Community School Dist., 393 U. S. 503, 507 (1969). “At the same time ... we have necessarily recognized that the discretion of the States and local school boards in matters of education must be exercised in a manner that comports with the transcendent imperatives of the First Amendment.” Board of Education, Island Trees Union Free School Dist. No. 26 n. Pico, 457 U. S. 853, 864 (1982). The Court has been particularly vigilant in monitoring compliance with the Establishment Clause in elementary and this “fundamental concept of liberty” applies to the States. Cantwell v. Connecticut, 310 U. S. 296, 303 (1940). 4 The Lemon test has been applied in all cases since its adoption in 1971, except in Marsh v. Chambers, 463 U. S. 783 (1983), where the Court held that the Nebraska Legislature’s practice of opening a session with a prayer by a chaplain paid by the State did not violate the Establishment Clause. The Court based its conclusion in that case on the historical acceptance of the practice. Such a historical approach is not useful in determining the proper roles of church and state in public schools, since free public education was virtually nonexistent at the time the Constitution was adopted. See Wallace v. Jaffree, 472 U. S. 38, 80 (1985) (O’Connor, J., concurring in judgment) (citing Abington School Dist. v. Schempp, 374 U. S. 203, 238, and n. 7 (1963) (Brennan, J., concurring)). 584 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. secondary schools. Families entrust public schools with the education of their children, but condition their trust on the understanding that the classroom will not purposely be used to advance religious views that may conflict with the private beliefs of the student and his or her family. Students in such institutions are impressionable and their attendance is involuntary. See, e. g., Grand Rapids School Dist. v. Ball, 473 U. S. 373, 383 (1985); Wallace v. Jaffree, 472 U. S. 38, 60, n. 51 (1985); Meek v. Pittenger, 421 U. S. 349, 369 (1975); Abington School Dist. v. Schempp, 374 U. S. 203, 252-253 (1963) (Brennan, J., concurring). The State exerts great authority and coercive power through mandatory attendance requirements, and because of the students’ emulation of teachers as role models and the children’s susceptibility to peer pressure.5 See Bethel School Dist. No. 403 v. Fraser, supra, at 683; Wallace n. Jaffree, supra, at 81 (O’Connor, J., concurring in judgment). Furthermore, “[t]he public school is at once the symbol of our democracy and the most pervasive means for promoting our common destiny. In no activity of the State is it more vital to keep out divisive forces than in its schools . . . .” Illinois ex rel. McCollum v. Board of Education, 333 U. S. 203, 231 (1948) (opinion of Frankfurter, J.). Consequently, the Court has been required often to invalidate statutes which advance religion in public elementary and secondary schools. See, e. g., Grand Rapids School Dist. v. Ball, supra (school district’s use of religious school teachers in public schools); Wallace n. Jaffree, supra (Alabama statute authorizing moment of silence for school prayer); Stone v. 5 The potential for undue influence is far less significant with regard to college students who voluntarily enroll in courses. “This distinction warrants a difference in constitutional results.” Abington School Dist. v. Schempp, supra, at 253 (Brennan, J., concurring). Thus, for instance, the Court has not questioned the authority of state colleges and universities to offer courses on religion or theology. See Widmar v. Vincent, 454 U. S. 263, 271 (1981) (Powell, J.); id., at 281 (Stevens, J., concurring in judgment). EDWARDS v. AGUILLARD 585 578 Opinion of the Court Graham, 449 U. S. 39 (1980) (posting copy of Ten Commandments on public classroom wall); Epperson n. Arkansas, 393 U. S. 97 (1968) (statute forbidding teaching of evolution); Abington School Dist. v. Schempp, supra (daily reading of Bible); Engel n. Vitale, 370 U. S. 421, 430 (1962) (recitation of “denominationally neutral” prayer). Therefore, in employing the three-pronged Lemon test, we must do so mindful of the particular concerns that arise in the context of public elementary and secondary schools. We now turn to the evaluation of the Act under the Lemon test. Ill Lemon's first prong focuses on the purpose that animated adoption of the Act. “The purpose prong of the Lemon test asks whether government’s actual purpose is to endorse or disapprove of religion.” Lynch v. Donnelly, 465 U. S. 668, 690 (1984) (O’Connor, J., concurring). A governmental intention to promote religion is clear when the State enacts a law to serve a religious purpose. This intention may be evidenced by promotion of religion in general, see Wallace n. Jaffree, supra, at 52-53 (Establishment Clause protects individual freedom of conscience “to select any religious faith or none at all”), or by advancement of a particular religious belief, e. g., Stone v. Graham, supra, at 41 (invalidating requirement to post Ten Commandments, which are “undeniably a sacred text in the Jewish and Christian faiths”) (footnote omitted); Epperson v. Arkansas, supra, at 106 (holding that banning the teaching of evolution in public schools violates the First Amendment since “teaching and learning” must not “be tailored to the principles or prohibitions of any religious sect or dogma”). If the law was enacted for the purpose of endorsing religion, “no consideration of the second or third criteria [of Lemon] is necessary.” Wallace v. Jaffree, supra, at 56. In this case, appellants have identified no clear secular purpose for the Louisiana Act. 586 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. True, the Act’s stated purpose is to protect academic freedom. La. Rev. Stat. Ann. §17:286.2 (West 1982). This phrase might, in common parlance, be understood as referring to enhancing the freedom of teachers to teach what they will. The Court of Appeals, however, correctly concluded that the Act was not designed to further that goal.6 We find no merit in the State’s argument that the “legislature may not [have] use[d] the terms ‘academic freedom’ in the correct legal sense. They might have [had] in mind, instead, a basic concept of fairness; teaching all of the evidence.” Tr. of Oral Arg. 60. Even if “academic freedom” is read to mean “teaching all of the evidence” with respect to the origin of human beings, the Act does not further this purpose. The goal of providing a more comprehensive science curriculum is not furthered either by outlawing the teaching of evolution or by requiring the teaching of creation science. A While the Court is normally deferential to a State’s articulation of a secular purpose, it is required that the statement 6 The Court of Appeals stated that “[a]cademic freedom embodies the principle that individual instructors are at liberty to teach that which they deem to be appropriate in the exercise of their professional judgment.” 765 F. 2d, at 1257. But, in the State of Louisiana, courses in public schools are prescribed by the State Board of Education and teachers are not free, absent permission, to teach courses different from what is required. Tr. of Oral Arg. 44-46. “Academic freedom,” at least as it is commonly understood, is not a relevant concept in this context. Moreover, as the Court of Appeals explained, the Act “requires, presumably upon risk of sanction or dismissal for failure to comply, the teaching of creation-science whenever evolution is taught. Although states may prescribe public school curriculum concerning science instruction under ordinary circumstances, the compulsion inherent in the Balanced Treatment Act is, on its face, inconsistent with the idea of academic freedom as it is universally understood.” 765 F. 2d, at 1257 (emphasis in original). The Act actually serves to diminish academic freedom by removing the flexibility to teach evolution without also teaching creation science, even if teachers determine that such curriculum results in less effective and comprehensive science instruction. EDWARDS v. AGUILLARD 587 578 Opinion of the Court of such purpose be sincere and not a sham. See Wallace v. Jaffree, 472 U. S., at 64 (Powell, J., concurring); id., at 75 (O’Connor, J., concurring in judgment); Stone v. Graham, supra, at 41; Abington School Dist. v. Schempp, 374 U. S., at 223-224. As Justice O’Connor stated in Wallace: “It is not a trivial matter, however, to require that the legislature manifest a secular purpose and omit all sectarian endorsements from its laws. That requirement is precisely tailored to the Establishment Clause’s purpose of assuring that Government not intentionally endorse religion or a religious practice.” 472 U. S., at 75 (concurring in judgment). It is clear from the legislative history that the purpose of the legislative sponsor, Senator Bill Keith, was to narrow the science curriculum. During the legislative hearings, Senator Keith stated: “My preference would be that neither [creationism nor evolution] be taught.” 2 App. E-621. Such a ban on teaching does not promote—indeed, it undermines — the provision of a comprehensive scientific education. It is equally clear that requiring schools to teach creation science with evolution does not advance academic freedom. The Act does not grant teachers a flexibility that they did not already possess to supplant the present science curriculum with the presentation of theories, besides evolution, about the origin of life. Indeed, the Court of Appeals found that no law prohibited Louisiana public school teachers from teaching any scientific theory. 765 F. 2d, at 1257. As the president of the Louisiana Science Teachers Association testified, “[a]ny scientific concept that’s based on established fact can be included in our curriculum already, and no legislation allowing this is necessary.” 2 App. E-616. The Act provides Louisiana schoolteachers with no new authority. Thus the stated purpose is not furthered by it. The Alabama statute held unconstitutional in Wallace v. Jaffree, supra, is analogous. In Wallace, the State characterized its new law as one designed to provide a 1-minute period for meditation. We rejected that stated purpose as in 588 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. sufficient, because a previously adopted Alabama law already provided for such a 1-minute period. Thus, in this case, as in Wallace, “[a]ppellants have not identified any secular purpose that was not fully served by [existing state law] before the enactment of [the statute in question].” 472 U. S., at 59. Furthermore, the goal of basic “fairness” is hardly furthered by the Act’s discriminatory preference for the teaching of creation science and against the teaching of evolution.7 While requiring that curriculum guides be developed for creation science, the Act says nothing of comparable guides for evolution. La. Rev. Stat. Ann. § 17:286.7A (West 1982). Similarly, resource services are supplied for creation science but not for evolution. § 17:286.7B. Only “creation scientists” can serve on the panel that supplies the resource services. Ibid. The Act forbids school boards to discriminate against anyone who “chooses to be a creation-scientist” or to teach “creationism,” but fails to protect those who choose to teach evolution or any other noncreation science theory, or who refuse to teach creation science. § 17:286.4C. If the Louisiana Legislature’s purpose was solely to maximize the comprehensiveness and effectiveness of science instruction, it would have encouraged the teaching of all scientific theories about the origins of humankind.8 But under 7 The Creationism Act’s provisions appear among other provisions prescribing the courses of study in Louisiana’s public schools. These other provisions, similar to those in other States, prescribe courses of study in such topics as driver training, civics, the Constitution, and free enterprise. None of these other provisions, apart from those associated with the Creationism Act, nominally mandates “equal time” for opposing opinions within a specific area of learning. See, e. g., La. Rev. Stat. Ann. §§ 17:261-17:281 (West 1982 and Supp. 1987). 8 The dissent concludes that the Act’s purpose was to protect the academic freedom of students, and not that of teachers. Post, at 628. Such a view is not at odds with our conclusion that if the Act’s purpose was to provide comprehensive scientific education (a concern shared by students and teachers, as well as parents), that purpose was not advanced by the statute’s provisions. Supra, at 587. EDWARDS v. AGUILLARD 589 578 Opinion of the Court the Act’s requirements, teachers who were once free to teach any and all facets of this subject are now unable to do so. Moreover, the Act fails even to ensure that creation science will be taught, but instead requires the teaching of this theory only when the theory of evolution is taught. Thus we agree with the Court of Appeals’ conclusion that the Act does not serve to protect academic freedom, but has the distinctly different purpose of discrediting “evolution by counterbalancing its teaching at every turn with the teaching of creationism . . . .” 765 F. 2d, at 1257. B Stone v. Graham invalidated the State’s requirement that the Ten Commandments be posted in public classrooms. “The Ten Commandments are undeniably a sacred text in the Jewish and Christian faiths, and no legislative recitation of a supposed secular purpose can blind us to that fact.” 449 U. S., at 41 (footnote omitted). As a result, the contention that the law was designed to provide instruction on a “fundamental legal code” was “not sufficient to avoid conflict with the First Amendment.” Ibid. Similarly Abington School Dist. n. Schempp held unconstitutional a statute “requiring the selection and reading at the opening of the school day of verses from the Holy Bible and the recitation of the Lord’s Prayer by the students in unison,” despite the proffer of such secular purposes as the “promotion of moral values, the con Moreover, it is astonishing that the dissent, to prove its assertion, relies on a section of the legislation that was eventually deleted by the legislature. Compare § 3702 in 1 App. E-292 (text of section prior to amendment) with La. Rev. Stat. Ann. § 17:286.2 (West 1982). The dissent contends that this deleted section—which was explicitly rejected by the Louisiana Legislature-reveals the legislature’s “obviously intended meaning of the statutory terms ‘academic freedom.’” Post, at 628. Quite to the contrary, Boudreaux, the main expert relied on by the sponsor of the Act, cautioned the legislature that the words “academic freedom” meant “freedom to teach science.” 1 App. E-429. His testimony was given at the time the legislature was deciding whether to delete this section of the Act. 590 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. tradiction to the materialistic trends of our times, the perpetuation of our institutions and the teaching of literature.” 374 U. S., at 223. As in Stone and Abington, we need not be blind in this case to the legislature’s preeminent religious purpose in enacting this statute. There is a historic and contemporaneous link between the teachings of certain religious denominations and the teaching of evolution.9 It was this link that concerned the Court in Epperson v. Arkansas, 393 U. S. 97 (1968), which also involved a facial challenge to a statute regulating the teaching of evolution. In that case, the Court reviewed an Arkansas statute that made it unlawful for an instructor to teach evolution or to use a textbook that referred to this scientific theory. Although the Arkansas antievolution law did not explicitly state its predominate religious purpose, the Court could not ignore that “[t]he statute was a product of the upsurge of ‘fundamentalist’ religious fervor” that has long viewed this particular scientific theory as contradicting the literal interpretation of the Bible. Id., at 98, 106-107.10 After reviewing the history of antievolution statutes, the Court determined that “there can be no doubt that the motivation for the [Arkansas] law was the same [as other antievolution statutes]: to suppress the teaching of a theory which, it was thought, ‘denied’ the divine creation of man.” Id., at 109. The Court found that there can be no legitimate 9See McLean v. Arkansas Bd. of Ed., 529 F. Supp. 1255, 1258-1264 (ED Ark. 1982) (reviewing historical and contemporary antagonisms between the theory of evolution and religious movements). 10 The Court evaluated the statute in light of a series of antievolution statutes adopted by state legislatures dating back to the Tennessee statute that was the focus of the celebrated Scopes trial in 1925. Epperson n. Arkansas, 393 U. S., at 98, 101, n. 8, and 109. The Court found the Arkansas statute comparable to this Tennessee “monkey law,” since both gave preference to “ ‘religious establishments which have as one of their tenets or dogmas the instantaneous creation of man.’” Id., at 103, n. 11 (quoting Scopes v. State, 154 Tenn. 105, 126, 289 S. W. 363, 369 (1927) (Chambliss, J., concurring)). EDWARDS v. AGUILLARD 591 578 Opinion of the Court state interest in protecting particular religions from scientific views “distasteful to them,” id., at 107 (citation omitted), and concluded “that the First Amendment does not permit the State to require that teaching and learning must be tailored to the principles or prohibitions of any religious sect or dogma,” id., at 106. These same historic and contemporaneous antagonisms between the teachings of certain religious denominations and the teaching of evolution are present in this case. The preeminent purpose of the Louisiana Legislature was clearly to advance the religious viewpoint that a supernatural being created humankind.11 The term “creation science” was defined as embracing this particular religious doctrine by those responsible for the passage of the Creationism Act. Senator Keith’s leading expert on creation science, Edward Boudreaux, testified at the legislative hearings that the theory of creation science included belief in the existence of a supernatural creator. See 1 App. E-421—E-422 (noting that “creation scientists” point to high probability that life was “created by an intelligent mind”).12 Senator Keith also cited testimony from other experts to support the creation-science view that “a creator [was] responsible for the universe and everything in it.”13 2 App. E-497. The legislative history 11 While the belief in the instantaneous creation of humankind by a supernatural creator may require the rejection of every aspect of the theory of evolution, an individual instead may choose to accept some or all of this scientific theory as compatible with his or her spiritual outlook. See Tr. of Oral Arg. 23-29. 12 Boudreaux repeatedly defined creation science in terms of a theory that supports the existence of a supernatural creator. See, e. g., 2 App. E-501—E-502 (equating creation science with a theory pointing “to conditions of a creator”); 1 App. E-153—E-154 (“Creation . . . requires the direct involvement of a supernatural intelligence”). The lead witness at the hearings introducing the original bill, Luther Sunderland, described creation science as postulating “that everything was created by some intelligence or power external to the universe.” Id., at E-9—E-10. 13 Senator Keith believed that creation science embodied this view: “One concept is that a creator however you define a creator was responsible for 592 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. therefore reveals that the term “creation science,” as contemplated by the legislature that adopted this Act, embodies the religious belief that a supernatural creator was responsible for the creation of humankind. Furthermore, it is not happenstance that the legislature required the teaching of a theory that coincided with this religious view. The legislative history documents that the Act’s primary purpose was to change the science curriculum of public schools in order to provide persuasive advantage to a particular religious doctrine that rejects the factual basis of evolution in its entirety. The sponsor of the Creationism Act, Senator Keith, explained during the legislative hearings that his disdain for the theory of evolution resulted from the support that evolution supplied to views contrary to his own religious beliefs. According to Senator Keith, the theory of evolution was consonant with the “cardinal principiéis] of religious humanism, secular humanism, theological liberalism, aetheistism [sic]” 1 App. E-312—E-313; see also 2 App. E-499—E-500. The state senator repeatedly stated that scientific evidence supporting his religious views should be included in the public school curriculum to redress the fact that the theory of evolution incidentally coincided with what he characterized as religious beliefs antithetical to his own.14 everything that is in this world. The other concept is that it just evolved. ” Id., at E-280. Besides Senator Keith, several of the most vocal legislators also revealed their religious motives for supporting the bill in the official legislative history. See, e. g., id., at E-441, E-443 (Sen. Saunders noting that bill was amended so that teachers could refer to the Bible and other religious texts to support the creation-science theory); 2 App. E-561 — E-562, E-610 (Rep. Jenkins contending that the existence of God was a scientific fact). 14 See, e. g., 1 App. E-74—E-75 (noting that evolution is contrary to his family’s religious beliefs); id., at E-313 (contending that evolution advances religions contrary to his own); id., at E-357 (stating that evolution is “almost a religion” to science teachers); id., at E-418 (arguing that evolution is cornerstone of some religions contrary to his own); 2 App. E-763— E-764 (author of model bill, from which Act is derived, sent copy of the model bill to Senator Keith and advised that “I view this whole battle as EDWARDS v. AGUILLARD 593 578 Opinion of the Court The legislation therefore sought to alter the science curriculum to reflect endorsement of a religious view that is antagonistic to the theory of evolution. In this case, the purpose of the Creationism Act was to restructure the science curriculum to conform with a particular religious viewpoint. Out of many possible science subjects taught in the public schools, the legislature chose to affect the teaching of the one scientific theory that historically has been opposed by certain religious sects. As in Epperson, the legislature passed the Act to give preference to those religious groups which have as one of their tenets the creation of humankind by a divine creator. The “overriding fact” that confronted the Court in Epperson was “that Arkansas’ law selects from the body of knowledge a particular segment which it proscribes for the sole reason that it is deemed to conflict with ... a particular interpretation of the Book of Genesis by a particular religious group.” 393 U. S., at 103. Similarly, the Creationism Act is designed either to promote the theory of creation science which embodies a particular religious tenet by requiring that creation science be taught whenever evolution is taught or to prohibit the teaching of a scientific theory disfavored by certain religious sects by forbidding the teaching of evolution when creation science is not also taught. The Establishment Clause, however, “forbids alike the preference of a religious doctrine or the prohibition of theory which is deemed antagonistic to a particular dogma.” Id., at 106-107 (emphasis added). Because the primary purpose of the Creationism Act is to advance a particular religious belief, the Act endorses religion in violation of the First Amendment. We do not imply that a legislature could never require that scientific critiques of prevailing scientific theories be taught. Indeed, the Court acknowledged in Stone that its decision one between God and anti-God forces .... [I]f evolution is permitted to continue ... it will continue to be made to appear that a Supreme Being is unnecessary . . .”). 594 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. forbidding the posting of the Ten Commandments did not mean that no use could ever be made of the Ten Commandments, or that the Ten Commandments played an exclusively religious role in the history of Western Civilization. 449 U. S., at 42. In a similar way, teaching a variety of scientific theories about the origins of humankind to schoolchildren might be validly done with the clear secular intent of enhancing the effectiveness of science instruction. But because the primary purpose of the Creationism Act is to endorse a particular religious doctrine, the Act furthers religion in violation of the Establishment Clause.15 IV Appellants contend that genuine issues of material fact remain in dispute, and therefore the District Court erred in granting summary judgment. Federal Rule of Civil Procedure 56(c) provides that summary judgment “shall be rendered forthwith if the pleadings, depositions, answers to interrogatories, and admissions on file, together with the affidavits, if any, show that there is no genuine issue as to any material fact and that the moving party is entitled to a judgment as a matter of law.” A court’s finding of improper purpose behind a statute is appropriately determined by the statute on its face, its legislative history, or its interpretation by a responsible administrative agency. See, e. g., Wallace v. Jaffree, 472 U. S., at 56-61; Stone v. Graham, 449 U. S., at 41-42; Epperson v. Arkansas, 393 U. S., at 103-109. The plain meaning of the statute’s words, enlightened by their context and the contemporaneous legislative history, can control the determination of legislative purpose. See Wallace v. Jaffree, supra, at 74 (O’Connor, J., concurring in judgment); Richards n. United States, 369 U. S. 1, 9 (1962); Jay 15 Neither the District Court nor the Court of Appeals found a clear secular purpose, while both agreed that the Creationism Act’s primary purpose was to advance religion. “When both courts below are unable to discern an arguably valid secular purpose, this Court normally should hesitate to find one.” Wallace v. Jaffree, 472 U. S., at 66 (Powell, J., concurring). EDWARDS v. AGUILLARD 595 578 Opinion of the Court v. Boyd, 351 U. S. 345, 357 (1956). Moreover, in determining the legislative purpose of a statute, the Court has also considered the historical context of the statute, e. g., Epperson n. Arkansas, supra, and the specific sequence of events leading to passage of the statute, e. g., Arlington Heights v. Metropolitan Housing Dev. Corp., 429 U. S. 252 (1977). In this case, appellees’ motion for summary judgment rested on the plain language of the Creationism Act, the legislative history and historical context of the Act, the specific sequence of events leading to the passage of the Act, the State Board’s report on a survey of school superintendents, and the correspondence between the Act’s legislative sponsor and its key witnesses. Appellants contend that affidavits made by two scientists, two theologians, and an education administrator raise a genuine issue of material fact and that summary judgment was therefore barred. The affidavits define creation science as “origin through abrupt appearance in complex form” and allege that such a viewpoint constitutes a true scientific theory. See App. to Brief for Appellants A-7 to A-40. We agree with the lower courts that these affidavits do not raise a genuine issue of material fact. The existence of “uncontroverted affidavits” does not bar summary judgment.16 Moreover, the postenactment testimony of outside experts is of little use in determining the Louisiana Legislature’s purpose in enacting this statute. The Louisiana Legislature did hear and rely on scientific experts in passing the bill,17 but none of the persons making the affidavits produced by the ap 16 There is “no express or implied requirement in Rule 56 that the moving party support its motion with affidavits or other similar materials negating the opponent’s claim.” Celotex Corp. v. Catrett, 477 U. S. 317, 323 (1986) (emphasis in original). 17 The experts, who were relied upon by the sponsor of the bill and the legislation’s other supporters, testified that creation science embodies the religious view that there is a supernatural creator of the universe. See, supra, at 591-592. 596 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. pellants participated in or contributed to the enactment of the law or its implementation.18 The District Court, in its discretion, properly concluded that a Monday-morning “battle of the experts” over possible technical meanings of terms in the statute would not illuminate the contemporaneous purpose of the Louisiana Legislature when it made the law.19 We therefore conclude that the District Court did not err in finding that appellants failed to raise a genuine issue of material fact, and in granting summary judgment.20 V The Louisiana Creationism Act advances a religious doctrine by requiring either the banishment of the theory of evolution from public school classrooms or the presentation of a religious viewpoint that rejects evolution in its entirety. 18 Appellants contend that the affidavits are relevant because the term “creation science” is a technical term similar to that found in statutes that regulate certain scientific or technological developments. Even assuming, arguendo, that “creation science” is a term of art as represented by appellants, the definition provided by the relevant agency provides a better insight than the affidavits submitted by appellants in this case. In a 1981 survey conducted by the Louisiana Department of Education, the school superintendents in charge of implementing the provisions of the Creationism Act were asked to interpret the meaning of “creation science” as used in the statute. About 75 percent of Louisiana’s superintendents stated that they understood “creation science” to be a religious doctrine. 2 App. E-798—E-799. Of this group, the largest proportion of superintendents interpreted creation science, as defined by the Act, to mean the literal interpretation of the Book of Genesis. The remaining superintendents believed that the Act required teaching the view that “the universe was made by a creator.” Id., at E-799. 19 The Court has previously found the postenactment elucidation of the meaning of a statute to be of little relevance in determining the intent of the legislature contemporaneous to the passage of the statute. See Wallace y. Jaffree, 472 U. S., at 57, n. 45; id., at 75 (O’Connor, J., concurring in judgment). 20 Numerous other Establishment Clause cases that found state statutes to be unconstitutional have been disposed of without trial. E. g., Larkin v. GrendeVs Den, Inc., 459 U. S. 116 (1982); Lemon v. Kurtzman, 403 U. S. 602 (1971); Engel n. Vitale, 370 U. S. 421 (1962). EDWARDS v. AGUILLARD 597 578 Powell, J., concurring The Act violates the Establishment Clause of the First Amendment because it seeks to employ the symbolic and financial support of government to achieve a religious purpose. The judgment of the Court of Appeals therefore is Affirmed. Justice Powell, with whom Justice O’Connor joins, concurring. I write separately to note certain aspects of the legislative history, and to emphasize that nothing in the Court’s opinion diminishes the traditionally broad discretion accorded state and local school officials in the selection of the public school curriculum. I This Court consistently has applied the three-pronged test of Lemon n. Kurtzman, 403 U. S. 602 (1971), to determine whether a particular state action violates the Establishment Clause of the Constitution.1 See, e. g., Grand Rapids School Dist. v. Ball, 473 U. S. 373, 383 (1985) (“We have particularly relied on Lemon in every case involving the sensitive relationship between government and religion in the education of our children”). The first requirement of the Lemon test is that the challenged statute have a “secular legislative purpose.” Lemon n. Kurtzman, supra, at 612. See Committee for Public Education & Religious Liberty v. Nyquist, 413 U. S. 756, 773 (1973). If no valid secular purpose can be identified, then the statute violates the Establishment Clause. A “The starting point in every case involving construction of a statute is the language itself.” Blue Chip Stamps v. Manor Drug Stores, 421 U. S. 723, 756 (1975) (Powell, J., ’As the Court recognizes, ante, at 583, n. 4, the one exception to this consistent application of the Lemon test is Marsh n. Chambers, 463 U. S. 783 (1983). 598 OCTOBER TERM, 1986 Powell, J., concurring 482 U. S. concurring). The Balanced Treatment for Creation-Science and Evolution-Science Act (Act or Balanced Treatment Act), La. Rev. Stat. Ann. §17:286.1 et seq. (West 1982), provides in part: “[P]ublic schools within [the] state shall give balanced treatment to creation-science and to evolution-science. Balanced treatment of these two models shall be given in classroom lectures taken as a whole for each course, in textbook materials taken as a whole for each course, in library materials taken as a whole for the sciences and taken as a whole for the humanities, and in other educational programs in public schools, to the extent that such lectures, textbooks, library materials, or educational programs deal in any way with the subject of the origin of man, life, the earth, or the universe. When creation or evolution is taught, each shall be taught as a theory, rather than as proven scientific fact.” § 17:286.4(A). “Balanced treatment” means “providing whatever information and instruction in both creation and evolution models the classroom teacher determines is necessary and appropriate to provide insight into both theories in view of the textbooks and other instructional materials available for use in his classroom.” § 17:286.3(1). “Creation-science” is defined as “the scientific evidences for creation and inferences from those scientific evidences.” §17:286.3(2). “Evolution-science” means “the scientific evidences for evolution and inferences from those scientific evidences.” §17:286.3(3). Although the Act requires the teaching of the scientific evidences of both creation and evolution whenever either is taught, it does not define either term. “A fundamental canon of statutory construction is that, unless otherwise defined, words will be interpreted as taking their ordinary, contemporary, common meaning.” Perrin n. United States, 444 U. S. 37, 42 (1979). The “doctrine or theory of creation” is commonly defined as “holding that matter, the various forms of life, and the world were created by a transcendent God out EDWARDS v. AGUILLARD 599 578 Powell, J., concurring of nothing.” Webster’s Third New International Dictionary 532 (unabridged 1981). “Evolution” is defined as “the theory that the various types of animals and plants have their origin in other preexisting types, the distinguishable differences being due to modifications in successive generations.” Id., at 789. Thus, the Balanced Treatment Act mandates that public schools present the scientific evidence to support a theory of divine creation whenever they present the scientific evidence to support the theory of evolution. “[C]oncepts concerning God or a supreme being of some sort are manifestly religious .... These concepts do not shed that religiosity merely because they are presented as a philosophy or as a science.” Malnak v. Yogi, 440 F. Supp. 1284, 1322 (NJ 1977), aff’d per curiam, 592 F. 2d 197 (CA3 1979). From the face of the statute, a purpose to advance a religious belief is apparent. A religious purpose alone is not enough to invalidate an act of a state legislature. The religious purpose must predominate. See Wallace n. Jaffree, 472 U. S. 38, 56 (1985); id., at 64 (Powell, J., concurring); Lynch v. Donnelly, 465 U. S. 668, 681, n. 6 (1984). The Act contains a statement of purpose: to “protec[t] academic freedom.” § 17:286.2. This statement is puzzling. Of course, the “academic freedom” of teachers to present information in public schools, and students to receive it, is broad. But it necessarily is circumscribed by the Establishment Clause. “Academic freedom” does not encompass the right of a legislature to structure the public school curriculum in order to advance a particular religious belief. Epperson v. Arkansas, 393 U. S. 97, 106 (1968). Nevertheless, I read this statement in the Act as rendering the purpose of the statute at least ambiguous. Accordingly, I proceed to review the legislative history of the Act. B In June 1980, Senator Bill Keith introduced Senate Bill 956 in the Louisiana Legislature. The stated purpose of the bill 600 OCTOBER TERM, 1986 Powell, J., concurring 482 U. S. was to “assure academic freedom by requiring the teaching of the theory of creation ex nihilo in all public schools where the theory of evolution is taught.” 1 App. E-l.2 The bill defined the “theory of creation ex nihilo” as “the belief that the origin of the elements, the galaxy, the solar system, of life, of all the species of plants and animals, the origin of man, and the origin of all things and their processes and relationships were created ex nihilo and fixed by God.” Id., at E-la— E-lb. This theory was referred to by Senator Keith as “scientific creationism.” Id., at E-2. While a Senate committee was studying scientific creationism, Senator Keith introduced a second draft of the bill, requiring balanced treatment of “evolution-science” and “creation-science.” Id., at E-108. Although the Keith bill prohibited “instruction in any religious doctrine or materials,” id., at E-302, it defined “creation-science” to include “the scientific evidences and related inferences that indicate (a) sudden creation of the universe, energy, and life from nothing; (b) the insufficiency of mutation and natural selection in bringing about development of all living kinds from a single organism; (c) changes only within fixed limits or originally created kinds of plants and animals; (d) separate ancestry for man and apes; (e) explanation of the earth’s geology by catastrophism, including the occurrence of a worldwide flood; and (f) a 2 Creation “ex nihilo” means creation “from nothing” and has been found to be an “inherently religious concept.” McLean n. Arkansas Board of Education, 529 F. Supp. 1255, 1266 (ED Ark. 1982). The District Court in McLean found: “The argument that creation from nothing in [§] 4(a)(1) [of the substantially similar Arkansas Balanced Treatment Act] does not involve a supernatural deity has no evidentiary or rational support. To the contrary, ‘creation out of nothing’ is a concept unique to Western religions. In traditional Western religious thought, the conception of a creator of the world is a conception of God. Indeed, creation of the world ‘out of nothing’ is the ultimate religious statement because God is the only actor.” Id., at 1265. EDWARDS v. AGUILLARD 601 578 Powell, J., concurring relatively recent inception of the earth and living kinds.” Id., at E-298—E-299. Significantly, the model Act on which the Keith bill relied was also the basis for a similar statute in Arkansas. See McLean v. Arkansas Board of Education, 529 F. Supp. 1255 (ED Ark. 1982). The District Court in McLean carefully examined this model Act, particularly the section defining creation science, and concluded that “[b]oth [its] concepts and wording. . . convey an inescapable religiosity.” Id., at 1265. The court found that “[t]he ideas of [this section] are not merely similar to the literal interpretation of Genesis; they are identical and parallel to no other story of creation.” Ibid. The complaint in McLean was filed on May 27, 1981. On May 28, the Louisiana Senate committee amended the Keith bill to delete the illustrative list of scientific evidences. According to the legislator who proposed the amendment, it was “not intended to try to gut [the bill] in any way, or defeat the purpose [for] which Senator Keith introduced [it],” 1 App. E-432, and was not viewed as working “any violence to the bill.” Id., at E-438. Instead, the concern was “whether this should be an all inclusive list.” Ibid. The legislature then held hearings on the amended bill that became the Balanced Treatment Act under review. The principal creation scientist to testify in support of the Act was Dr. Edward Boudreaux. He did not elaborate on the nature of creation science except to indicate that the “scientific evidences” of the theory are “the objective information of science [that] point[s] to conditions of a creator.” 2 id., at E-501—E-502. He further testified that the recognized creation scientists in the United States, who “numbe[r] something like a thousand [and] who hold doctorate and masters degrees in all areas of science,” are affiliated with either or both the Institute for Creation Research and the Creation Research Society. Id., at E-503—E-504. Information on both of these organizations is part of the legislative history, 602 OCTOBER TERM, 1986 Powell, J., concurring 482 U. S. and a review of their goals and activities sheds light on the nature of creation science as it was presented to, and understood by, the Louisiana Legislature. The Institute for Creation Research is an affiliate of the Christian Heritage College in San Diego, California. The Institute was established to address the “urgent need for our nation to return to belief in a personal, omnipotent Creator, who has a purpose for His creation and to whom all people must eventually give account.” 1 id., at E-197. A goal of the Institute is “a revival of belief in special creation as the true explanation of the origin of the world.” Therefore, the Institute currently is working on the “development of new methods for teaching scientific creationism in public schools.” Id., at E-197—E-199. The Creation Research Society (CRS) is located in Ann Arbor, Michigan. A member must subscribe to the following statement of belief: “The Bible is the written word of God, and because it is inspired throughout, all of its assertions are historically and scientifically true.” 2 id., at E-583. To study creation science at the CRS; a member must accept “that the account of origins in Genesis is a factual presentation of simple historical truth.” Ibid.3 3 The District Court in McLean noted three other elements of the CRS statement of belief to which members must subscribe: “ ‘[i] All basic types of living things, including man, were made by direct creative acts of God during Creation Week as described in Genesis. Whatever biological changes have occurred since Creation have accomplished only changes within the original created kinds, [ii] The great Flood described in Genesis, commonly referred to as the Noachian Deluge, was an historical event, world-wide in its extent and effect, [iii] Finally, we are an organization of Christian men of science, who accept Jesus Christ as our Lord and Savior. The account of the special creation of Adam and Eve as one man and one woman, and their subsequent Fall into sin, is the basis for our belief in the necessity of a Savior for all mankind. Therefore, salvation can come only thru (sic) accepting Jesus Christ as our Savior.’ ” 529 F. Supp., at 1260, n. 7. EDWARDS v. AGUILLARD 603 578 Powell, J., concurring C When, as here, “both courts below are unable to discern an arguably valid secular purpose, this Court normally should hesitate to find one.” Wallace v. Jaffree, 472 U. S., at 66 (Powell, J., concurring). My examination of the language and the legislative history of the Balanced Treatment Act confirms that the intent of the Louisiana Legislature was to promote a particular religious belief. The legislative history of the Arkansas statute prohibiting the teaching of evolution examined in Epperson v. Arkansas, 393 U. S. 97 (1968), was strikingly similar to the legislative history of the Balanced Treatment Act. In Epperson, the Court found: “It is clear that fundamentalist sectarian conviction was and is the law’s reason for existence. Its antecedent, Tennessee’s ‘monkey law,’ candidly stated its purpose: to make it unlawful ‘to teach any theory that denies the story of the Divine Creation of man as taught in the Bible, and to teach instead that man has descended from a lower order of animals.’ Perhaps the sensational publicity attendant upon the Scopes trial induced Arkansas to adopt less explicit language. It eliminated Tennessee’s reference to ‘the story of the Divine creation of man’ as taught in the Bible, but there is no doubt that the motivation for the law was the same: to suppress the teaching of a theory which, it was thought, ‘denied’ the divine creation of man.” Id., at 107-109 (footnotes omitted). Here, it is clear that religious belief is the Balanced Treatment Act’s “reason for existence.” The tenets of creation science parallel the Genesis story of creation,4 and this is a 4 After hearing testimony from numerous experts, the District Court in McLean concluded that “[t]he parallels between [the definition section of the model Act] and Genesis are quite specific.” Id., at 1265, n. 19. It found the concepts of “sudden creation from nothing,” a worldwide flood of divine origin, and “kinds” to be derived from Genesis; “relatively recent inception” to mean “an age of the earth from 6,000 to 10,000 years” and to 604 OCTOBER TERM, 1986 Powell, J., concurring 482 U. S. religious belief. “[N]o legislative recitation of a supposed secular purpose can blind us to that fact.” Stone v. Graham, 449 U. S. 39, 41 (1980). Although the Act as finally enacted does not contain explicit reference to its religious purpose, there is no indication in the legislative history that the deletion of “creation ex nihilo” and the four primary tenets of the theory was intended to alter the purpose of teaching creation science. Instead, the statements of purpose of the sources of creation science in the United States make clear that their purpose is to promote a religious belief. I find no persuasive evidence in the legislative history that the legislature’s purpose was any different. The fact that the Louisiana Legislature purported to add information to the school curriculum rather than detract from it as in Epperson does not affect my analysis. Both legislatures acted with the unconstitutional purpose of structuring the public school curriculum to make it compatible with a particular religious belief: the “divine creation of man.” That the statute is limited to the scientific evidences supporting the theory does not render its purpose secular. In reaching its conclusion that the Act is unconstitutional, the Court of Appeals “[did] not deny that the underpinnings of creationism may be supported by scientific evidence.” 765 F. 2d 1251, 1256 (1985). And there is no need to do so. Whatever the academic merit of particular subjects or theories, the Establishment Clause limits the discretion of state officials to pick and choose among them for the purpose of promoting a particular religious belief. The language of the statute and its legislative history convince me that the Louisiana Legislature exercised its discretion for this purpose in this case. be based “on the genealogy of the Old Testament using the rather astronomical ages assigned to the patriarchs”; and the “separate ancestry of man and ape” to focus on “the portion of the theory of evolution which Fundamentalists find most offensive.” Ibid, (citing Epperson v. Arkansas, 393 U. S. 97 (1968)). EDWARDS v. AGUILLARD 605 578 Powell, J., concurring II Even though I find Louisiana’s Balanced Treatment Act unconstitutional, I adhere to the view “that the States and locally elected school boards should have the responsibility for determining the educational policy of the public schools.” Board of Education, Island Trees Union Free School Dist. No. 26n. Pico, 457 U. S. 853, 893 (1982) (Powell, J., dissenting). A decision respecting the subj ect matter to be taught in public schools does not violate the Establishment Clause simply because the material to be taught “ ‘happens to coincide or harmonize with the tenets of some or all religions.’ ” Harris n. McRae, 448 U. S. 297, 319 (1980) (quoting McGowan v. Maryland, 366 U. S. 420, 442 (1961)). In the context of a challenge under the Establishment Clause, interference with the decisions of these authorities is warranted only when the purpose for their decisions is clearly religious. The history of the Religion Clauses of the First Amendment has been chronicled by this Court in detail. See, e. g., Everson v. Board of Education, 330 U. S. 1, 8-14 (1947); Engel v. Vitale, 370 U. S. 421, 425-430 (1962); McGowan v. Maryland, supra, at 437-442. Therefore, only a brief review at this point may be appropriate. The early settlers came to this country from Europe to escape religious persecution that took the form of forced support of state-established churches. The new Americans thus reacted strongly when they perceived the same type of religious intolerance emerging in this country. The reaction in Virginia, the home of many of the Founding Fathers, is instructive. George Mason’s draft of the Virginia Declaration of Rights was adopted by the House of Burgesses in 1776. Because of James Madison’s influence, the Declaration of Rights embodied the guarantee of free exercise of religion, as opposed to toleration. Eight years later, a provision prohibiting the establishment of religion became a part of Virginia law when James Madison’s Memorial and Remonstrance against Re 606 OCTOBER TERM, 1986 Powell, J., concurring 482 U. S. ligious Assessments, written in response to a proposal that all Virginia citizens be taxed to support the teaching of the Christian religion, spurred the legislature to consider and adopt Thomas Jefferson’s Bill for Establishing Religious Freedom. See Committee for Public Education & Religious Liberty n. Nyquist, 413 U. S., at 770, n. 28. Both the guarantees of free exercise and against the establishment of religion were then incorporated into the Federal Bill of Rights by its drafter, James Madison. While the “meaning and scope of the First Amendment” must be read “in light of its history and the evils it was designed forever to suppress,” Everson v. Board of Education, supra, at 14-15, this Court has also recognized that “this Nation’s history has not been one of entirely sanitized separation between Church and State.” Committee for Public Education & Religious Liberty v. Nyquist, supra, at 760. “The fact that the Founding Fathers believed devotedly that there was a God and that the unalienable rights of man were rooted in Him is clearly evidenced in their writings, from the Mayflower Compact to the Constitution itself.” Abington School District n. Schempp, 374 U. S. 203, 213 (1963).5 The Court properly has noted “an unbroken history of official acknowledgment... of the role of religion in American life.” Lynch v. Donnelly, 465 U. S., at 674, and has recognized that these references to “our religious heritage” are constitutionally acceptable. Id., at 677. As a matter of history, schoolchildren can and should properly be informed of all aspects of this Nation’s religious heritage. I would see no constitutional problem if schoolchildren were taught the nature of the Founding Father’s religious beliefs and how these beliefs affected the attitudes 5 John Adams wrote to Thomas Jefferson: “[T]he Bible is the best book in the world. It contains more of my little philosophy than all the libraries I have seen; and such parts of it as I cannot reconcile to my little philosophy, I postpone for future investigation.” Letter of Dec. 25, 1813, 10 Works of John Adams 85 (1856). EDWARDS v. AGUILLARD 607 578 Powell, J., concurring of the times and the structure of our government.6 Courses in comparative religion of course are customary and constitutionally appropriate.7 In fact, since religion permeates our history, a familiarity with the nature of religious beliefs is necessary to understand many historical as well as contemporary events.8 In addition, it is worth noting that the Estab- 6 There is an enormous variety of religions in the United States. The Encyclopedia of American Religions (2d ed. 1987) describes 1,347 religious organizations. The United States Census Bureau groups the major American religions into: Buddhist Churches of America; Eastern Churches; Jews; Old Catholic, Polish National Catholic, and Armenian Churches; The Roman Catholic Church; Protestants; and Miscellaneous. Statistical Abstract of the United States 50 (106th ed. 1986). Our country has become strikingly multireligious as well as multiracial and multiethnic. This fact, perhaps more than anything one could write, demonstrates the wisdom of including the Establishment Clause in the First Amendment. States’ proposals for what became the Establishment Clause evidence the goal of accommodating competing religious beliefs. See, e. g., New York’s Resolutionof Ratification reprinted in 2 Documentary History of the Constitution 190, 191 (1894) (“[N]o Religious Sect or Society ought to be favoured or established by Law in preference of others”). 7 State-sponsored universities in Louisiana already offer courses integrating religious studies into the curriculum. Approximately half of the state-sponsored universities offer one or more courses involving religion. As an example, Louisiana State University at Baton Rouge offers seven courses: Introduction to Religion, Old Testament, New Testament, Faith and Doubt, Jesus in History and Tradition, Eastern Religions, and Philosophy of Religion. Of course, the difference in maturity between college-age and secondary students may affect the constitutional analysis of a particular public school policy. See Widmar v. Vincent, 454 U. S. 263, 274, n. 14 (1981). Nevertheless, many general teaching guides suggest that education as to the nature of various religious beliefs could be integrated into a secondary school curriculum in a manner consistent with the Constitution. See, e. g., C. Kniker, Teaching about Religion in Public Schools (1985); Religion in Elementary Social Studies Project, Final Report (Fla. State Univ. 1976); L. Karp, Teaching the Bible as Literature in Public Schools (1973). 8 For example, the political controversies in Northern Ireland, the Middle East, and India cannot be understood properly without reference to the underlying religious beliefs and the conflicts they tend to generate. 608 OCTOBER TERM, 1986 White, J., concurring in judgment 482 U. S. lishment Clause does not prohibit per se the educational use of religious documents in public school education. Although this Court has recognized that the Bible is “an instrument of religion,” Abington School District v. Schempp, supra, at 224, it also has made clear that the Bible “may constitutionally be used in an appropriate study of history, civilization, ethics, comparative religion, or the like.” Stone v. Graham, 449 U. S., at 42 (citing Abington School District v. Schempp, supra, at 225). The book is, in fact, “the world’s all-time best seller”9 with undoubted literary and historic value apart from its religious content. The Establishment Clause is properly understood to prohibit the use of the Bible and other religious documents in public school education only when the purpose of the use is to advance a particular religious belief. Ill In sum, I find that the language and the legislative history of the Balanced Treatment Act unquestionably demonstrate that its purpose is to advance a particular religious belief. Although the discretion of state and local authorities over public school curricula is broad, “the First Amendment does not permit the State to require that teaching and learning must be tailored to the principles or prohibitions of any religious sect or dogma.” Epperson n. Arkansas, 393 U. S., at 106. Accordingly, I concur in the opinion of the Court and its judgment that the Balanced Treatment Act violates the Establishment Clause of the Constitution. Justice White, concurring in the judgment. As it comes to us, this is not a difficult case. Based on the historical setting and plain language of the Act both courts construed the statutory words “creation science” to refer to a religious belief, which the Act required to be taught if evolu 9 See N. Y. Times, May 10, 1981, section 2, p. 24, col. 3; N. McWhirter, 1986 Guiness Book of World Records 144 (the Bible is the world’s most widely distributed book). EDWARDS v. AGUILLARD 609 578 White, J., concurring in judgment tion was taught. In other words, the teaching of evolution was conditioned on the teaching of a religious belief. Both courts concluded that the state legislature’s primary purpose was to advance religion and that the statute was therefore unconstitutional under the Establishment Clause. We usually defer to courts of appeals on the meaning of a state statute, especially when a district court has the same view. Of course, we have the power to disagree, and the lower courts in a particular case may be plainly wrong. But if the meaning ascribed to a state statute by a court of appeals is a rational construction of the statute, we normally accept it. Brockett v. Spokane Arcades, Inc., 472 U. S. 491, 499-500 (1985); Chardon v. Fumero Soto, 462 U. S. 650, 654-655, n. 5 (1983); Haring n. Prosise, 462 U. S. 306, 314, n. 8 (1983); Pierson v. Ray, 386 U. S. 547, 558, n. 12 (1967); General Box Co. v. United States, 351 U. S. 159, 165 (1956). We do so because we believe “that district courts and courts of appeals are better schooled in and more able to interpret the laws of their respective States.” Brockett v. Spokane Arcades, supra, at 500. Brockett also indicates that the usual rule applies in First Amendment cases. Here, the District Judge, relying on the terms of the Act, discerned its purpose to be the furtherance of a religious belief, and a panel of the Court of Appeals agreed. Of those four judges, two are Louisianians. I would accept this view of the statute. Even if as an original matter I might have arrived at a different conclusion based on a reading of the statute and the record before us, I cannot say that the two courts below are so plainly wrong that they should be reversed. Rehearing en banc was denied by an 8-7 vote, the dissenters expressing their disagreement with the panel decision. The disagreement, however, was over the construction of the Louisiana statute, particularly the assessment of its purpose, and offers no justification for departing from the usual rule counseling against de novo constructions of state statutes. 610 OCTOBER TERM, 1986 Scalia, J., dissenting 482 U. S. If the Court of Appeals’ construction is to be accepted, so is its conclusion that under our prior cases the Balanced Treatment Act is unconstitutional because its primary purpose is to further a religious belief by imposing certain requirements on the school curriculum. Unless, therefore, we are to reconsider the Court’s decisions interpreting the Establishment Clause, I agree that the judgment of the Court of Appeals must be affirmed. Justice Scalia, with whom The Chief Justice joins, dissenting. Even if I agreed with the questionable premise that legislation can be invalidated under the Establishment Clause on the basis of its motivation alone, without regard to its effects, I would still find no justification for today’s decision. The Louisiana legislators who passed the “Balanced Treatment for Creation-Science and Evolution-Science Act” (Balanced Treatment Act), La. Rev. Stat. Ann. §§17:286.1-17:286.7 (West 1982), each of whom had sworn to support the Constitution,1 were well aware of the potential Establishment Clause problems and considered that aspect of the legislation with great care. After seven hearings and several months of study, resulting in substantial revision of the original proposal, they approved the Act overwhelmingly and specifically articulated the secular purpose they meant it to serve. Although the record contains abundant evidence of the sincerity of that purpose (the only issue pertinent to this case), the Court today holds, essentially on the basis of “its visceral knowledge regarding what must have motivated the legislators,” 778 F. 2d 225, 227 (CA5 1985) (Gee, J., dissenting) (emphasis added), that the members of the Louisiana Legislature knowingly violated their oaths and then lied about it. I dissent. Had requirements of the Balanced Treatment Act that 1 Article VI, cl. 3, of the Constitution provides that “the Members of the several State Legislatures . . . shall be bound by Oath or Affirmation, to support this Constitution.” EDWARDS v. AGUILLARD 611 578 Scalia, J., dissenting are not apparent on its face been clarified by an interpretation of the Louisiana Supreme Court, or by the manner of its implementation, the Act might well be found unconstitutional; but the question of its constitutionality cannot rightly be disposed of on the gallop, by impugning the motives of its supporters. I This case arrives here in the following posture: The Louisiana Supreme Court has never been given an opportunity to interpret the Balanced Treatment Act, State officials have never attempted to implement it, and it has never been the subject of a full evidentiary hearing. We can only guess at its meaning. We know that it forbids instruction in either “creation-science” or “evolution-science” without instruction in the other, § 17:286.4A, but the parties are sharply divided over what creation science consists of. Appellants insist that it is a collection of educationally valuable scientific data that has been censored from classrooms by an embarrassed scientific establishment. Appellees insist it is not science at all but thinly veiled religious doctrine. Both interpretations of the intended meaning of that phrase find considerable support in the legislative history. At least at this stage in the litigation, it is plain to me that we must accept appellants’ view of what the statute means. To begin with, the statute itself defines “creation-science” as “the scientific evidences for creation and inferences from those scientific evidences.” §17:286.3(2) (emphasis added). If, however, that definition is not thought sufficiently helpful, the means by which the Louisiana Supreme Court will give the term more precise content is quite clear—and again, at this stage in the litigation, favors the appellants’ view. “Creation science” is unquestionably a “term of art,” see Brief for 72 Nobel Laureates et al. as Amici Curiae 20, and thus, under Louisiana law, is “to be interpreted according to [its] received meaning and acceptation with the learned in the art, trade or profession to which [it] refer[s].” La. Civ. 612 OCTOBER TERM, 1986 Scalia, J., dissenting 482 U. S. Code Ann., Art. 15 (West 1952).2 The only evidence in the record of the “received meaning and acceptation” of “creation science” is found in five affidavits filed by appellants. In those affidavits, two scientists, a philosopher, a theologian, and an educator, all of whom claim extensive knowledge of creation science, swear that it is essentially a collection of scientific data supporting the theory that the physical universe and life within it appeared suddenly and have not changed substantially since appearing. See App. to Juris. Statement A-19 (Kenyon); id., at A-36 (Morrow); id., at A-41 (Miethe). These experts insist that creation science is a strictly scientific concept that can be presented without religious reference. See id., at A-19—A-20, A-35 (Kenyon); id., at A-36—A-38 (Morrow); id., at A-40, A-41, A-43 (Miethe); id., at A-47, A-48 (Most); id., at A-49 (Clinkert). At this point, then, we must assume that the Balanced Treatment Act does not require the presentation of religious doctrine. Nothing in today’s opinion is plainly to the contrary, but what the statute means and what it requires are of rather little concern to the Court. Like the Court of Appeals, 765 F. 2d 1251, 1253, 1254 (CA5 1985), the Court finds it necessary to consider only the motives of the legislators who supported the Balanced Treatment Act, ante, at 586, 593-594, 596. After examining the statute, its legislative history, and its historical and social context, the Court holds that the Louisiana Legislature acted without “a secular legislative purpose” and that the Act therefore fails the “purpose” prong of the three-part test set forth in Lemon v. Kurtzman, 403 U. S. 602, 612 (1971). As I explain below, infra, at 636-640, 2 Thus the popular dictionary definitions cited by Justice Powell, ante, at 598-599 (concurring opinion), and appellees, see Brief for Appellees 25, 26; Tr. of Oral Arg. 32, 34, are utterly irrelevant, as are the views of the school superintendents cited by the majority, ante, at 595, n. 18. Three-quarters of those surveyed had “[n]o” or “[l]imited” knowledge of “creation-science theory,” and not a single superintendent claimed “[e]x-tensive” knowledge of the subject. 2 App. E-798. EDWARDS v. AGUILLARD 613 578 Scalia, J., dissenting I doubt whether that “purpose” requirement of Lemon is a proper interpretation of the Constitution; but even if it were, I could not agree with the Court’s assessment that the requirement was not satisfied here. This Court has said little about the first component of the Lemon test. Almost invariably, we have effortlessly discovered a secular purpose for measures challenged under the Establishment Clause, typically devoting no more than a sentence or two to the matter. See, e. g., Witters n. Washington Dept, of Services for Blind, 474 U. S. 481,485-486 (1986); Grand Rapids School District v. Ball, 473 U. S. 373, 383 (1985); Mueller v. Allen, 463 U. S. 388, 394-395 (1983); Larkin v. GrendeVs Den, Inc., 459 U. S. 116, 123-124 (1982); Widmar n. Vincent, 454 U. S. 263, 271 (1981); Committee for Public Education & Religious Liberty n. Regan, 444 U. S. 646, 654, 657 (1980); Wolman n. Walter, 433 U. S. 229, 236 (1977) (plurality opinion); Meek v. Pittenger, 421 U. S. 349, 363 (1975); Committee for Public Education & Religious Liberty v. Nyquist, 413 U. S. 756, 773 (1973); Levitt v. Committee for Public Education & Religious Liberty, 413 U. S. 472, 479-480, n. 7 (1973); Tilton v. Richardson, 403 U. S. 672, 678-679 (1971) (plurality opinion); Lemon v. Kurtzman, supra, at 613. In fact, only once before deciding Lemon, and twice since, have we invalidated a law for lack of a secular purpose. See Wallace v. Jaffree, 472 U. S. 38 (1985); Stone v. Graham, 449 U. S. 39 (1980) (per curiam); Epperson n. Arkansas, 393 U. S. 97 (1968). Nevertheless, a few principles have emerged from our cases, principles which should, but to an unfortunately large extent do not, guide the Court’s application of Lemon today. It is clear, first of all, that regardless of what “legislative purpose” may mean in other contexts, for the purpose of the Lemon test it means the “actual” motives of those responsible for the challenged action. The Court recognizes this, see ante, at 585, as it has in the past, see, e. g., Witters v. Washington Dept, of Services for Blind, supra, at 486; Wallace v. 614 OCTOBER TERM, 1986 Scalia, J., dissenting 482 U. S. Jaffree, supra, at 56. Thus, if those legislators who supported the Balanced Treatment Act in fact acted with a “sincere” secular purpose, ante, at 587, the Act survives the first component of the Lemon test, regardless of whether that purpose is likely to be achieved by the provisions they enacted. Our cases have also confirmed that when the Lemon Court referred to “a secular . . . purpose,” 403 U. S., at 612, it meant “a secular purpose.” The author of Lemon, writing for the Court, has said that invalidation under the purpose prong is appropriate when “there [is] no question that the statute or activity was motivated wholly by religious considerations.” Lynch v. Donnelly, 465 U. S. 668, 680 (1984) (Burger, C. J.) (emphasis added); see also id., at 681, n. 6; Wallace n. Jaffree, supra, at 56 (“[T]he First Amendment requires that a statute must be invalidated if it is entirely motivated by a purpose to advance religion”) (emphasis added; footnote omitted). In all three cases in which we struck down laws under the Establishment Clause for lack of a secular purpose, we found that the legislature’s sole motive was to promote religion. See Wallace v. Jaffree, supra, at 56, 57, 60; Stone v. Graham, supra, at 41, 43, n. 5; Epperson v. Arkansas, supra, at 103, 107-108; see also Lynch n. Donnelly, supra, at 680 (describing Stone and Epperson as cases in which we invalidated laws “motivated wholly by religious considerations”). Thus, the majority’s invalidation of the Balanced Treatment Act is defensible only if the record indicates that the Louisiana Legislature had no secular purpose. It is important to stress that the purpose forbidden by Lemon is the purpose to “advance religion.” 403 U. S., at 613; accord, ante, at 585 (“promote” religion); Witters v. Washington Dept, of Services for Blind, supra, at 486 (“endorse religion”); Wallace n. Jaffree, 472 U; S., at 56 (“advance religion”); ibid, (“endorse . . . religion”); Committee for Public Education & Religious Liberty v. Nyquist, supra, at 788 (“‘advancing’ . . . religion”); Levitt v. Committee for EDWARDS v. AGUILLARD 615 578 Scalia, J., dissenting Public Education & Religious Liberty, supra, at 481 (“advancing religion”); Walz n. Tax Common of New York City, 397 U. S. 664, 674 (1970) (“establishing, sponsoring, or supporting religion”); Board of Education n. Allen, 392 U. S. 236, 243 (1968) (“‘advancement or inhibition of religion’”) (quoting Abington School Dist. v. Schempp, 374 U. S. 203, 222 (1963)). Our cases in no way imply that the Establishment Clause forbids legislators merely to act upon their religious convictions. We surely would not strike down a law providing money to feed the hungry or shelter the homeless if it could be demonstrated that, but for the religious beliefs of the legislators, the funds would not have been approved. Also, political activism by the religiously motivated is part of our heritage. Notwithstanding the majority’s implication to the contrary, ante, at 589-591, we do not presume that the sole purpose of a law is to advance religion merely because it was supported strongly by organized religions or by adherents of particular faiths. See Walz v. Tax Common of New York City, supra, at 670; cf. Harris v. McRae, 448 U. S. 297, 319-320 (1980). To do so would deprive religious men and women of their right to participate in the political process. Today’s religious activism may give us the Balanced Treatment Act, but yesterday’s resulted in the abolition of slavery, and tomorrow’s may bring relief for famine victims. Similarly, we will not presume that a law’s purpose is to advance religion merely because it “‘happens to coincide or harmonize with the tenets of some or all religions,’” Harris v. McRae, supra, at 319 (quoting McGowan n. Maryland, 366 U. S. 420, 442 (1961)), or because it benefits religion, even substantially. We have, for example, turned back Establishment Clause challenges to restrictions on abortion funding, Harris v. McRae, supra, and to Sunday closing laws, McGowan v. Maryland, supra, despite the fact that both “agre[e] with the dictates of [some] Judaeo-Christian religions,” id., at 442. “In many instances, the Congress or state legislatures conclude that the general welfare of soci 616 OCTOBER TERM, 1986 Scalia, J., dissenting 482 U. S. ety, wholly apart from any religious considerations, demands such regulation.” Ibid. On many past occasions we have had no difficulty finding a secular purpose for governmental action far more likely to advance religion than the Balanced Treatment Act. See, e. g., Mueller v. Allen, 463 U. S., at 394-395 (tax deduction for expenses of religious education); Wolman n. Walter, 433 U. S., at 236 (plurality opinion) (aid to religious schools); Meek n. Pittenger, 421 U. S., at 363 (same); Committee for Public Education & Religious Liberty v. Nyquist, 413 U. S., at 773 (same); Lemon n. Kurtzman, 403 U. S., at 613 (same); Walz n. Tax Comm’n of New York City, supra, at 672 (tax exemption for church property); Board of Education v. Allen, supra, at 243 (textbook loans to students in religious schools). Thus, the fact that creation science coincides with the beliefs of certain religions, a fact upon which the majority relies heavily, does not itself justify invalidation of the Act. Finally, our cases indicate that even certain kinds of governmental actions undertaken with the specific intention of improving the position of religion do not “advance religion” as that term is used in Lemon. 403 U. S., at 613. Rather, we have said that in at least two circumstances government must act to advance religion, and that in a third it may do so. First, since we have consistently described the Establishment Clause as forbidding not only state action motivated by the desire to advance religion, but also that intended to “disapprove,” “inhibit,” or evince “hostility” toward religion, see, e. g., ante, at 585 (“‘disapprove’”) (quoting Lynch v. Donnelly, supra, at 690 (O’Connor, J., concurring)); Lynch v. Donnelly, supra, at 673 (“hostility”); Committee for Public Education & Religious Liberty n. Nyquist, supra, at 788 (“ ‘inhibi[t]’ ”); and since we have said that governmental “neutrality” toward religion is the preeminent goal of the First Amendment, see, e. g., Grand Rapids School District v. Ball, 473 U. S., at 382; Roemer v. Maryland Public Works Bd., 426 U. S. 736, 747 (1976) (plurality opinion); EDWARDS v. AGUILLARD 617 578 Scalia, J., dissenting Committee for Public Education & Religious Liberty v. Nyquist, supra, at 792-793; a State which discovers that its employees are inhibiting religion must take steps to prevent them from doing so, even though its purpose would clearly be to advance religion. Cf. Walz v. Tax Comm’n of New York City, supra, at 673. Thus, if the Louisiana Legislature sincerely believed that the State’s science teachers were being hostile to religion, our cases indicate that it could act to eliminate that hostility without running afoul of Lemon's purpose test. Second, we have held that intentional governmental advancement of religion is sometimes required by the Free Exercise Clause. For example, in Hobbie n. Unemployment Appeals Comm’n of Fla., 480 U. S. 136 (1987); Thomas n. Review Bd., Indiana Employment Security Div., 450 U. S. 707 (1981); Wisconsin v. Yoder, 406 U. S. 205 (1972); and Sherbert v. Verner, 374 U. S. 398 (1963), we held that in some circumstances States must accommodate the beliefs of religious citizens by exempting them from generally applicable regulations. We have not yet come close to reconciling Lemon and our Free Exercise cases, and typically we do not really try. See, e. g., Hobbie v. Unemployment Appeals Comm’n of Fla., supra, at 144-145; Thomas n. Review Bd., Indiana Employment Security Div., supra, at 719-720. It is clear, however, that members of the Louisiana Legislature were not impermissibly motivated for purposes of the Lemon test if they believed that approval of the Balanced Treatment Act was required by the Free Exercise Clause. We have also held that in some circumstances government may act to accommodate religion, even if that action is not required by the First Amendment. See Hobbie v. Unemployment Appeals Comm’n of Fla., supra, at 144-145. It is well established that “[t]he limits of permissible state accommodation to religion are by no means co-extensive with the noninterference mandated by the Free Exercise Clause.” Walz v. Tax Comm’n of New York City, supra, at 673; 618 OCTOBER TERM, 1986 Scalia, J., dissenting 482 U. S. see also Gillette v. United States, 401 U. S. 437, 453 (1971). We have implied that voluntary governmental accommodation of religion is not only permissible, but desirable. See, e. g., ibid. Thus, few would contend that Title VII of the Civil Rights Act of 1964, which both forbids religious discrimination by private-sector employers, 78 Stat. 255, 42 U. S. C. §2000e-2(a)(l), and requires them reasonably to accommodate the religious practices of their employees, §2000e(j), violates the Establishment Clause, even though its “purpose” is, of course, to advance religion, and even though it is almost certainly not required by the Free Exercise Clause. While we have warned that at some point, accommodation may devolve into “an unlawful fostering of religion,” Hobbie n. Unemployment Appeals Comm’n of Fla., supra, at 145, we have not suggested precisely (or even roughly) where that point might be. It is possible, then, that even if the sole motive of those voting for the Balanced Treatment Act was to advance religion, and its passage was not actually required, or even believed to be required, by either the Free Exercise or Establishment Clauses, the Act would nonetheless survive scrutiny under Lemon’s purpose test. One final observation about the application of that test: Although the Court’s opinion gives no hint of it, in the past we have repeatedly affirmed “our reluctance to attribute unconstitutional motives to the States.” Mueller v. Allen, supra, at 394; see also Lynch v. Donnelly, 465 U. S., at 699 (Brennan, J., dissenting). We “presume that legislatures act in a constitutional manner.” Illinois n. Krull, 480 U. S. 340, 351 (1987); see also Clements n. Fashing, 457 U. S. 957, 963 (1982) (plurality opinion); Rostker n. Goldberg, 453 U. S. 57, 64 (1981); McDonald v. Board of Election Comm’rs of Chicago, 394 U. S. 802, 809 (1969). Whenever we are called upon to judge the constitutionality of an act of a state legislature, “we must have ‘due regard to the fact that this Court is not exercising a primary judgment but is sitting in judgment EDWARDS v. AGUILLARD 619 578 Scalia, J., dissenting upon those who also have taken the oath to observe the Constitution and who have the responsibility for carrying on government.’” Rostker n. Goldberg, supra, at 64 (quoting Joint Anti-Fascist Refugee Committee n. McGrath, 341 U. S. 123, 164 (1951) (Frankfurter, J., concurring)). This is particularly true, we have said, where the legislature has specifically considered the question of a law’s constitutionality. Ibid. With the foregoing in mind, I now turn to the purposes underlying adoption of the Balanced Treatment Act. II A We have relatively little information upon which to judge the motives of those who supported the Act. About the only direct evidence is the statute itself and transcripts of the seven committee hearings at which it was considered. Unfortunately, several of those hearings were sparsely attended, and the legislators who were present revealed little about their motives. We have no committee reports, no floor debates, no remarks inserted into the legislative history, no statement from the Governor, and no postenactment statements or testimony from the bill’s sponsor or any other legislators. Cf. Wallace v. Jaffree, 472 U. S., at 43, 56-57. Nevertheless, there is ample evidence that the majority is wrong in holding that the Balanced Treatment Act is without secular purpose. At the outset, it is important to note that the Balanced Treatment Act did not fly through the Louisiana Legislature on wings of fundamentalist religious fervor—which would be unlikely, in any event, since only a small minority of the State’s citizens belong to fundamentalist religious denominations. See B. Quinn, H. Anderson, M. Bradley, P. Goetting, & P. Shriver, Churches and Church Membership in the United States 16 (1982). The Act had its genesis (so to speak) in legislation introduced by Senator Bill Keith in June 620 OCTOBER TERM, 1986 Scalia, J., dissenting 482 U. S. 1980. After two hearings before the Senate Committee on Education, Senator Keith asked that his bill be referred to a study commission composed of members of both Houses of the Louisiana Legislature. He expressed hope that the joint committee would give the bill careful consideration and determine whether his arguments were “legitimate.” 1 App. E-29—E-30. The committee met twice during the interim, heard testimony (both for and against the bill) from several witnesses, and received staff reports. Senator Keith introduced his bill again when the legislature reconvened. The Senate Committee on Education held two more hearings and approved the bill after substantially amending it (in part over Senator Keith’s objection). After approval by the full Senate, the bill was referred to the House Committee on Education. That committee conducted a lengthy hearing, adopted further amendments, and sent the bill on to the full House, where it received favorable consideration. The Senate concurred in the House amendments and on July 20, 1981, the Governor signed the bill into law. Senator Keith’s statements before the various committees that considered the bill hardly reflect the confidence of a man preaching to the converted. He asked his colleagues to “keep an open mind” and not to be “biased” by misleading characterizations of creation science. Id., at E-33. He also urged them to “look at this subject on its merits and not on some preconceived idea.” Id., at E-34; see also 2 id., at E-491. Senator Keith’s reception was not especially warm. Over his strenuous objection, the Senate Committee on Education voted 5-1 to amend his bill to deprive it of any force; as amended, the bill merely gave teachers permission to balance the teaching of creation science or evolution with the other. 1 id., at E-442—E-461. The House Committee restored the “mandatory” language to the bill by a vote of only 6-5, 2 id., at E-626—E-627, and both the full House (by vote of 52-35), id., at E-700-E-706, and full Senate (23-15), id., at E-735—E-738, had to repel further efforts to gut the bill. EDWARDS v. AGUILLARD 621 578 Scalia, J., dissenting The legislators understood that Senator Keith’s bill involved a “unique” subject, 1 id., at E-106 (Rep. M. Thompson), and they were repeatedly made aware of its potential constitutional problems, see, e. g., id., at E-26—E-28 (McGehee); id., at E-38—E-39 (Sen. Keith); id., at E-241— E-242 (Rossman); id., at E-257 (Probst); id., at E-261 (Beck); id., at E-282 (Sen. Keith). Although the Establishment Clause, including its secular purpose requirement, was of substantial concern to the legislators, they eventually voted overwhelmingly in favor of the Balanced Treatment Act: The House approved it 71-19 (with 15 members absent), 2 id., at E-716—E-722; the Senate 26-12 (with all members present), id., at E-741—E-744. The legislators specifically designated the protection of “academic freedom” as the purpose of the Act. La. Rev. Stat. Ann. § 17:286.2 (West 1982). We cannot accurately assess whether this purpose is a “sham,” ante, at 587, until we first examine the evidence presented to the legislature far more carefully than the Court has done. Before summarizing the testimony of Senator Keith and his supporters, I wish to make clear that I by no means intend to endorse its accuracy. But my views (and the views of this Court) about creation science and evolution are (or should be) beside the point. Our task is not to judge the debate about teaching the origins of life, but to ascertain what the members of the Louisiana Legislature believed. The vast majority of them voted to approve a bill which explicitly stated a secular purpose; what is crucial is not their wisdom in believing that purpose would be achieved by the bill, but their sincerity in believing it would be. Most of the testimony in support of Senator Keith’s bill came from the Senator himself and from scientists and educators he presented, many of whom enjoyed academic credentials that may have been regarded as quite impressive by members of the Louisiana Legislature. To a substantial extent, their testimony was devoted to lengthy, and, to the layman, seemingly expert scientific expositions on the origin 622 OCTOBER TERM, 1986 Scalia, J., dissenting 482 U. S. of life. See, e. g., 1 App. E-ll—E-18 (Sunderland); id., at E-50—E-60 (Boudreaux); id., at E-86—E-89 (Ward); id., at E-130—E-153 (Boudreaux paper); id., at E-321—E-326 (Boudreaux); id., at E-423—E-428 (Sen. Keith). These scientific lectures touched upon, inter alia, biology, paleontology, genetics, astronomy, astrophysics, probability analysis, and biochemistry. The witnesses repeatedly assured committee members that “hundreds and hundreds” of highly respected, internationally renowned scientists believed in creation science and would support their testimony. See, e. g., id., at E-5 (Sunderland); id., at E-76 (Sen. Keith); id., at E-100—E-101 (Reiboldt); id., at E-327—E-328 (Boudreaux); 2 id., at E-503—E-504 (Boudreaux). Senator Keith and his witnesses testified essentially as set forth in the following numbered paragraphs: (1) There are two and only two scientific explanations for the beginning of life3—evolution and creation science. 1 id., at E-6 (Sunderland); id., at E-34 (Sen. Keith); id., at E-280 (Sen. Keith); id., at E-417—E-418 (Sen. Keith). Both are bona fide “sciences.” Id., at E-6—E-7 (Sunderland); id., at E-12 (Sunderland); id., at E-416 (Sen. Keith); id., at E-427 (Sen. Keith); 2 id., at E-491—E-492 (Sen. Keith); id., at E-497—E-498 (Sen. Keith). Both posit a theory of the origin of life and subject that theory to empirical testing. Evolution posits that life arose out of inanimate chemical compounds and has gradually evolved over millions of years. Creation science posits that all life forms now on earth appeared suddenly and relatively recently and have changed little. Since there are only two possible explanations of the origin of life, any evidence that tends to disprove the theory of evolution necessarily tends to prove the theory of creation science, and vice versa. For example, the abrupt appearance in the fossil record of complex life, and the extreme rar 3 Although creation scientists and evolutionists also disagree about the origin of the physical universe, both proponents and opponents of Senator Keith’s bill focused on the question of the beginning of life. EDWARDS v. AGUILLARD 623 578 Scalia, J., dissenting ity of transitional life forms in that record, are evidence for creation science. 1 id., at E-7 (Sunderland); id., at E-12— E-18 (Sunderland); id., at E-45—E-60 (Boudreaux); id., at E-67 (Harlow); id., at E-130—E-153 (Boudreaux paper); id., at E-423-E-428 (Sen. Keith). (2) The body of scientific evidence supporting creation science is as strong as that supporting evolution. In fact, it may be stronger. Id., at E-214 (Young statement); id., at E-310 (Sen. Keith); id., at E-416 (Sen. Keith); 2 id., at E-492 (Sen. Keith). The evidence for evolution is far less compelling than we have been led to believe. Evolution is not a scientific “fact,” since it cannot actually be observed in a laboratory. Rather, evolution is merely a scientific theory or “guess.” 1 id., at E-20—E-21 (Morris); id., at E-85 (Ward); id., at E-100 (Reiboldt); id., at E-328—E-329 (Boudreaux); 2 id., at E-506 (Boudreaux). It is a very bad guess at that. The scientific problems with evolution are so serious that it could accurately be termed a “myth.” 1 id., at E-85 (Ward); id., at E-92—E-93 (Kalivoda); id., at E-95—E-97 (Sen. Keith); id., at E-154 (Boudreaux paper); id., at E-329 (Boudreaux); id., at E-453 (Sen. Keith); 2 id., at E-505— E-506 (Boudreaux); id., at E-516 (Young). (3) Creation science is educationally valuable. Students exposed to it better understand the current state of scientific evidence about the origin of life. 1 id., at E-19 (Sunderland); id., at E-39 (Sen. Keith); id., at E-79 (Kalivoda); id., at E-308 (Sen. Keith); 2 id., at E-513—E-514 (Morris). Those students even have a better understanding of evolution. 1 id., at E-19 (Sunderland). Creation science can and should be presented to children without any religious content. Id., at E-12 (Sunderland); id., at E-22 (Sanderford); id., at E-35—E-36 (Sen. Keith); id., at E-101 (Reiboldt); id., at E-279-E-280 (Sen. Keith); id., at E-282 (Sen. Keith). (4) Although creation science is educationally valuable and strictly scientific, it is now being censored from or misrepresented in the public schools. Id., at E-19 (Sunderland); id., 624 OCTOBER TERM, 1986 Scalia, J., dissenting 482 U. S. at E-21 (Morris); id., at E-34 (Sen. Keith); id., at E-37 (Sen. Keith); id., at E-42 (Sen. Keith); id., at E-92 (Kafi-voda); id., at E-97—E-98 (Reiboldt); id., at E-214 (Young statement); id., at E-218 (Young statement); id., at E-280 (Sen. Keith); id., at E-309 (Sen. Keith); 2 id., at E-513 (Morris). Evolution, in turn, is misrepresented as an absolute truth. 1 id., at E-63 (Harlow); id., at E-74 (Sen. Keith); id., at E-81 (Kalivoda); id., at E-214 (Young statement); 2 id., at E-507 (Harlow); id., at E-513 (Morris); id., at E-516 (Young). Teachers have been brainwashed by an entrenched scientific establishment composed almost exclusively of scientists to whom evolution is like a “religion.” These scientists discriminate against creation scientists so as to prevent evolution’s weaknesses from being exposed. 1 id., at E-61 (Boudreaux); id., at E-63—E-64 (Harlow); id., at E-78—E-79 (Kalivoda); id., at E-80 (Kalivoda); id., at E-95—E-97 (Sen. Keith); id., at E-129 (Boudreaux paper); id., at E-218 (Young statement); id., at E-357 (Sen. Keith); id., at E-430 (Boudreaux). (5) The censorship of creation science has at least two harmful effects. First, it deprives students of knowledge of one of the two scientific explanations for the origin of life and leads them to believe that evolution is proven fact; thus, their education suffers and they are wrongly taught that science has proved their religious beliefs false. Second, it violates the Establishment Clause. The United States Supreme Court has held that secular humanism is a religion. Id., at E-36 (Sen. Keith) (referring to Torcaso v. Watkins, 367 U. S. 488, 495, n. 11 (1961)); 1 App. E-418 (Sen. Keith); 2 id., at E-499 (Sen. Keith). Belief in evolution is a central tenet of that religion. 1 id., at E-282 (Sen. Keith); id., at E-312— E-313 (Sen. Keith); id., at E-317 (Sen. Keith); id., at E-418 (Sen. Keith); 2 id., at E-499 (Sen. Keith). Thus, by censoring creation science and instructing students that evolution is fact, public school teachers are now advancing religion in violation of the Establishment Clause. 1 id., at E-2—E-4 EDWARDS v. AGUILLARD 625 578 Scalia, J., dissenting (Sen. Keith); id., at E-36—E-37, E-39 (Sen. Keith); id., at E-154—E-155 (Boudreaux paper); id., at E-281—E-282 (Sen. Keith); id., at E-313 (Sen. Keith); id., at E-315— E-316 (Sen. Keith); id., at E-317 (Sen. Keith); 2 id., at E-499—E-500 (Sen. Keith). Senator Keith repeatedly and vehemently denied that his purpose was to advance a particular religious doctrine. At the outset of the first hearing on the legislation, he testified: “We are not going to say today that you should have some kind of religious instructions in our schools. . . . We are not talking about religion today. ... I am not proposing that we take the Bible in each science class and read the first chapter of Genesis.” 1 id., at E-35. At a later hearing, Senator Keith stressed: “[T]o . . . teach religion and disguise it as creationism ... is not my intent. My intent is to see to it that our textbooks are not censored.” Id., at E-280. He made many similar statements throughout the hearings. See, e. g., id., at E-41; id., at E-282; id., at E-310; id., at E-417; see also id., at E-44 (Boudreaux); id., at E-80 (Kalivoda). We have no way of knowing, of course, how many legislators believed the testimony of Senator Keith and his witnesses. But in the absence of evidence to the contrary,4 we 4 Although appellees and amici dismiss the testimony of Senator Keith and his witnesses as pure fantasy, they did not bother to submit evidence of that to the District Court, making it difficult for us to agree with them. The State, by contrast, submitted the affidavits of two scientists, a philosopher, a theologian, and an educator, whose academic credentials are rather impressive. See App. to Juris. Statement A-17—A-18 (Kenyon); id., at A-36 (Morrow); id., at A-39—A-40 (Miethe); id., at A-46—A-47 (Most); id., at A-49 (Clinkert). Like Senator Keith and his witnesses, the affiants swear that evolution and creation science are the only two scientific explanations for the origin of life, see id., at A-19—A-20 (Kenyon); id., at A-38 (Morrow); id., at A-41 (Miethe); that creation science is strictly scientific, see id., at A-18 (Kenyon); id., at A-36 (Morrow); id., at A-40—A-41 (Miethe); id., at A-49 (Clinkert); that creation science is simply a collection of scientific data that supports the hypothesis that life appeared on earth suddenly and has changed little, see id., at A-19 (Kenyon); id., at A-36 626 OCTOBER TERM, 1986 Scalia, J., dissenting 482 U. S. have to assume that many of them did. Given that assumption, the Court today plainly errs in holding that the Louisiana Legislature passed the Balanced Treatment Act for exclusively religious purposes. B Even with nothing more than this legislative history to go on, I think it would be extraordinary to invalidate the Balanced Treatment Act for lack of a valid secular purpose. Striking down a law approved by the democratically elected representatives of the people is no minor matter. “The cardinal principle of statutory construction is to save and not to destroy. We have repeatedly held that as between two possible interpretations of a statute, by one of which it would be unconstitutional and by the other valid, our plain duty is to adopt that which will save the act.” NLRB v. Jones & Laughlin Steel Corp., 301 U. S. 1, 30 (1937). So, too, it seems to me, with discerning statutory purpose. Even if the legislative history were silent or ambiguous about the existence of a secular purpose—and here it is not—the statute should survive Lemon’s purpose test. But even more validation than mere legislative history is present here. The Louisiana Legislature explicitly set forth its secular purpose (Morrow); id., at A-41 (Miethe); that hundreds of respected scientists believe in creation science, see id., at A-20 (Kenyon); that evidence for creation science is as strong as evidence for evolution, see id., at A-21 (Kenyon); id., at A-34—A-35 (Kenyon); id., at A-37—A-38 (Morrow); that creation science is educationally valuable, see id., at A-19 (Kenyon); id., at A-36 (Morrow); id., at A-38—A-39 (Morrow); id., at A-49 (Clinkert); that creation science can be presented without religious content, see id., at A-19 (Kenyon); id., at A-35 (Kenyon); id., at A-36 (Morrow); id., at A-40 (Miethe); id., at A-43—A-44 (Miethe); id., at A-47 (Most); id., at A-49 (Clinkert); and that creation science is now censored from classrooms while evolution is misrepresented as proven fact, see id., at A-20 (Kenyon); id., at A-35 (Kenyon); id., at A-39 (Morrow); id., at A-50 (Clinkert). It is difficult to conclude on the basis of these affidavits—the only substantive evidence in the record—that the laymen serving in the Louisiana Legislature must have disbelieved Senator Keith or his witnesses. EDWARDS v. AGUILLARD 627 578 Scalia, J., dissenting (“protecting academic freedom”) in the very text of the Act. La. Rev. Stat. §17:286.2 (West 1982). We have in the past repeatedly relied upon or deferred to such expressions, see, e. g., Committee for Public Education & Religious Liberty v. Regan, 444 U. S., at 654; Meek v. Pittenger, 421 U. S., at 363, 367-368; Committee for Public Education & Religious Liberty v. Nyquist, 413 U. S., at 773; Levitt n. Committee for Public Education & Religious Liberty, 413 U. S., at 479-480, n. 7; Tilton v. Richardson, 403 U. S., at 678-679 (plurality opinion); Lemon v. Kurtzman, 403 U. S., at 613; Board of Education n. Allen, 392 U. S., at 243. The Court seeks to evade the force of this expression of purpose by stubbornly misinterpreting it, and then finding that the provisions of the Act do not advance that misinterpreted purpose, thereby showing it to be a sham. The Court first surmises that “academic freedom” means “enhancing the freedom of teachers to teach what they will,” ante, at 586— even though “academic freedom” in that sense has little scope in the structured elementary and secondary curriculums with which the Act is concerned. Alternatively, the Court suggests that it might mean “maximiz[ing] the comprehensiveness and effectiveness of science instruction,” ante, at 588— though that is an exceedingly strange interpretation of the words, and one that is refuted on the very face of the statute. See § 17:286.5. Had the Court devoted to this central question of the meaning of the legislatively expressed purpose a small fraction of the research into legislative history that produced its quotations of religiously motivated statements by individual legislators, it would have discerned quite readily what “academic freedom” meant: students9 freedom from indoctrination. The legislature wanted to ensure that students would be free to decide for themselves how life began, based upon a fair and balanced presentation of the scientific evidence—that is, to protect “the right of each [student] voluntarily to determine what to believe (and what not to believe) free of any coercive pressures from the State.” Grand 628 OCTOBER TERM, 1986 Scalia, J., dissenting 482 U. S. Rapids School District v. Ball, 473 U. S., at 385. The legislature did not care whether the topic of origins was taught; it simply wished to ensure that when the topic was taught, students would receive “‘all of the evidence.’” Ante, at 586 (quoting Tr. of Oral Arg. 60). As originally introduced, the “purpose” section of the Balanced Treatment Act read: “This Chapter is enacted for the purposes of protecting academic freedom . . . of students . . . and assisting students in their search for truth.” 1 App. E-292 (emphasis added). Among the proposed findings of fact contained in the original version of the bill was the following: “Public school instruction in only evolution-science . . . violates the principle of academic freedom because it denies students a choice between scientific models and instead indoctrinates them in evolution science alone.” Id., at E-295 (emphasis added).5 Senator Keith unquestionably understood “academic freedom” to mean “freedom from indoctrination.” See id., at E-36 (purpose of bill is “to protect academic freedom by providing student choice”); id., at E-283 (purpose of bill is to protect “academic freedom” by giving students a “choice” rather than subjecting them to “indoctrination on origins”). If one adopts the obviously intended meaning of the statutory term “academic freedom,” there is no basis whatever for concluding that the purpose they express is a “sham.” Ante, 5 The majority finds it “astonishing” that I would cite a portion of Senator Keith’s original bill that was later deleted as evidence of the legislature’s understanding of the phrase “academic freedom.” Ante, at 589, n. 8. What is astonishing is the majority’s implication that the deletion of that section deprives it of value as a clear indication of what the phrase meant—there and in the other, retained, sections of the bill. The Senate Committee on Education deleted most of the lengthy “purpose” section of the bill (with Senator Keith’s consent) because it resembled legislative “findings of fact,” which, committee members felt, should generally not be incorporated in legislation. The deletion had absolutely nothing to do with the manner in which the section described “academic freedom.” See 1 App. E-314-E-320; id., at E-440-E-442. EDWARDS v. AGUILLARD 629 578 Scalia, J., dissenting at 587. To the contrary, the Act pursues that purpose plainly and consistently. It requires that, whenever the subject of origins is covered, evolution be “taught as a theory, rather than as proven scientific fact” and that scientific evidence inconsistent with the theory of evolution (viz., “creation science”) be taught as well. La. Rev. Stat. Ann. § 17:286.4A (West 1982). Living up to its title of “Balanced Treatment for Creation-Science and Evolution-Science Act,” §17.286.1, it treats the teaching of creation the same way. It does not mandate instruction in creation science, § 17:286.5; forbids teachers to present creation science “as proven scientific fact,” § 17:286.4A; and bans the teaching of creation science unless the theory is (to use the Court’s terminology) “discred-it[ed] ‘. . . at every turn’” with the teaching of evolution. Ante, at 589 (quoting 765 F. 2d, at 1257). It surpasses understanding how the Court can see in this a purpose “to restructure the science curriculum to conform with a particular religious viewpoint,” ante, at 593, “to provide a persuasive advantage to a particular religious doctrine,” ante, at 592, “to promote the theory of creation science which embodies a particular religious tenet,” ante, at 593, and “to endorse a particular religious doctrine,” ante, at 594. The Act’s reference to “creation” is not convincing evidence of religious purpose. The Act defines creation science as “scientific evidenc[e],” § 17:286.3(2) (emphasis added), and Senator Keith and his witnesses repeatedly stressed that the subject can and should be presented without religious content. See supra, at 623. We have no basis on the record to conclude that creation science need be anything other than a collection of scientific data supporting the theory that life abruptly appeared on earth. See n. 4, supra. Creation science, its proponents insist, no more must explain whence life came than evolution must explain whence came the inanimate materials from which it says life evolved. But even if that were not so, to posit a past creator is not to posit the eternal and personal God who is the object of religious veneration. 630 OCTOBER TERM, 1986 Scalia, J., dissenting 482 U. S. Indeed, it is not even to posit the “unmoved mover” hypothesized by Aristotle and other notably nonfundamentalist philosophers. Senator Keith suggested this when he referred to “a creator however you define a creator.” 1 App. E-280 (emphasis added). The Court cites three provisions of the Act which, it argues, demonstrate a “discriminatory preference for the teaching of creation science” and no interest in “academic freedom.” Ante, at 588. First, the Act prohibits discrimination only against creation scientists and those who teach creation science. §17:286.4C. Second, the Act requires local school boards to develop and provide to science teachers “a curriculum guide on presentation of creation-science.” § 17:286.7A. Finally, the Act requires the Governor to designate seven creation scientists who shall, upon request, assist local school boards in developing the curriculum guides. §17:286.7B. But none of these provisions casts doubt upon the sincerity of the legislators’ articulated purpose of “academic freedom”—unless, of course, one gives that term the obviously erroneous meanings preferred by the Court. The Louisiana legislators had been told repeatedly that creation scientists were scorned by most educators and scientists, who themselves had an almost religious faith in evolution. It is hardly surprising, then, that in seeking to achieve a balanced, “nonindoctrinating” curriculum, the legislators protected from discrimination only those teachers whom they thought were suffering from discrimination. (Also, the legislators were undoubtedly aware of Epperson v. Arkansas, 393 U. S. 97 (1968), and thus could quite reasonably have concluded that discrimination against evolutionists was already prohibited.) The two provisions respecting the development of curriculum guides are also consistent with “academic freedom” as the Louisiana Legislature understood the term. Witnesses had informed the legislators that, because of the hostility of most scientists and educators to creation science, the topic had been censored from or badly misrepresented in elementary EDWARDS v. AGUILLARD 631 578 Scalia, J., dissenting and secondary school texts. In light of the unavailability of works on creation science suitable for classroom use (a fact appellees concede, see Brief for Appellees 27, 40) and the existence of ample materials on evolution, it was entirely reasonable for the legislature to conclude that science teachers attempting to implement the Act would need a curriculum guide on creation science, but not on evolution, and that those charged with developing the guide would need an easily accessible group of creation scientists. Thus, the provisions of the Act of so much concern to the Court support the conclusion that the legislature acted to advance “academic freedom.” The legislative history gives ample evidence of the sincerity of the Balanced Treatment Act’s articulated purpose. Witness after witness urged the legislators to support the Act so that students would not be “indoctrinated” but would instead be free to decide for themselves, based upon a fair presentation of the scientific evidence, about the origin of life. See, e. g., 1 App. E-18 (Sunderland) (“all that we are advocating” is presenting “scientific data” to students and “letting [them] make up their own mind[s]”); id., at E-19— E-20 (Sunderland) (Students are now being “indoctrinated” in evolution through the use of “censored school books. . . . All that we are asking for is [the] open unbiased education in the classroom . . . your students deserve”); id., at E-21 (Morris) (“A student cannot [make an intelligent decision about the origin of life] unless he is well informed about both [evolution and creation science]”); id., at E-22 (Sanderford) (“We are asking very simply [that] . . . creationism [be presented] alongside . . . evolution and let people make their own mind[s] up”); id., at E-23 (Young) (the bill would require teachers to live up to their “obligation to present all theories” and thereby enable “students to make judgments themselves”); id., at E-44 (Boudreaux) (“Our intention is truth and as a scientist, I am interested in truth”); id., at E-60— E-61 (Boudreaux) (“[W]e [teachers] are guilty of a lot of 632 OCTOBER TERM, 1986 Scalia, J., dissenting 482 U. S. brainwashing. . . . We have a duty to . . . [present the] truth” to students “at all levels from gradeschool on through the college level”); id., at E-79 (Kalivoda) (“This [hearing] is being held I think to determine whether children will benefit from freedom of information or if they will be handicapped educationally by having little or no information about creation”); id., at E-80 (Kalivoda) (“I am not interested in teaching religion in schools. ... I am interested in the truth and [students] having the opportunity to hear more than one side”); id., at E-98 (Reiboldt) (“The students have a right to know there is an alternate creationist point of view. They have a right to know the scientific evidences which suppor[t] that alternative”); id., at E-218 (Young statement) (passage of the bill will ensure that “communication of scientific ideas and discoveries may be unhindered”); 2 id., at E-514 (Morris) (“[A]re we going to allow [students] to look at evolution, to look at creationism, and to let one or the other stand or fall on its own merits, or will we by failing to pass this bill. . . deny students an opportunity to hear another viewpoint?”); id., at E-516—E-517 (Young) (“We want to give the children here in this state an equal opportunity to see both sides of the theories”). Senator Keith expressed similar views. See, e. g., 1 id., at E-36; id., at E-41; id., at E-280; id., at E-283. Legislators other than Senator Keith made only a few statements providing insight into their motives, but those statements cast no doubt upon the sincerity of the Act’s articulated purpose. The legislators were concerned primarily about the manner in which the subject of origins was presented in Louisiana schools—specifically, about whether scientifically valuable information was being censored and students misled about evolution. Representatives Cain, Jenkins, and F. Thompson seemed impressed by the scientific evidence presented in support of creation science. See 2 id., at E-530 (Rep. F. Thompson); id., at E-533 (Rep. Cain); id., at E-613 (Rep. Jenkins). At the first study commission hearing, Senator Picard and Representative M. Thompson ques EDWARDS v. AGUILLARD 633 578 Scalia, J., dissenting tioned Senator Keith about Louisiana teachers’ treatment of evolution and creation science. See 1 id., at E-71—E-74. At the close of the hearing, Representative M. Thompson told the audience: “We as members of the committee will also receive from the staff information of what is currently being taught in the Louisiana public schools. We really want to see [it]. I. . . have no idea in what manner [biology] is presented and in what manner the creationist theories [are] excluded in the public school[s]. We want to look at what the status of the situation is.” Id., at E-104. Legislators made other comments suggesting a concern about censorship and misrepresentation of scientific information. See, e. g., id., at E-386 (Sen. McLeod); 2 id., at E-527 (Rep. Jenkins); id., at E-528 (Rep. M. Thompson); id., at E-534 (Rep. Fair). It is undoubtedly true that what prompted the legislature to direct its attention to the misrepresentation of evolution in the schools (rather than the inaccurate presentation of other topics) was its awareness of the tension between evolution and the religious beliefs of many children. But even appellees concede that a valid secular purpose is not rendered impermissible simply because its pursuit is prompted by concern for religious sensitivities. Tr. of Oral Arg. 43, 56. If a history teacher falsely told her students that the bones of Jesus Christ had been discovered, or a physics teacher that the Shroud of Turin had been conclusively established to be inexplicable on the basis of natural causes, I cannot believe (despite the majority’s implication to the contrary, see ante, at 592-593) that legislators or school board members would be constitutionally prohibited from taking corrective action, simply because that action was prompted by concern for the religious beliefs of the misinstructed students. In sum, even if one concedes, for the sake of argument, that a majority of the Louisiana Legislature voted for the Balanced Treatment Act partly in order to foster (rather 634 OCTOBER TERM, 1986 Scalia, J., dissenting 482 U. S. than merely eliminate discrimination against) Christian fundamentalist beliefs, our cases establish that that alone would not suffice to invalidate the Act, so long as there was a genuine secular purpose as well. We have, moreover, no adequate basis for disbelieving the secular purpose set forth in the Act itself, or for concluding that it is a sham enacted to conceal the legislators’ violation of their oaths of office. I am astonished by the Court’s unprecedented readiness to reach such a conclusion, which I can only attribute to an intellectual predisposition created by the facts and the legend of Scopes v. State, 154 Tenn. 105, 289 S. W. 363 (1927)—an instinctive reaction that any governmentally imposed requirements bearing upon the teaching of evolution must be a manifestation of Christian fundamentalist repression. In this case, however, it seems to me the Court’s position is the repressive one. The people of Louisiana, including those who are Christian fundamentalists, are quite entitled, as a secular matter, to have whatever scientific evidence there may be against evolution presented in their schools, just as Mr. Scopes was entitled to present whatever scientific evidence there was for it. Perhaps what the Louisiana Legislature has done is unconstitutional because there is no such evidence, and the scheme they have established will amount to no more than a presentation of the Book of Genesis. But we cannot say that on the evidence before us in this summary judgment context, which includes ample uncontradicted testimony that “creation science” is a body of scientific knowledge rather than revealed belief. Infinitely less can we say (or should we say) that the scientific evidence for evolution is so conclusive that no one could be gullible enough to believe that there is any real scientific evidence to the contrary, so that the legislation’s stated purpose must be a lie. Yet that illiberal judgment, that Scopes-in-reverse, is ultimately the basis on which the Court’s facile rejection of the Louisiana Legislature’s purpose must rest. EDWARDS v. AGUILLARD 635 578 Scalia, J., dissenting Since the existence of secular purpose is so entirely clear, and thus dispositive, I will not go on to discuss the fact that, even if the Louisiana Legislature’s purpose were exclusively to advance religion, some of the well-established exceptions to the impermissibility of that purpose might be applicable— the validating intent to eliminate a perceived discrimination against a particular religion, to facilitate its free exercise, or to accommodate it. See supra, at 617-618. I am not in any case enamored of those amorphous exceptions, since I think them no more than unpredictable correctives to what is (as the next Part of this opinion will discuss) a fundamentally unsound rule. It is surprising, however, that the Court does not address these exceptions, since the context of the legislature’s action gives some reason to believe they may be applicable.6 6 As the majority recognizes, ante, at 592, Senator Keith sincerely believed that “secular humanism is a bona fide religion,” 1 App. E-36; see also id., at E-418; 2 id., at E-499, and that “evolution is the cornerstone of that religion,” 1 id., at E-418; see also id., at E-282; id., at E-312—E-313; id., at E-317; 2 id., at E-499. The Senator even told his colleagues that this Court had “held” that secular humanism was a religion. See 1 id., at E-36, id., at E-418; 2 id., at E-499. (In Torcaso v. Watkins, 367 U. S. 488, 495, n. 11 (1961), we did indeed refer to “Secular Humanism” as a “religio[n].”) Senator Keith and his supporters raised the “religion” of secular humanism not, as the majority suggests, to explain the source of their “disdain for the theory of evolution,” ante, at 592, but to convince the legislature that the State of Louisiana was violating the Establishment Clause because its teachers were misrepresenting evolution as fact and depriving students of the information necessary to question that theory. 1 App. E-2-E-4 (Sen. Keith); id., at E-36-E-37, E-39 (Sen. Keith); id., at E-154—E-155 (Boudreaux paper); id., at E-281—E-282 (Sen. Keith); id., at E-317 (Sen. Keith); 2 id., at E-499—E-500 (Sen. Keith). The Senator repeatedly urged his colleagues to pass his bill to remedy this Establishment Clause violation by ensuring state neutrality in religious matters, see, e. g., 1 id., at E-36; id., at E-39; id., at E-313, surely a permissible purpose under Lemon. Senator Keith’s argument may be questionable, but nothing in the statute or its legislative history gives us reason to doubt his sincerity or that of his supporters. 636 OCTOBER TERM, 1986 Scalia, J., dissenting 482 U. S. Because I believe that the Balanced Treatment Act had a secular purpose, which is all the first component of the Lemon test requires, I would reverse the judgment of the Court of Appeals and remand for further consideration. Ill I have to this point assumed the validity of the Lemon “purpose” test. In fact, however, I think the pessimistic evaluation that The Chief Justice made of the totality of Lemon is particularly applicable to the “purpose” prong: it is “a constitutional theory [that] has no basis in the history of the amendment it seeks to interpret, is difficult to apply and yields unprincipled results . . . .” Wallace n. Jaffree, 472 U. S., at 112 (Rehnquist, J., dissenting). Our cases interpreting and applying the purpose test have made such a maze of the Establishment Clause that even the most conscientious governmental officials can only guess what motives will be held unconstitutional. We have said essentially the following: Government may not act with the purpose of advancing religion, except when forced to do so by the Free Exercise Clause (which is now and then); or when eliminating existing governmental hostility to religion (which exists sometimes); or even when merely accommodating governmentally uninhibited religious practices, except that at some point (it is unclear where) intentional accommodation results in the fostering of religion, which is of course unconstitutional. See supra, at 614-618. But the difficulty of knowing what vitiating purpose one is looking for is as nothing compared with the difficulty of knowing how or where to find it. For while it is possible to discern the objective “purpose” of a statute (i. e., the public good at which its provisions appear to be directed), or even the formal motivation for a statute where that is explicitly set forth (as it was, to no avail, here), discerning the subjective motivation of those enacting the statute is, to be honest, almost always an impossible task. The number of possible EDWARDS v. AGUILLARD 637 578 Scalia, J., dissenting motivations, to begin with, is not binary, or indeed even finite. In the present case, for example, a particular legislator need not have voted for the Act either because he wanted to foster religion or because he wanted to improve education. He may have thought the bill would provide jobs for his district, or may have wanted to make amends with a faction of his party he had alienated on another vote, or he may have been a close friend of the bill’s sponsor, or he may have been repaying a favor he owed the majority leader, or he may have hoped the Governor would appreciate his vote and make a fundraising appearance for him, or he may have been pressured to vote for a bill he disliked by a wealthy contributor or by a flood of constituent mail, or he may have been seeking favorable publicity, or he may have been reluctant to hurt the feelings of a loyal staff member who worked on the bill, or he may have been settling an old score with a legislator who opposed the bill, or he may have been mad at his wife who opposed the bill, or he may have been intoxicated and utterly unmotivated when the vote was called, or he may have accidentally voted “yes” instead of “no,” or, of course, he may have had (and very likely did have) a combination of some of the above and many other motivations. To look for the sole purpose of even a single legislator is probably to look for something that does not exist. Putting that problem aside, however, where ought we to look for the individual legislator’s purpose? We cannot of course assume that every member present (if, as is unlikely, we know who or even how many they were) agreed with the motivation expressed in a particular legislator’s preenactment floor or committee statement. Quite obviously, “[w]hat motivates one legislator to make a speech about a statute is not necessarily what motivates scores of others to enact it.” United States v. O’Brien, 391 U. S. 367, 384 (1968). Can we assume, then, that they all agree with the motivation expressed in the staff-prepared committee reports they might have read—even though we are unwilling to 638 OCTOBER TERM, 1986 Scalia, J., dissenting 482 U. S. assume that they agreed with the motivation expressed in the very statute that they voted for? Should we consider postenactment floor statements? Or postenactment testimony from legislators, obtained expressly for the lawsuit? Should we consider media reports on the realities of the legislative bargaining? All of these sources, of course, are eminently manipulable. Legislative histories can be contrived and sanitized, favorable media coverage orchestrated, and postenactment recollections conveniently distorted. Perhaps most valuable of all would be more objective indications—for example, evidence regarding the individual legislators’ religious affiliations. And if that, why not evidence regarding the fervor or tepidity of their beliefs? Having achieved, through these simple means, an assessment of what individual legislators intended, we must still confront the question (yet to be addressed in any of our cases) how many of them must have the invalidating intent. If a state senate approves a bill by vote of 26 to 25, and only one of the 26 intended solely to advance religion, is the law unconstitutional? What if 13 of the 26 had that intent? What if 3 of the 26 had the impermissible intent, but 3 of the 25 voting against the bill were motivated by religious hostility or were simply attempting to “balance” the votes of their impermissibly motivated colleagues? Or is it possible that the intent of the bill’s sponsor is alone enough to invalidate it—on a theory, perhaps, that even though everyone else’s intent was pure, what they produced was the fruit of a forbidden tree? Because there are no good answers to these questions, this Court has recognized from Chief Justice Marshall, see Fletcher v. Peck, 6 Cranch 87, 130 (1810), to Chief Justice Warren, United States v. O'Brien, supra, at 383-384, that determining the subjective intent of legislators is a perilous enterprise. See also Palmer n. Thompson, 403 U. S. 217, 224-225 (1971); Epperson n. Arkansas, 393 U. S., at 113 (Black, J., concurring). It is perilous, I might note, not just for the judges who will very likely reach the wrong result, EDWARDS v. AGUILLARD 639 578 Scalia, J., dissenting but also for the legislators who find that they must assess the validity of proposed legislation—and risk the condemnation of having voted for an unconstitutional measure—not on the basis of what the legislation contains, nor even on the basis of what they themselves intend, but on the basis of what others have in mind. Given the many hazards involved in assessing the subjective intent of governmental decisionmakers, the first prong of Lemon is defensible, I think, only if the text of the Establishment Clause demands it. That is surely not the case. The Clause states that “Congress shall make no law respecting an establishment of religion.” One could argue, I suppose, that any time Congress acts with the intent of advancing religion, it has enacted a “law respecting an establishment of religion”; but far from being an unavoidable reading, it is quite an unnatural one. I doubt, for example, that the Clayton Act, 38 Stat. 730, as amended, 15 U. S. C. §12 et seq., could reasonably be described as a “law respecting an establishment of religion” if bizarre new historical evidence revealed that it lacked a secular purpose, even though it has no discernible nonsecular effect. It is, in short, far from an inevitable reading of the Establishment Clause that it forbids all governmental action intended to advance religion; and if not inevitable, any reading with such untoward consequences must be wrong. In the past we have attempted to justify our embarrassing Establishment Clause jurisprudence7 on the ground that it 7 Professor Choper summarized our school aid cases thusly: “[A] provision for therapeutic and diagnostic health services to parochial school pupils by public employees is invalid if provided in the parochial school, but not if offered at a neutral site, even if in a mobile unit adjacent to the parochial school. Reimbursement to parochial schools for the expense of administering teacher-prepared tests required by state law is invalid, but the state may reimburse parochial schools for the expense of administering state-prepared tests. The state may lend school textbooks to parochial school pupils because, the Court has explained, the books can be checked in advance for religious content and are ‘self-policing’; but the 640 OCTOBER TERM, 1986 Scalia, J., dissenting 482 U. S. “sacrifices clarity and predictability for flexibility.” Committee for Public Education & Religious Liberty v. Regan, 444 U. S., at 662. One commentator has aptly characterized this as “a euphemism . . . for . . . the absence of any principled rationale.” Choper, supra n. 7, at 681. I think it time that we sacrifice some “flexibility” for “clarity and predictability.” Abandoning Lemon’s purpose test—a test which exacerbates the tension between the Free Exercise and Establishment Clauses, has no basis in the language or history of the Amendment, and, as today’s decision shows, has wonderfully flexible consequences—would be a good place to start. state may not lend other seemingly self-policing instructional items such as tape recorders and maps. The state may pay the cost of bus transportation to parochial schools, which the Court has ruled are ‘permeated’ with religion; but the state is forbidden to pay for field trip transportation visits ‘to governmental, industrial, cultural, and scientific centers designed to enrich the secular studies of students.’ ” Choper, The Religion Clauses of the First Amendment: Reconciling the Conflict, 41 U. Pitt. L. Rev. 673, 680-681 (1980) (footnotes omitted). Since that was written, more decisions on the subject have been rendered, but they leave the theme of chaos securely unimpaired. See, e. g., Aguilar v. Felton, 473 U. S. 402 (1985); Grand Rapids School District v. Ball, 473 U. S. 373 (1985). FRAZIER v. HEEBE 641 Syllabus FRAZIER u HEEBE, CHIEF JUDGE, UNITED STATES DISTRICT COURT FOR THE EASTERN DISTRICT OF LOUISIANA, et al. CERTIORARI TO THE UNITED STATES COURT OF APPEALS FOR THE FIFTH CIRCUIT No. 86-475. Argued April 29, 1987—Decided June 19, 1987 Petitioner, an attorney who maintained both his residence and his law office in Mississippi and who was a member of the Mississippi and Louisiana State Bars, was denied admission to the Bar of the United States District Court for the Eastern District of Louisiana because he neither lived nor had an office in Louisiana, as required by the court’s local Rule 21.2. He was also ineligible under the court’s Rule 21.3.1, which requires continuous and uninterrupted Louisiana residence or maintenance of a Louisiana law office for continuing eligibility in the bar. He sought a writ of prohibition from the Court of Appeals, alleging that the restrictions in the Rules were unconstitutional on their face and as applied to him. The court remanded the case to the District Court for appropriate proceedings and entry of an appealable judgment. That court upheld Rule 21.2 as constitutional. The Court of Appeals affirmed. Held: The District Court was not empowered to adopt Rules requiring members of the Louisiana Bar who apply for admission to its bar to live, or maintain an office, in Louisiana. Pp. 645-651. (a) A district court has discretion to adopt local rules that are necessary to carry out its business, including rules governing admission to its bar. However, this Court may exercise its inherent supervisory power (as it does here) to ensure that local rules are consistent with principles of right and justice. Pp. 645-646. (b) Rule 21.2’s residence requirement is unnecessary and arbitrarily discriminates against out-of-state attorneys who are members of the Louisiana Bar and are willing to pay the necessary fees and dues in order to be admitted to the Eastern District Bar. There is no reason to believe that such attorneys are less competent than resident attorneys. Moreover, other more effective means of ensuring the competence of bar members are available to the district courts, including examination or seminar attendance requirements. Nor does an alleged need for immediate availability of attorneys require a blanket rule that denies all nonresident attorneys admission to a district court bar. As a practical matter, a high percentage of nonresident attorneys willing to take the state bar examination and pay the annual dues will reside in places reasonably 642 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. convenient to the district court. Moreover, modern communication systems make it possible to minimize the problem of unavailability, and district courts also have alternative means to ensure prompt attendance at important conferences. Pp. 646-649. (c) The in-state office requirement is similarly unnecessary and irrational. It is not imposed on a lawyer residing in Louisiana whose only office is out-of-state and who is equally as unavailable to the court as a nonresident lawyer with an out-of-state office. Nor does the mere fact that an attorney has an office in Louisiana warrant the assumption that he or she is more competent than an out-of-state member of the state bar. Moreover, any need the court may have to ensure the availability of attorneys does not justify the in-state office requirement. There is no link between residency within a State and proximity to a courthouse. P. 650. (d) The contention that nonresident lawyers are not totally foreclosed from Eastern District practice because they can appear pro hac vice is unpersuasive. Such alternative does not allow the nonresident attorney to practice on the same terms as a resident member of the bar. In order to appear pro hac vice under the District Court’s Rules, a lawyer must associate with a member of the court’s bar. Such association imposes a financial and administrative burden on nonresident counsel. Furthermore, “local” counsel may be located much farther from the courthouse than the out-of-state counsel. Pp. 650-651. 788 F. 2d 1049, reversed. Brennan, J., delivered the opinion of the Court, in which White, Marshall, Blackmun, Powell, and Stevens, JJ., joined. Rehnquist, C. J., filed a dissenting opinion, in which O’Connor and Scalia, JJ., joined, post, p. 651. Cornish F. Hitchcock argued the cause for petitioner. With him on the briefs were Alan B. Morrison and Gary L. Roberts. Curtis R. Boisfontaine argued the cause and filed a brief for respondents.* Justice Brennan delivered the opinion of the Court. The question for decision is whether a United States District Court may require that applicants for general admission * Lawrence A. Salibra II filed a brief for the American Corporate Counsel Association as amicus curiae urging reversal. FRAZIER v. HEEBE 643 641 Opinion of the Court to its bar either reside or maintain an office in the State where that court sits. I Petitioner David Frazier is an attorney having both his residence and his law office in Pascagoula, Mississippi. An experienced litigator, he is a member of the Mississippi and Louisiana State Bars, and also of the Bars of the United States Courts of Appeals for the Fifth and Eleventh Circuits and the United States District Court for the Southern District of Mississippi. In April 1982, Frazier applied for admission to the Bar of the United States District Court for the Eastern District of Louisiana. His application was denied because he neither lived nor had an office in Louisiana, as required by the court’s local Rule 21.2. In addition, Frazier was ineligible for admission under the court’s local Rule 21.3.1, which requires continuous and uninterrupted Louisiana residence or maintenance of a Louisiana law office for continuing eligibility in that bar. Frazier challenged these District Court Rules by petitioning for a writ of prohibition from the Court of Appeals for the Fifth Circuit. The petition alleged that the restrictions in Rules 21.2 and 21.3.1 were unconstitutional, on their face and as applied to him. The Court of Appeals did not rule on the petition, but remanded the case to the District Court for the Eastern District for appropriate proceedings and entry of an appealable judgment. All the judges of the Eastern District recused themselves. The matter was assigned to Judge Edwin Hunter, a Senior Judge of the Western District of Louisiana. The District Court held a 1-day bench trial in which two District Court Judges, two Magistrates, and the Clerk of the Eastern District testified in support of the challenged Rules. Frazier challenged the District Court Rules on several constitutional grounds, primarily under the equal protection requirement of the Due Process Clause of the Fifth Amend 644 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. ment.1 Applying the standard of intermediate scrutiny, the District Court upheld Rule 21.2 as constitutional.2 594 F. Supp. 1173, 1179 (1984). The District Court found that the Rule serves the important Government objective of the efficient administration of justice. Ibid. It relied on testimony by court officials that proximity to the New Orleans courthouse is important when emergencies arise during proceedings, and that participation by nonresident attorneys complicates the scheduling of routine court matters. Id., at 1183-1184. The court also found that the office requirement is not unduly restrictive and that it increases the availability of an attorney to the court. Finally, it stated the failure to require in-state attorneys to open a local office was reasonable, since such attorneys “must of necessity open an office,” and, even absent an office, an instate attorney is likely to be available. Ibid. Without further explanation, the court declared that the in-state attorney’s admission to the bar “does not raise the same concern for the efficient administration of justice that admission of nonresident attorneys does.” Ibid. After reviewing petitioner’s other claims, the District Court denied Frazier’s petition for extraordinary relief and dismissed his suit. The Court of Appeals affirmed over a dissent. 788 F. 2d 1049 (1986). The court found that the discrimination at issue did not warrant heightened scrutiny, and held that the 1 Petitioner also contended that the local Rules violated the Commerce Clause, the Full Faith and Credit Clause, the Privileges and Immunities Clause, and the First and Fourteenth Amendments of the Federal Constitution. 2 In determining the level of review appropriate for the federal equal protection challenge, the court determined that no fundamental constitutional right was implicated and that Frazier was not a member of a suspect class. The court therefore concluded that strict scrutiny was unnecessary. The court did not determine whether intermediate or deferential scrutiny was required for classifications based on state residency, because it concluded that, even under intermediate scrutiny, Rule 21.2 was constitutional. 594 F. Supp. 1173, 1180-1182 (1984). FRAZIER v. HEEBE 645 641 Opinion of the Court exclusion was rationally related to the District Court’s goal of promoting lawyer competence and availability for hearings. It characterized the testimony before the District Court as “of one voice: lawyers admitted pro hac vice, who neither reside nor maintain an office in Louisiana, fail to comply with the local rules and impede the efficient administration of justice more than members of the bar of the Eastern District.” Id., at 1054. The court also noted that out-of-state attorneys were not unduly disadvantaged by this restriction, since they could affiliate with Louisiana counsel and appear pro hac vice. Id., at 1054—1055. Finally, the court denied petitioner’s alternative request to invalidate these Rules through use of the Court of Appeals’ supervisory power over District Courts in that Circuit. The court expressed its reluctance to exercise its supervisory authority because the Fifth Circuit Judicial Council was at that time reviewing the local Rules of the District Courts in the Circuit. Id., at 1055. We granted certiorari, 479 U. S. 960 (1986), and now reverse. Pursuant to our supervisory authority, we hold that the District Court was not empowered to adopt its local Rules to require members of the Louisiana Bar who apply for admission to its bar to live in, or maintain an office in, Louisiana where that court sits. We therefore need not address the constitutional questions presented. II We begin our analysis by recognizing that a district court has discretion to adopt local rules that are necessary to carry out the conduct of its business. See 28 U. S. C. §§1654, 2071; Fed. Rule Civ. Proc. 83. This authority includes the regulation of admissions to its own bar. A district court’s discretion in promulgating local rules is not, however, without limits. This Court may exercise its inherent supervisory power to ensure that these local rules are consistent with “‘the principles of right and justice.’” In re Ruffalo, 390 U. S. 544, 554 (1968) (White, J., concurring) (citation omit 646 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. ted); see In re Snyder, 472 U. S. 634, 643 (1985); Theard n. United States, 354 U. S. 278, 282 (1957); Ex parte Burr, 9 Wheat. 529, 530 (1824).3 Section 2071 requires that local rules of a district court “shall be consistent with” the “rules of practice and procedure prescribed by the Supreme Court.”4 Today we invoke our supervisory authority to prohibit arbitrary discrimination against members of the Louisiana Bar, residing and having their office out-of-state, who are otherwise qualified to join the Bar of the Eastern District. In the present case, our attention is focused on the requirements imposed by Rule 21.2 of the Eastern District of Louisiana,5 namely that, to be admitted to the bar, an attorney must reside or maintain an office in Louisiana. Respondents assert that these requirements facilitate the efficient administration of justice, because nonresident attorneys allegedly are less competent and less available to the court than resident attorneys. We disagree. We find both requirements to be unnecessary and irrational. Rule 21.2’s requirement of residence in Louisiana arbitrarily discriminates against out-of-state attorneys who have passed the Louisiana bar examination and are willing to pay the necessary fees and dues in order to be admitted to the Eastern District Bar. No empirical evidence was introduced 3 See also Flanders, Local Rules in Federal District Courts: Usurpation, Legislation, or Information, 14 Loyola (LA) L. Rev. 213, 252-256 (1981); Martineau, The Supreme Court and State Regulation of the Legal Profession, 8 Hastings Const. L. Q. 199, 234-236 (1981); Note, The Supervisory Power of the Federal Courts, 76 Harv. L. Rev. 1656, 1656-1657 (1963). 4 Section 2072 confirms the supervisory authority that the Court has over lower federal courts: “The Supreme Court shall have the power to prescribe by general rules, the forms of process, writs, pleadings, and motions, and the practice and procedure of the district courts and courts of appeals of the United States in civil actions. ...” 28 U. S. C. §2072. The local rules must also be consistent with Acts of Congress. 28 U. S. C. §2071. Congress thus far has chosen to leave regulation of the federal bars to the courts. Petitioner does not challenge the requirement of Rule 21.2 that an attorney must be a member in good standing of the Louisiana Bar. FRAZIER v. HEEBE 647 641 Opinion of the Court at trial to demonstrate why this class of attorneys, although members of the Louisiana Bar, should be excluded from the Eastern District’s Bar.6 Instead, the evidence was limited almost exclusively to experiences with pro hac vice practitioners, who unlike petitioner, were not members of the Louisiana Bar. Tr. 153. Experience with this category of onetime or occasional practitioners does not provide a basis for predicting the behavior of attorneys, who are members of the Louisiana Bar and who seek to practice in the Eastern District on a regular basis. Indeed, there is no reason to believe that nonresident attorneys who have passed the Louisiana bar examination are less competent than resident attorneys. The competence of the former group in local and federal law has been tested and demonstrated to the same extent as that of Louisiana lawyers, and its members are equally qualified. We are unwilling to assume that “a nonresident lawyer—any more than a resident—would disserve his clients by failing to familiarize himself [or herself] with the [local] rules.” Supreme Court of New Hampshire n. Piper, 470 U. S. 274, 285 (1985).7 The 6 During the bench trial, there was only one occasion when a witness, testifying in favor of the local Rules, distinguished between nonresident members of the Louisiana Bar and pro hac vice practitioners. In that instance, the witness could offer anecdotal testimony about only two nonresident members of the Louisiana Bar. Tr. 214-215 (testimony of Magistrate Wynne). 7 In Supreme Court of New Hampshire v. Piper, 470 U. S. 274 (1985), the Court held that a Rule by a State Supreme Court that limited bar admission to state residents violated the Privileges and Immunities Clause of Art. IV, §2. In the context of that case, the Court considered several contentions quite similar to those presented here. The Court rejected the notion that nonresident attorneys should be presumed to be less competent or less available than resident attorneys. 470 U. S., at 285-286. We held that a State may discriminate against nonresident attorneys only where its reasons are substantial and the difference in treatment bears a close relationship to those reasons. Rules that discriminate against nonresident attorneys are even more difficult to justify in the context of federal-court practice than they are in the 648 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. Court has previously recognized that a nonresident lawyer is likely to have a substantial incentive, as a practical matter, to learn and keep abreast of local rules. Ibid. A lawyer’s application to a particular bar is likely to be based on the expectation of considerable local practice, since it requires the personal investment of taking the state bar examination and paying fees and annual dues. Moreover, other more effective means of ensuring the competence of bar members are available to the district courts, including examination or seminar attendance requirements. Complete exclusion is unnecessary. We also do not believe that an alleged need for immediate availability of attorneys in some proceedings requires a blanket rule that denies all nonresident attorneys admission to a district-court bar. If attorney availability is a significant problem, the Rules are poorly crafted to remedy it. For example, the Rules presume that a lawyer in Shreveport, Louisiana, which is located more than 300 miles from the New Orleans courthouse of the Eastern District, is more likely or able to attend a conference than a lawyer such as petitioner, who is only 110 miles away, but must cross a state boundary on his way to the court. As a practical matter, a high per-area of state-court practice, where laws and procedures may differ substantially from State to State. See Comisky & Patterson, The Case for a Federally Created National Bar by Rule or by Legislation, 55 Temp. L. Q. 945, 960-964 (1982). There is a growing body of specialized federal law and a more mobile federal bar, accompanied by an increased demand for specialized legal services regardless of state boundaries. See Simonelli, State Regulation of a Federal License to Practice Law, 56 N. Y. State Bar J. 15 (May 1984). The Court’s supervisory power over federal courts allows the Court to intervene to protect the integrity of the federal system, while its authority over state-court bars is limited to enforcing federal constitutional requirements. Because of these differences, the Court has repeatedly emphasized, for example, that disqualification from membership from a state bar does not necessarily lead to disqualification from a federal bar. See Theard v. United States, 354 U. S. 278, 282 (1957); Selling v. Radford, 243 U. S. 46, 49 (1917); cf. Sperry v. Florida ex rel. Florida Bar, 373 U. S. 379, 385-387 (1963). FRAZIER v. HEEBE 649 641 Opinion of the Court centage of nonresident attorneys willing to take the state bar examination and pay the annual dues will reside in places “reasonably convenient” to the District Court. Cf. 470 U. S., at 286-287. Moreover, modern communication systems, including conference telephone arrangements, make it possible to minimize the problem of unavailability. Finally, district courts have alternative means to ensure prompt attendance at important conferences. For instance, they may impose sanctions on lawyers who fail to appear on schedule. Indeed, the Eastern District has adopted Rule 21.8.1, which specifically requires that sanctions be imposed on lawyers who fail to appear at hearings.8 We therefore conclude that the residency requirement imposed by the Eastern District is unnecessary and arbitrarily discriminates against out-of-state attorneys. Similarly, we find the in-state office requirement unnecessary and irrational. First, the requirement is not imposed on in-state attorneys. A resident lawyer is allowed to maintain his or her only office outside of Louisiana. A resident lawyer with an out-of-state office is equally as unavailable to the court as a nonresident lawyer with an out-of-state office. In addition, the mere fact that an attorney has an office in Louisiana surely does not warrant the assumption that he or she is more competent than an out-of-state member of the state bar. Requiring petitioner to have a Louisiana address and telephone number, and an in-state answering service will not elevate his or her understanding of the local Rules. As the failure to require in-state attorneys to have an in-state office reveals, the location of a lawyer’s office simply has nothing to do with his or her intellectual ability or experience in litigating cases in Federal District Court. 8 Furthermore, the Court noted in Piper that “[t]he trial court, by rule or as an exercise of discretion, may require any lawyer who resides at a great distance to retain a local attorney who will be available for unscheduled meetings and hearings.” 470 U. S., at 287. 650 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. We further conclude that any need the court may have to ensure the availability of attorneys does not justify the instate office requirement. As observed with regard to state residency requirements, there is no link between residency within a State and proximity to a courthouse. The office requirement does not specify that counsel be in the Eastern District, but only that the attorney have an office somewhere in the State, regardless of how far that office is from the courthouse.9 Thus, we conclude that neither the residency requirement nor the office requirement of the local Rules is justified.10 Respondents contend that nonresident lawyers are not totally foreclosed from Eastern District practice because they can appear pro hac vice. In Piper, however, we recognized that this alternative does not allow the nonresident attorney to practice “on the same terms as a resident member of the bar.” 470 U. S., at 277, n. 2. An attorney not licensed by a district court must repeatedly file motions for each appearance on a pro hac vice basis. 594 F. Supp., at 1177. In addition, in order to appear pro hac vice under local Rule 21.5, a lawyer must also associate with a member of the Eastern District Bar, who is required to sign all court documents.11 594 F. Supp., at 1177. This association, of course, imposes a financial and administrative burden on nonresident counsel.12 9 For example, if a lawyer in Port Arthur, Texas, opened a branch office just across the state line in Lake Charles, Louisiana, he or she could join the Eastern District Bar even though that office was twice as far from the courthouse in New Orleans as is petitioner’s office. “Under Rule 21.3.1, a lawyer must maintain an in-state residence or office not only at the time of admission, but also for as long as the lawyer desires to remain a member of the Eastern District Bar. This Rule serves only to extend the unfairness of Rule 21.2. We therefore also find this local Rule to be unnecessary and irrational. 11 Under Rule 21.6, a District Court may grant a waiver of local-counsel association only if it would be a hardship for an out-of-state client. 12 From the lawyer’s standpoint, he or she will be at a significant disadvantage in attracting clients. Clients would have to be willing to provide FRAZIER v. HEEBE 651 641 Rehnquist, C. J., dissenting Furthermore, it is ironic that “local” counsel may be located much farther away from the New Orleans courthouse than the out-of-state counsel. Thus, the availability of appearance pro hac vice is not a reasonable alternative for an out-of-state attorney who seeks general admission to the Eastern District’s Bar.13 Reversed. Chief Justice Rehnquist, with whom Justice O’Connor and Justice Scalia join, dissenting. We have previously held that this Court may, in the exercise of its “supervisory authority,” modify or reverse judgments of lower federal courts in accordance with principles derived neither from the United States Constitution nor from any Act of Congress. United States v. Hasting, 461 U. S. 499, 505 (1983); Cupp v. Naughten, 414 U. S. 141, 146 (1973). Such a power, we have reasoned, inheres in any appellate court called upon “to review proceedings of trial courts and to reverse judgments of such courts which the appellate court concludes were wrong.” Ibid. In the present case the Court expands the notion of supervisory authority to allow it to review and revise local Rules of a District Court that regulate admission to the bar of that court. But it does not follow from the fact that we may reverse or modify a judgment of compensation for the necessary association with a local lawyer who will duplicate the principal lawyer’s efforts. The effect of such a rule is to drive up the cost of litigation and to steer business almost exclusively to the instate bar. A client may have a number of excellent reasons to select a nonlocal lawyer: his or her regular lawyer most familiar with the legal issues may be nonlocal; a nonresident lawyer may practice a specialty not available locally; or a client may be involved in an unpopular cause with which local lawyers are reluctant to be associated. See Piper, 470 U. S., at 281. 13 Furthermore, in many District Courts the decision on whether to grant pro hac vice status to an out-of-state attorney is purely discretionary and therefore is not a freely available alternative. See Supreme Court of New Hampshire v. Piper, supra, at 277, n. 2; Leis v. Flynt, 439 U. S. 438, 442 (1979). 652 OCTOBER TERM, 1986 Rehnquist, C. J., dissenting 482 U. S. another federal court which we believe to be wrong that we may set aside a rule promulgated by that court governing admission to its own bar on a similar basis. Congress has provided in 28 U. S. C. §2071 that the district courts may prescribe rules for the conduct of their business.1 It is clear from 28 U. S. C. § 1654 that the authority provided in § 2071 includes the authority of a district court to regulate the membership of its bar.2 See United States v. Hvass, 355 U. S. 570, 575 (1958). Neither these sections nor Federal Rule of Civil Procedure 83,3 which also governs the rulemaking power of district courts, gives any intimation that this Court possesses “supervisory power” over rules adopted in accordance with these provisions. Indeed, the history of these provisions demonstrates the broad discretion possessed by district courts in promulgating their own rules. At one 1 Section 2071 provides: “The Supreme Court and all courts established by Act of Congress may from time to time prescribe rules for the conduct of their business. Such rules shall be consistent with Acts of Congress and rules of practice and procedure prescribed by the Supreme Court.” 2 Section 1654 provides: “In all courts of the United States the parties may plead and conduct their own cases personally or by counsel as, by the rules of such courts, respectively, are permitted to manage and conduct causes therein.” 8 Federal Rule of Civil Procedure 83 provides: “Each district court by action of a majority of the judges thereof may from time to time, after giving appropriate public notice and an opportunity to comment, make and amend rules governing its practice not inconsistent with these rules. A local rule so adopted shall take effect upon the date specified by the district court and shall remain in effect unless amended by the district court or abrogated by the judicial council of the circuit in which the district is located. Copies of rules and amendments so made by any district court shall upon their promulgation be furnished to the judicial council and the Administrative Office of the United States Courts and be made available to the public. In all cases not provided for by rule, the district judges and magistrates may regulate their practice in any manner not inconsistent with these rules or those of the district in which they act.” FRAZIER v. HEEBE 653 641 Rehnquist, C. J., dissenting time former Equity Rule 79 required that district court rules be made “[w]ith the concurrence of a majority of the circuit judges for the circuit,” but that restriction was abolished by former 28 U. S. C. § 731 (1940 ed.), which provided the basis for Federal Rule of Civil Procedure 83. And no enabling Act has ever required the approval of this Court, or a majority of the Justices thereof, for the promulgation of district court rules. The Court finds that the Rules Enabling Act, 28 U. S. C. § 2072, “confirms” its power to decide whether local rules are rational and necessary. Ante, at 646, n. 4. That Act, however, has heretofore been regarded as statutory authorization for this Court’s promulgation of rules of procedure itself, and not as a grant of power to review the wisdom of rules adopted by a district court in default of any action by this Court. See, e. g., Burlington Northern R. Co. v. Woods, 480 U. S. 1, 5, n. 3 (1987); Hanna n. Plumer, 380 U. S. 460, 463-466, 471-474 (1965). To the extent that the Rules Enabling Act can be viewed as “confirming” this Court’s power to review the wisdom of district court rules, Federal Rule of Civil Procedure 83 suggests that this Court has apparently relinquished that power to the Judicial Councils of the Circuits. Rule 83, as recently amended in 1985, provides detailed procedures governing the adoption and amendment of district court rules. Under these procedures, a district court may make and amend rules by action of a majority of the judges of the court after notice and an opportunity for comment by the public are provided. The district court rules shall “remain in effect unless amended by the district court or abrogated by the judicial council of the circuit in which the district is located.” Fed. Rule Civ. Proc. 83 (emphasis added). If there were a role for this Court to entertain ad hoc challenges to district court rules on the basis of necessity or rationality alone, one would 654 OCTOBER TERM, 1986 Rehnquist, C. J., dissenting 482 U. S. think that it would have been provided for in the orderly procedures of Rule 83.4 Unquestionably the rule of a district court relating to membership in its bar may not violate the United States Constitution and must conform to any Act of Congress conferring authority in that respect. One denied admission to the bar by a rule which violates either the Constitution or an applicable statute may of course obtain review of that decision in this Court, and a reversal of the decision if his claims are well founded. But today’s decision rests upon no such grounds.5 Prior cases addressing challenges to the validity of local rules have confined their analyses to four inquiries: whether the rule conflicts with an Act of Congress; whether the rule conflicts with the rules of procedure promulgated by this Court; whether the rule is constitutionally infirm; and whether the subject matter governed by the rule is not within the power of a lower federal court to regulate. See, e. g., Colgrove v. Battin, 413 U. S. 149, 159-160, 162-164 (1973); Miner n. Atlass, 363 U. S. 641, 651-652 (1960); Story v. Liv 4 As noted by the Court, the Court of Appeals for the Fifth Circuit rejected petitioner’s request to exercise its authority under Rule 83 to invalidate local Rules 21.2 and 21.3.1, noting that the Fifth Circuit Judicial Conference is presently reviewing the local rules of the District Courts of the Circuit. In light of this pending review, the Court’s action today is particularly disruptive of the procedures established by Rule 83. 6 The Court declares its prerogative to review district court rules governing bar admission standards to determine whether they are consistent with “the principles of right and justice.” Ante, at 645. Yet the “law and justice” standard cited by the Court derives from cases in which this Court has reviewed attorney disbarment decisions by lower federal courts. See In re Ruffalo, 390 U. S. 544, 554 (1968) (White, J., concurring in result); Theard v. United States, 354 U. S. 278, 282 (1957); Selling v. Radford, 243 U. S. 46, 51 (1917). The Court is unable to cite an example in which this standard has been used to evaluate the validity of a local rule governing bar admission requirements. Although Theard n. United States, supra, and In re Ruffalo, supra, involved District Court and Court of Appeals rules governing disbarment proceedings, the validity of those rules was not questioned by the Court in those cases. FRAZIER v. HEEBE 655 641 Rehnquist, C. J., dissenting ingston, 13 Pet. 359, 368 (1839). The Court today does not suggest that the local Rules at issue here are invalid for any of these reasons, and instead determines merely that, in its view, the Rules are “unnecessary and irrational.” Ante, at 646, 650, n. 10. This newfound and quite unwarranted authority contrasts starkly with the observations of Chief Justice Marshall, writing for the Court in Ex parte Burr, 9 Wheat. 529 (1824): “Some doubts are felt in this Court respecting the extent of its authority as to the conduct of the Circuit and District Courts towards their officers; but without deciding on this question, the Court is not inclined to interpose, unless it were in a case where the conduct of the Circuit or District Court was irregular, or was flagrantly improper.” Id., at 530. The force behind the Court’s reluctance in Ex parte Burr to interfere with a lower court’s bar membership decision was its recognition that a federal court possesses nearly exclusive authority over such matters. Id., at 531. This recognition is reflected throughout this Court’s cases. See, e. g., Ex parte Secombe, 19 How. 9, 12-13 (1857); Ex parte Garland, 4 Wall. 333, 379 (1867); see also In re Snyder, 472 U. S. 634, 643 (1985). Petitioner contends that the local rules in question here violate the equal protection component of the Due Process Clause of the Fifth Amendment, but the Court, having waved its supervisory wand, finds it unnecessary to address this question. For the reasons stated by the Court of Appeals for the Fifth Circuit, I conclude that the local rules do not classify so arbitrarily or irrationally as to run afoul of the Fifth Amendment Due Process Clause. I would therefore affirm the judgment of the Court of Appeals. 656 OCTOBER TERM, 1986 Syllabus 482 U. S. GOODMAN et al. v. LUKENS STEEL CO. ET AL. CERTIORARI TO THE UNITED STATES COURT OF APPEALS FOR THE THIRD CIRCUIT No. 85-1626. Argued April 1, 1987—Decided June 19, 1987* In 1973, petitioners in No. 85-1626 (hereinafter petitioners), including individual employees of Lukens Steel Co. (Lukens), brought suit in Federal District Court against Lukens and the employees’ collective-bargaining agents (Unions), asserting racial discrimination claims under Title VII of the Civil Rights Act of 1964 and 42 U. S. C. § 1981. The court held that Pennsylvania’s 6-year statute of limitations governing contract claims applied to § 1981 claims, that Lukens had discriminated in certain respects, and that the Unions were also guilty of discriminatory practices in failing to challenge Lukens’ discriminatory discharges of probationary employees, in failing and refusing to assert instances of racial discrimination as grievances, and in tolerating and tacitly encouraging racial harassment. The court entered injunctive orders against Lukens and the Unions, reserving damages issues for further proceedings. The Court of Appeals held that Pennsylvania’s 2-year statute of limitations governing personal injury actions, rather than the 6-year statute, controlled the § 1981 claims, but affirmed the liability judgment against the Unions. Held: 1. The Court of Appeals was correct in selecting the Pennsylvania 2-year limitations period governing personal injury actions as the most analogous state statute of limitations to govern all § 1981 suits. Section 1981 speaks not only to personal rights to contract, but also to personal rights to sue, to testify, and to equal rights under all laws for the security of persons and property; and all persons are to be subject to like punishments, taxes, and burdens of every kind. Cf. Wilson v. Garcia, 471 U. S. 261. The Court of Appeals also properly concluded that the 2-year statute should be applied retroactively to petitioners here, even though the court overruled its prior 1977 and 1978 decisions that refused to apply Pennsylvania’s 2-year personal injury statute to the §1981 claims involved in those cases. Chevron Oil Co. v. Huson, 404 U. S. 97, advises that nonretroactivity is appropriate in certain circumstances, including when the decision overrules clear Circuit precedent on which the complaining party is entitled to rely. However, until the Court of *Together with No. 85-2010, United Steelworkers of America, AFL-CIO-CLC, et al. v. Goodman et al., also on certiorari to the same court. GOODMAN V. LUKENS STEEL CO. 657 656 Syllabus Appeals’ 1977 decision, there had been no authoritative specification of which statute of limitations applied to an employee’s § 1981 claims, and hence no clear precedent on which petitioners could have relied when they filed their complaint in 1973. As for the other Chevron factors, applying the 2-year statute here will not frustrate any federal law or result in inequity to the workers who are charged with knowledge that it was an unsettled question as to how far back from the date of filing their complaint the damages period would reach. Pp. 660-664. 2. The courts below properly held that the Unions violated Title VII and § 1981. Because both courts agreed on the facts pertaining to whether the Unions had treated blacks and whites differently and intended to discriminate on the basis of race, this Court will not examine the record, absent the Unions’ showing of extraordinary reasons for doing so. There is no merit to the Unions’ contention that the judgment rests on the erroneous legal premise that Title VII and § 1981 are violated if a union passively sits by and does not affirmatively oppose the employer’s racially discriminatory employment practices. In fact, both courts below concluded that the case against the Unions was one of more than mere acquiescence, in that the Unions deliberately chose not to assert claims of racial discrimination by the employer. Nor is there any merit to the argument that the only basis for Title VII liability was § 703(c)(3)’s prohibition against a union’s causing or attempting to cause illegal discrimination by an employer, which was not supported by the record. Both courts found that the Unions had discriminated on the basis of race by the way in which they represented the workers, and the Court of Appeals held that the deliberate choice not to process grievances violated § 703(c)(1), the plain language of which supports such conclusion. Furthermore, the District Court properly rejected the Unions’ explanation that, in order not to antagonize the employer, they did not include racial discrimination claims in grievances claiming other contract violations. A union that intentionally fails to assert discrimination claims, either to avoid antagonizing the employer and thus to improve chances of success on other issues, or in deference to the desires of its white membership, is liable under both Title VII and § 1981, regardless of whether, as a subjective matter, its leaders are favorably disposed toward minorities. Pp. 664-669. 777 F. 2d 113, affirmed. White, J., delivered the opinion of the Court, in which Rehnquist, C. J., and Stevens, J., joined, in Part I of which Powell and Scalia, JJ., joined, and in Part II of which Brennan, Marshall, and Black-mun, JJ., joined. Brennan, J., filed an opinion concurring in part and dissenting in part, in which Marshall and Blackmun, JJ., joined, 658 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. post, p. 669. Powell, J., filed an opinion concurring in part and dissenting in part, in which Scalia, J., joined, and in Parts I, II, III, and IV of which O’Connor, J., joined, post. p. 680. O’Connor, J., filed an opinion concurring in the judgment in part and dissenting in part, post, p. 689. Robert M. Weinberg argued the cause for petitioners in No. 85-2010 and for respondent Union in No. 85-1626. With him on the briefs were Julia Penny Clark, Michael H. Gottesman, Laurence Gold, David Silberman, Bernard Kleiman, and Carl Frankel. William H. Ewing argued the cause for petitioners in No. 85-1626 and for respondents in No. 85-2010. With him on the briefs for petitioners in No. 85-1626 were Arnold P. Borish and Daniel Segal. Messrs. Ewing, Borish, Segal, and Leslie A. Hayes filed a brief for respondents in No. 85-2010. t Justice White delivered the opinion of the Court. In 1973, individual employees1 of Lukens Steel Company (Lukens) brought this suit on behalf of themselves and others, asserting racial discrimination claims under Title VII of the Civil Rights Act of 1964, 78 Stat. 253, as amended, 42 U. S. C. §2000e et seq.,2 and 42 U. S. C. tBriefs of amici curiae urging reversal were filed for the United States by Solicitor General Fried, Assistant Attorney General Reynolds, Deputy Solicitor General Ayer, Deputy Assistant Attorney General Carvin, Roger Clegg, and David K. Flynn; and for the Lawyer’s Committee for Civil Rights Under Law et al. by Robert F. Mullen, Harold R. Tyler, James Robertson, Norman Redlich, William L. Robinson. Judith A. Winston, Richard T. Seymour, Joan Bertin, Judith L. Lichtman, Grover G. Hankins, Antonia Hernandez, and E. Richard Larson. Robert E. Williams and Douglas S. McDowell filed a brief for the Equal Employment Advisory Council as amicus curiae urging affirmance. ’The United Political Action Committee of Chester County was also a plaintiff in the case. 2 The part of Title VII relevant to the suit against the Unions is 42 U. S. C. § 2000e-2(c), which provides: “(c) Labor organization practices “It shall be an unlawful employment practice for a labor organization— GOODMAN V. LUKENS STEEL CO. 659 656 Opinion of the Court § 19813 against their employer and their collective-bargaining agents, the United Steelworkers of America and two of its local unions (Unions).4 After a bench trial, the District Court specified the periods for which Title VII claims could be litigated; it also reaffirmed a pretrial order that the Pennsylvania 6-year statute of limitations governing claims on contracts, replevin, and trespass6 applied to the §1981 claims and that claims with respect to the period after July 14, 1967, were accordingly not barred. On the merits, the District Court found that Lukens had discriminated in certain respects, but that in others plaintiffs had not made out a case.6 The District Court concluded that the Unions were also guilty of discriminatory practices, specifically in failing “(1) to exclude or to expel from its membership, or otherwise to discriminate against, any individual because of his race, color, religion, sex, or national origin; “(2) to limit, segregate, or classify its membership or applicants for membership, or to classify or fail or refuse to refer for employment any individual, in any way which would deprive or tend to deprive any individual of employment opportunities, or would limit such employment opportunities or otherwise adversely affect his status as an employee or as an applicant for employment, because of such individual’s race, color, religion, sex, or national origin; or “(3) to cause or attempt to cause an employer to discriminate against an individual in violation of this section.” 3 Section 1981 reads as follows: “All persons within the jurisdiction of the United States shall have the same right in every State and Territory to make and enforce contracts, to sue, be parties, give evidence, and to the full and equal benefit of all laws and proceedings for the security of persons and property as is enjoyed by white citizens, and shall be subject to like punishment, pains, penalties, taxes, licenses, and exactions of every kind, and to no other.” 4 United Steelworkers of America is the certified bargaining agent. The two locals act on its behalf. 5Pa. Stat. Ann., Tit. 12, §31 (Purdon 1931), repealed by Judiciary Act of 1976, Act No. 142, 1976 Pa. Laws 586. Under the 1976 Act, the new statute of limitations does not apply to claims arising prior to June 27, 1978. 6 The judgment against Lukens is not at issue in the cases brought here. 660 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. to challenge discriminatory discharges of probationary employees, failing and refusing to assert instances of racial discrimination as grievances, and in tolerating and tacitly encouraging racial harassment. 580 F. Supp. 1114 (ED Pa. 1984). The District Court entered separate injunctive orders against Lukens and the Unions, reserving damages issues for further proceedings. Lukens and the Unions appealed, challenging the District Court’s liability conclusions as well as its decision that the Pennsylvania 6-year statute of limitations, rather than the 2-year statute applicable to personal injuries, would measure the period of liability under § 1981. The Court of Appeals, differing with the District Court, held that the 2-year statute of limitations controlled but affirmed the liability judgment against the Unions. 777 F. 2d 113 (CA3 1985).7 The employees’ petition for certiorari in No. 85-1626 challenged the Court of Appeals’ choice of the § 1981 limitations period. The Unions’ petition in No. 85-2010 claimed error in finding them liable under Title VII and §1981. We granted both petitions, 479 U. S. 982 (1986). We address in Part I the limitations issue in No. 85-1626 and the Unions’ liability in Part II. I Because § 1981, like §§ 1982 and 1983, does not contain a statute of limitations, federal courts should select the most appropriate or analogous state statute of limitations. Wilson v. Garcia, 471 U. S. 261, 266-268 (1985); Runyon v. McCrary, 427 U. S. 160, 180-182 (1976); Johnson v. Railway Express Agency, Inc., 421 U. S. 454, 462 (1975). In Wilson, the reach of which is at issue in this case, there were three 7 Judge Garth dissented on the question of which statute of limitations to apply to the workers’ § 1981 claim. 777 F. 2d, at 130. He acknowledged that all § 1981 claims should be treated the same; but in his view, § 1981 claims involved injury to economic rights and the personal injury characterization adopted by the Court in Wilson was ill suited for claims arising under § 1981. GOODMAN V. LUKENS STEEL CO. 661 656 Opinion of the Court holdings: for the purpose of characterizing a claim asserted under § 1983, federal law, rather than state law, is controlling; a single state statute of limitations should be selected to govern all § 1983 suits; and because claims under § 1983 are in essence claims for personal injury, the state statute applicable to such claims should be borrowed. Petitioners in No. 85-1626 (hereafter petitioners), agree with the Court of Appeals that the first two Wilson holdings apply in §1981 cases, but insist that the third does not. Their submission is that § 1981 deals primarily with economic rights, more specifically the execution and enforcement of contracts, and that the appropriate limitations period to borrow is the one applicable to suits for interference with contractual rights, which in Pennsylvania was six years. The Court of Appeals properly rejected this submission. Section 1981 has a much broader focus than contractual rights. The section speaks not only of personal rights to contract, but personal rights to sue, to testify, and to equal rights under all laws for the security of persons and property; and all persons are to be subject to like punishments, taxes, and burdens of every kind. Section 1981 of the present Code was § 1977 of the Revised Statutes of 1874. Its heading was and is “Equal rights under the law” and is contained in a chapter entitled “Civil Rights.” Insofar as it deals with contracts, it declares the personal right to make and enforce contracts, a right, as the section has been construed, that may not be interfered with on racial grounds. The provision asserts, in effect, that competence and capacity to contract shall not depend upon race. It is thus part of a federal law barring racial discrimination, which, as the Court of Appeals said, is a fundamental injury to the individual rights of a person. Wilson’s characterization of §1983 claims is thus equally appropriate here, particularly since §1983 would reach state action that encroaches on the rights protected by § 1981. That § 1981 has far-reaching economic consequences does not change this conclusion, since such impact flows from 662 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. guaranteeing the personal right to engage in economically significant activity free from racially discriminatory interference. The Court of Appeals was correct in selecting the Pennsylvania 2-year limitations period governing personal injury actions. We also agree with the Court of Appeals that the 2-year statute, adopted in compliance with Wilson, should be applied in this case. The usual rule is that federal cases should be decided in accordance with the law existing at the time of decision. Gulf Offshore Co. n. Mobil Oil Corp., 453 U. S. 473, 486, n. 16 (1981); Thorpe v. Housing Authority of Durham, 393 U. S. 268, 281 (1969); United States v. Schooner Peggy, 1 Cranch 103, 109 (1801). But Chevron Oil Co. n. Huson, 404 U. S. 97 (1971), advises that nonretroactivity is appropriate in certain defined circumstances. There the Court held that a decision specifying the applicable state statute of limitations in another context should not be applied retroactively because the decision overruled clear Circuit precedent on which the complaining party was entitled to rely, because the new limitations period had been occasioned by a change in the substantive law the purpose of which would not be served by retroactivity, and because retroactive application would be inequitable. Petitioners argue that the same considerations are present here. We disagree. It is true, as petitioners point out, that the Court of Appeals decision in this case overruled prior Third Circuit cases, Meyers n. Pennypack Woods Home Ownership Assn., 559 F. 2d 894 (1977); Davis n. United States Steel Supply, Div. of United States Steel Corp., 581 F. 2d 335, 338, 341, n. 8 (1978), each of which had refused to apply the Pennsylvania 2-year personal injury statute of limitations to the §1981 claims involved in those cases. But until Meyers was decided in 1977, there had been no authoritative specification of which statute of limitations applied to an employee’s § 1981 claims, and hence no clear precedent on which petitioners GOODMAN V. LUKENS STEEL CO. 663 656 Opinion of the Court could have relied when they filed their complaint in this case in 1973. In a later case, Al-Khazraji n. St. Francis College, 784 F. 2d 505, 512-514 (1986), the Court of Appeals refused to apply retroactively the same 2-year statute in an employment discrimination § 1981 case because the case was filed when clear Circuit precedent specified a longer statute. Distinguishing its decision there from the case now before us, the Court of Appeals said: “In 1973, when the complaint was filed in the Goodman case, there was no established precedent in the Third Circuit to indicate the appropriate limitations period for Section 1981 claims.” 784 F. 2d, at 512. It was obviously for this reason that the Court of Appeals here said that its decision “should be given the customary retroactive effect.” 777 F. 2d, at 120. The court cited its prior decision in Smith v. Pittsburgh, 764 F. 2d 188 (1985),8 a post-Wilson case in which the Court of Appeals applied retroactively the 2-year statute in a § 1983 employment termination case because of the unsettled law in the Third and other Circuits. As for the remainder of the Chevron factors, applying the 2-year personal injury statute, which is wholly consistent with Wilson n. Garcia and with the general purposes of statutes of repose, will not frustrate any federal law or result in inequity to the workers who are charged with knowledge that it was an unsettled question as to how far back from the date of filing their complaint the damages period would 8 In the Smith case, the Third Circuit applied our three-part test in Chevron, in concluding that Wilson should be applied to the case then before it. The court remarked: “We have held that where application of the law had been erratic and inconsistent, without clear precedent on which plaintiff could reasonably rely in waiting to file suit, a subsequent Supreme Court decision on the applicable limitations period cannot be said to have overruled clear past precedent on which the litigants may have relied.” 764 F. 2d, at 194-195. The court went on to note that at the time plaintiffs in that case filed suit, the Third Circuit had not ruled definitively on which limitations period applied to the particular § 1983 claim at issue there. 664 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. reach. Accordingly, the Court of Appeals properly applied the 2-year statute of limitations to the present case.9 II This case was tried for 32 days in 1980. One-hundred fifty-seven witnesses testified and over 2,000 exhibits were introduced. On February 13, 1984, the District Court filed its findings and conclusions. In an introductory section discussing the relevant legal principles, the trial judge discussed, among other things, the nature of “disparate treatment” and “disparate impact” cases under Title VII, recognizing that in the former the plaintiff must prove not only disparate treatment, but trace its cause to intentional racial discrimination, an unnecessary element in disparate-impact cases. The District Court also emphasized that proof of discriminatory intent is crucial in §1981 cases and that such intent cannot be made out by showing only facially neutral conduct that burdens one race more than another. The District Court proceeded to find that the company had violated Title VII in several significant respects, including the discharge of employees during their probationary period, the toleration of racial harassment by employees, initial job assignments, promotions, and decisions on incentive pay. The court also found that in these identical ways the company had also violated § 1981, a finding the court could not have made without concluding that the company had intentionally discriminated on a racial basis in these respects. Similarly, the Unions were found to have discriminated on racial grounds in violation of both Title VII and § 1981 in certain ways: failing to challenge discriminatory discharges of probationary employees; failure and refusal to assert racial 9 The Court of Appeals recognized that giving retroactive effect to its statute of limitations holding would require reexamination of some of the liability determinations by the District Court in light of the shorter limitations period. GOODMAN V. LUKENS STEEL CO. 665 656 Opinion of the Court discrimination as a ground for grievances; and toleration and tacit encouragement of racial harassment. What the conduct of the Unions had been and whether they had treated blacks and whites differently were questions of historical fact that Federal Rule of Civil Procedure 52(a) enjoins appellate courts to accept unless clearly erroneous. So is the issue of whether the Unions intended to discriminate based on race. Anderson v. Bessemer City, 470 U. S. 564, 574 (1985); Pullman-Standard v. Swint, 456 U. S. 273, 287-288 (1982). The Court of Appeals did not set aside any of the District Court’s findings of fact that are relevant to this case. That is the way the case comes to us, and both courts below having agreed on the facts, we are not inclined to examine the record for ourselves absent some extraordinary reason for undertaking this task. Nothing the Unions have submitted indicates that we should do so. “A court of law, such as this Court is, rather than a court for correction of errors in factfinding, cannot undertake to review concurrent findings of fact by two courts below in the absence of a very obvious and exceptional showing of error.” Graver Mfg. Co. v. Linde Co., 336 U. S. 271, 275 (1949). See also United States v. Ceccolini, 435 U. S. 268, 273 (1978). Unless there are one or more errors of law inhering in the judgment below, as the Unions claim there are, we should affirm it. The Unions contend that the judgment against them rests on the erroneous legal premise that Title VII and § 1981 are violated if a union passively sits by and does not affirmatively oppose the employer’s racially discriminatory employment practices. It is true that the District Court declared that mere union passivity in the face of employer discrimination renders the union liable under Title VII and, if racial animus is properly inferrable, under § 1981 as well.”10 We need not 10 The first part of this statement must have been addressed to disparate impact, for discriminatory motive is required in disparate-treatment Title VII cases as it is in § 1981 claims. See Teamsters v. United States, 431 U. S. 324, 335-336, n. 15 (1977); General Building Contractors Assn., 666 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. discuss this rather abstract observation, for the court went on to say that the evidence proves “far more” than mere passivity.11 As found by the court, the facts were that since 1965, the collective-bargaining contract contained an express clause binding both the employer and the Unions not to discriminate on racial grounds; that the employer was discriminating against blacks in discharging probationary employees, which the Unions were aware of but refused to do anything about by way of filing proffered grievances or otherwise; that the Unions had ignored grievances based on instances of harassment which were indisputably racial in nature; and that the Unions had regularly refused to include assertions of racial discrimination in grievances that also asserted other contract violations.12 In affirming the District Court’s findings against the Unions, the Court of Appeals also appeared to hold that the Inc. v. Pennsylvania, 458 U. S. 375, 391 (1982). Because the District Court eventually found that in each respect the Unions violated both Title VII and § 1981 in exactly the same way, liability did not rest on a claim under Title VII that did not rest on intentional discrimination. 11 The District Court commented that there was substantial evidence, related to events occurring prior to the statute of limitations period, which “casts serious doubt on the unions’ total commitment to racial equality.” 580 F. Supp. 1114, 1157 (ED Pa. 1984). The District Court noted that it was the company, not the Unions, which pressed for a nondiscrimination clause in the collective-bargaining agreement. The District Court found that the Unions never took any action over the segregated locker facilities at Lukens and did not complain over other discriminatory practices by the company. The District Court found that when one employee approached the president of one of the local unions to complain about the segregated locker facilities in 1962, the president dissuaded him from complaining to the appropriate state agency. The District Court, however, found “inconclusive” the evidence offered in support of the employees’ claim that the Unions’ discriminated against blacks in their overall handling of grievances under the collective-bargaining agreement. 12 The District Court also found that although the Unions had objected to the company’s use of certain tests, they had never done so on racial grounds, even though they “were certainly chargeable with knowledge that many of the tests” had a racially disparate impact. Id., at 1159. GOODMAN v. LUKENS STEEL CO. 667 656 Opinion of the Court Unions had an affirmative duty to combat employer discrimination in the workplace. 777 F. 2d, at 126-127. But it, too, held that the case against the Unions was much stronger than one of mere acquiescence in that the Unions deliberately chose not to assert claims of racial discrimination by the employer. It was the Court of Appeals’ view that these intentional and knowing refusals discriminated against the victims who were entitled to have their grievances heard. The Unions submit that the only basis for any liability in this case under Title VII is § 703(c)(3), which provides that a Union may not “cause or attempt to cause an employer to discriminate against an individual in violation of this section,” 78 Stat. 256, 42 U. S. C. § 2000e-2(c)(3), and that nothing the District Court found and the Court of Appeals accepted justifies liability under this prohibition. We need not differ with the Unions on the reach of § 703(c)(3), for § 703(c)(1) makes it an unlawful practice for a Union to “exclude or to expel from its membership, or otherwise to discriminate against, any individual because of his race, color, religion, sex, or national origin.” 78 Stat. 255, 42 U. S. C. § 2000-2(c)(l). (Emphasis added.) Both courts below found that the Unions had indeed discriminated on the basis of race by the way in which they represented the workers, and the Court of Appeals expressly held that “[t]he deliberate choice not to process grievances also violated § 703(c)(1) of Title VII.” 777 F. 2d, at 127. The plain language of the statute supports this conclusion. The Court of Appeals is also faulted for stating that the Unions had violated their duty of fair representation, which the Unions assert has no relevance to this case. But we do not understand the Court of Appeals to have rested its affirmance on this ground, for as indicated above, it held that the Unions had violated § 703. The Unions insist that it was error to hold them liable for not including racial discrimination claims in grievances claiming other violations of the contract. The Unions followed 668 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. this practice, it was urged, because these grievances could be resolved without making racial allegations and because the employer would “get its back up” if racial bias was charged, thereby making it much more difficult to prevail. The trial judge, although initially impressed by this seemingly neutral reason for failing to press race discrimination claims, ultimately found the explanation “unacceptable” because the Unions also ignored grievances which involved racial harassment violating the contract covenant against racial discrimination but which did not also violate another provision. The judge also noted that the Unions had refused to complain about racially based terminations of probationary employees, even though the express undertaking not to discriminate protected this group of employees, as well as others, and even though, as the District Court found, the Unions knew that blacks were being discharged at a disproportionately higher rate than whites. In the judgment of the District Court, the virtual failure by the Unions to file any race-bias grievances until after this lawsuit started, knowing that the employer was practicing what the contract prevented, rendered the Unions’ explanation for their conduct unconvincing.13 As we understand it, there was no suggestion below that the Unions held any racial animus against or denigrated blacks generally. Rather, it was held that a collectivebargaining agent could not, without violating Title VII and 13 The District Court also rejected the Unions’ argument that much of the workers’ case involved discrimination by the company in making initial job assignments, and that it had no control over those assignments. The court found that once hired, new employees were entitled to the protection of the collective-bargaining agreement, including the protection afforded by the nondiscrimination clause: “To require blacks to continue to work in lower paying and less desirable jobs, in units disparately black, is to discriminate against them in violation of the collective bargaining agreement (and, of course, also in violation of Title VII). It is very clear, on the record in this case, that the defendant unions never sought to avail themselves of this rather obvious mechanism for protecting the interests of their members.” 580 F. Supp., at 1160. GOODMAN V. LUKENS STEEL CO. 669 656 Opinion of Brennan, J. §1981, follow a policy of refusing to file grievable racial discrimination claims however strong they might be and however sure the agent was that the employer was discriminating against blacks. The Unions, in effect, categorized racial grievances as unworthy of pursuit and, while pursuing thousands of other legitimate grievances, ignored racial discrimination claims on behalf of blacks, knowing that the employer was discriminating in violation of the contract. Such conduct, the courts below concluded, intentionally discriminated against blacks seeking a remedy for disparate treatment based on their race and violated both Title VII and § 1981. As the District Court said: “A union which intentionally avoids asserting discrimination claims, either so as not to antagonize the employer and thus improve its chances of success on other issues, or in deference to the perceived desires of its white membership, is liable under both Title [VII] and §1981, regardless of whether, as a subjective matter, its leaders were favorably disposed toward minorities.” 580 F. Supp., at 1160. The courts below, in our view, properly construed and applied Title VII and § 1981. Those provisions do not permit a union to refuse to file any and all grievances presented by a black person on the ground that the employer looks with disfavor on and resents such grievances. It is no less violative of these laws for a union to pursue a policy of rejecting disparate-treatment grievances presented by blacks solely because the claims assert racial bias and would be very troublesome to process. In both Nos. 85-1626 and 85-2010, the judgment of the Court of Appeals is affirmed. It is so ordered. Justice Brennan, with whom Justice Marshall and Justice Blackmun join, concurring in part and dissenting in part. I join Part II of the Court’s opinion, affirming the Court of Appeals’ decision that the Unions engaged in race discrimina 670 OCTOBER TERM, 1986 Opinion of Brennan, J. 482 U. S. tion in violation of 42 U. S. C. § 1981 and Title VII of the Civil Rights Act of 1964. I dissent, however, from Part I, which characterizes all §1981 actions as tort actions, and holds that they are subject to state statutes of limitations for personal injury. Section 1981, in its original conception and its current application, is primarily a proscription of race discrimination in the execution, administration, and enforcement of contracts. Our analysis in Wilson v. Garcia, 471 U. S. 261 (1985), requires us to hold, therefore, that § 1981 actions are governed by state statutes of limitations for interference with contractual relations. I In Wilson, the Court had to determine the most appropriate statute of limitations to apply to claims brought under § 1 of the Civil Rights Act of 1871, now codified at 42 U. S. C. § 1983. First, the Court decided that characterization of a § 1983 action, for the purpose of selecting a state statute of limitations, was a matter of federal law. 471 U. S., at 268-271. The Court then held that the federal interest in “uniformity, certainty, and the minimization of unnecessary litigation” required that all §1983 actions receive a single broad characterization for statute-of-limitations purposes. Id., at 275. For reasons identical to those stated in Wilson, the Court today concludes that § 1981, like § 1983, must receive a single broad characterization for statute-of-limitations purposes. I agree. The Court goes on to hold, however, that claims brought under §§ 1983 and 1981 should receive the same characterization, and here I part company with the Court. In Wilson, the Court relied on the history of § 1983 in its determination that claims under the statute were best characterized as tort actions for damages resulting from personal injury. The Court observed that § 1983, originally known as the Ku Klux Act, was enacted as part of the Civil Rights Act of 1871, and that the “specific historical catalyst” for § 1983 GOODMAN v. LUKENS STEEL CO. 671 656 Opinion of Brennan, J. was “the campaign of violence and deception in the South, fomented by the Ku Klux Klan.” Id., at 276. The Court highlighted the legislative history of § 1983, which made clear that Congress was attempting to stop a wave of murders, lynchings, and whippings and to eliminate “the refuge that local authorities extended to the authors of these outrageous incidents.” Ibid. From this, the Court concluded that “[t]he atrocities that concerned Congress in 1871 plainly sounded in tort.” Id., at 277. More specifically, the Court determined that, among the many types of tort claims filed under § 1983, the “action for the recovery of damages for personal injuries” was the most analogous common-law cause of action. Id., at 276. Performing a like historical analysis of § 1981, I conclude that it should be characterized as an action for recovery of damages for interference with contractual relations. Section 1981, originally enacted as § 1 of the Civil Rights Act of 1866,1 presently provides: “All persons within the jurisdiction of the United States shall have the same right in every State and Territory to make and enforce contracts, to sue, be parties, give evidence, and to the full and equal benefit of all laws and proceedings for the security of persons and property as is enjoyed by white citizens, and shall be subject to like punishment, pains, penalties, taxes, licenses, and exactions of every kind, and to no other.” Clearly, the “full and equal benefit” and “punishment” clauses guarantee numerous rights other than equal treatment in the execution, administration, and the enforcement of contracts. In this sense § 1981, like § 1983, is broadly concerned with “the equal status of every ‘person.’” Wilson, supra, at 277 (emphasis in original). But § 1981 was primar 1 It was reenacted, with minor changes, as § 16 of the Act of May 31, 1870, 16 Stat. 144, and was recodified in 1874. See Runyon v. McCrary, 427 U. S. 160, 168-169, n. 8 (1976). 672 OCTOBER TERM, 1986 Opinion of Brennan, J. 482 U. S. ily intended, and has been most frequently utilized, to remedy injury to a narrower category of contractual or economic rights. The main targets of the Civil Rights Act of 1866 were the “Black Codes,” enacted in Southern States after the Thirteenth Amendment was passed.2 Congress correctly perceived that the Black Codes were in fact poorly disguised substitutes for slavery: “They defined racial status; forbade blacks from pursuing certain occupations or professions (e. g. skilled artisans, merchants, physicians, preaching without a license); forbade owning firearms or other weapons; controlled the movement of blacks by systems of passes; 2 See B. Schwartz, From Confederation to Nation: The American Constitution 1835-1877, p. 191 (1973) (“The purpose of the act as explained by Lyman Trumbull, chairman of the Senate Judiciary Committee, in his address introducing the proposed legislation, was to carry into effect the Thirteenth Amendment by destroying the discrimination against the Negro that existed in the laws of the southern states, particularly the Black Codes enacted since emancipation”); id., at 193 (“Before the Thirteenth Amendment, slaves could not own property, and after emancipation the southern states enacted Black Codes to perpetuate this disability. This was the ‘incident of slavery’ which the 1866 statute was aimed at, relying for its enforcement on the Thirteenth Amendment”); 6 C. Fairman, History of the Supreme Court of the United States: Reconstruction and Reunion, 1864-1888, p. 110 (1971) (“Eight Southern legislatures were in session at some time in December 1865. Each addressed itself to the status of the Negro. . . . The Southern States had spoken, and the impact was felt in Congress from the moment it assembled. In a major aspect, the problem was economic”); K. Stampp, The Era of Reconstruction 1865-1877, p. 123 (1965) (“This condition of economic helplessness ... enabled the white landholders, with the aid of the Black Codes, to re-establish bondage in another form. The congressional Committee on Reconstruction heard a great deal of convincing testimony about the use of southern vagrancy laws and various extra-legal coercive devices to force Negroes back into agricultural labor under strict discipline. This testimony suggested that there was a close relationship between the securing of civil and political rights on the one hand and the establishment of economic independence on the other”). GOODMAN v. LUKENS STEEL CO. 673 656 Opinion of Brennan, J. required proof of residence; prohibited the congregation of groups of blacks; restricted blacks from residing in certain areas; and specified an etiquette of deference to whites, as, for example, by prohibiting blacks from directing insulting words at whites.” H. Hyman & W. Wiecek, Equal Justice Under Law 319 (1982).3 In addition, the “formidable hand of custom,” id., at 321, interposed itself between blacks and economic independence, forcing Congress to move against private, as well as state-sanctioned economic discrimination. See generally Runyon v. McCrary, 427 U. S. 160 (1976); Kohl, The Civil Rights Act of 1866, Its Hour Come Round At Last: Jones n. Alfred H. Mayer Co., 55 Va. L. Rev. 272, 279 (1969).4 Obviously, both the Black Codes and longstanding custom imposed a number of discriminatory prohibitions that were noneconomic, and the 39th Congress therefore had significant ’The Black Codes had “attenuated counterparts” in some Northern States, usually “prohibiting the ingress of blacks into the state, imposing Jim Crow in public facilities, or prohibiting blacks from voting.” H. Hyman & W. Wiecek, Equal Justice Under Law 320 (1982). 4 It has been pointed out that “the Black Codes told only part of the story” of the attempt to prevent blacks from controlling their own labor. Kohl, The Civil Rights Act of 1866, Its Hour Come Round At Last: Jones v. Alfred H. Mayer Co., 55 Va. L. Rev. 272, 279 (1969). The Joint Committee on Reconstruction heard testimony demonstrating that, even apart from the restrictions of formal law, black access to land and labor markets was in practice severely limited, that physical compulsion was used to force freedmen to sign employment contracts at low rates, that cartels of white plantation owners fixed the wages of black workers by agreement, and that whites refused to sell land to blacks. See id., at 279-283; see also Report of C. Schurz, S. Exec. Doc. No. 2, 39th Cong., 1st Sess., 24 (1865) (“The opposition to the negro’s controlling his own labor, carrying on business independently on his own account—in one word, working for his own benefit—showed itself in a variety of ways”). Section 1981 banned racial discrimination in contractual relations, whether individuals were expressly or constructively denied the right to contract because of race, or were provided a lesser opportunity than others, in the form of less favorable contract terms or unequal treatment, discouraging entry into contractual relations. 674 OCTOBER TERM, 1986 Opinion of Brennan, J. 482 U. S. concerns that lay outside the economic realm. Nonetheless, as the Court has often acknowledged,6 the Legislature’s central concerns in 1866 revolved around actions taken by the States and by private parties which consigned black Americans to lives of perpetual economic subservience to their former masters. These concerns were often denominated “civil rights” because, in the mid-19th century, “civil rights were commonly defined, especially by lawyers, as primarily economic.” Hyman & Wiecek, supra, at 299.6 Congress clearly believed that freedom would be empty for black men and women if they were not also assured an equal opportunity to engage in business, to work, and to bargain for sale of their labor. In the debates, it emerged time and again that Congress sought to identify and guarantee those rights that would enable a person to sustain an independent economic unit (a family) once the master-slave relation had been dismantled: 6 See General Building Contractors Assn. v. Pennsylvania, 458 U. S. 375, 386 (1982) (“The principal object of the legislation was to eradicate the Black Codes, laws enacted by Southern legislatures imposing a range of civil disabilities on freedmen”); Runyon, 427 U. S., at 172 (racial discrimination in the making and enforcement of contracts for education is a “classic violation of § 1981”); McDonald v. Santa Fe Trail Transportation Co., 427 U. S. 273, 295 (1976) (Section 1981 prohibits “discrimination in the making or enforcement of contracts”); Johnson v. Railway Express Agency, Inc., 421 U. S. 454, 459 (1975) (Section 1981 “on its face relates primarily to racial discrimination in the making and enforcement of contracts”); cf. Jones v. Alfred H. Mayer Co., 392 U. S. 409, 443 (1968) (the rights protected by 42 U. S. C. § 1982 would be mere “paper guarantee[s]” if Congress could not “assure that a dollar in the hands of a Negro will purchase the same thing as a dollar in the hands of a white man”). 6 See also Hyman & Wiecek, supra, at 300 (“There were many civil rights. How many, no one knew, although lawyers tended to classify them neatly in terms of primarily economic, contract relationships”); id., at 301 (“The right Americans . . . enjoyed[,] the opportunity to enter into almost limitless civil relationships, and to gain or lose from these involvements was considered a precious right. This right underlay what Republicans meant by free labor”). GOODMAN V. LUKENS STEEL CO. 675 656 Opinion of Brennan, J. “[Section 1981’s] object is to secure to a poor, weak class of laborers the right to make contracts for their labor, the power to enforce the payment of their wages, and the means of holding and enjoying the proceeds of their toil.” Cong. Globe, 39th Cong., 1st Sess., 1159 (1866) (Rep. Windom). “It is idle to say that a citizen shall have the right to life, yet to deny him the right to labor, whereby he alone can live. It is a mockery to say that a citizen may have a right to live, and yet deny him the right to make a contract to secure the privilege and the rewards of labor. It is worse than mockery to say that men may be clothed by the national authority with the character of citizens, yet may be stripped by State authority of the means by which citizens may exist.” Id., at 1833 (Rep. Lawrence).7 ’There are many passages to similar effect. See Cong. Globe, 39th Cong., 1st Sess., 1151 (1866) (Rep. Thayer): “Sir, if it is competent for the new-formed Legislatures of the rebel States to enact laws which oppress this large class of people who are dependent for protection upon the United States Government, to retain them still in a state of real servitude; if it is practicable for these Legislatures to pass laws and enforce laws which reduce this class of people to the condition of bondmen; laws which prevent the enjoyment of the fundamental rights of citizenship; laws which declare, for example, that they shall not have the privilege of purchasing a home for themselves and their families; laws which impair their ability to make contracts for labor in such manner as virtually to deprive them of the power of making such contracts, and which then declare them vagrants because they have no homes and because they have no employment; I say, if it is competent for these Legislatures to pass and enforce such laws, then I demand to know, of what practical value is the amendment abolishing slavery in the United States?” See also id., at 1160 (Rep. Windom): “[Blacks] are denied a home in which to shelter their families, prohibited from carrying on any independent business, and then arrested and sold as vagrants because they have no homes and no business. ‘Planters combine together to compel them to work for such wages as their former masters may dictate, and deny them the privilege of hiring to any one without the consent of the master; and in order to make it impossi- 676 OCTOBER TERM, 1986 Opinion of Brennan, J. 482 U. S. The historical origins of § 1981 therefore demonstrate its dominant concern with economic rights. The preeminence of this concern is even clearer if one looks at § 1981 in conjunction with 42 U. S. C. § 1982, which was simultaneously enacted. The plain language of § 1982 speaks squarely and exclusively to economic rights and relations. It provides that “[a]ll citizens of the United States shall have the same right, in every State and Territory, as is enjoyed by white citizens thereof to inherit, purchase, lease, sell, hold, and convey real and personal property.” Both §§ 1981 and 1982 were derived from § 1 of the Civil Rights Act of 1866; their wording and their identical legislative history have led the Court to construe them similarly. See Runyon, 427 U. S., at 171-173; Tillman n. Wheaton-Haven Recreation Assn., Inc., 410 U. S. 431, 440 (1973).8 Looking at §§1981 and 1982 in tandem, it is apparent that the primary thrust of the 1866 Congress was the provision of equal rights and treatment in the matrix of contractual and quasi-contractual relationships that form the economic sphere. The Court maintains that § 1981 must be characterized as a personal injury action because it is “part of a federal law barring racial discrimination,” which is “a fundamental injury to ble for them to seek employment elsewhere, the pass system is still enforced. ... Do you call that man free who cannot choose his own employer, or name the wages for which he will work? Do you call him a freeman who is denied the most sacred of all possessions, a home? Is he free who cannot bring a suit in court for the defense of his rights? Sir, if this be liberty, may none ever know what slavery is.” 8 The Court has previously acknowledged and relied upon the differing legislative histories and purposes of §§ 1981 and 1982 on the one hand, and § 1983 on the other, to demonstrate that the statutes should receive differing interpretations. See District of Columbia v. Carter, 409 U. S. 418 (1973) (holding that the District of Columbia is not a “State or Territory” under § 1983, although it is under § 1982). Cf. Monroe v. Pape, 365 U. S. 167, 205-206 (1961) (Frankfurter, J., dissenting in part) (“Different problems of statutory meaning are presented by two enactments deriving from different constitutional sources”). GOODMAN V. LUKENS STEEL CO. 677 656 Opinion of Brennan, J. the individual rights of a person.” Ante, at 661. If this reasoning is the real basis of Wilson, its historical analysis was completely superfluous. Any act of racism doubtless inflicts personal injury. At its core, it is an act of violence—a denial of another’s right to equal dignity.9 In many contexts, therefore, racially discriminatory acts are violations of federal civil rights laws. But the availability of a federal forum should not obscure the fact that the type of injury inflicted by discrimination will vary. Discrimination may overlap with almost all categories of legal claims, but does not wholly embrace any one. An assault, a breach of contract, an infliction of emotional distress, an unjust discharge, a refusal to hire or promote—all may be motivated by racial discrimination, but they are not for that reason the same type of legal claim. Bringing a claim under a civil rights Act should not alter this fact. Our analysis in Wilson requires us to differentiate between race discrimination that results in a tort and race discrimination that interferes with contractual relations.10 9 The tortious aspect of racial discrimination has been noted by the Court in Curtis v. Loether, 415 U. S. 189, 196, n. 10 (1974) (citing C. Gregory & H. Kalven, Cases and Materials on Torts 961 (2d ed. 1969)), in which Justice Marshall suggested that racial discrimination might eventually be treated as a “dignitary tort.” 10 The Court appears to argue that because “§ 1983 would reach state action that encroaches on the rights protected by § 1981,” ante, at 661, it is important that they have the same statute of limitations. As Judge Garth demonstrated below, this argument is without merit: “It is true that the same nucleus of operative fact sometimes could be characterized as either a § 1981 and/or § 1983 claim and thereby receive different limitations treatment if [a different statute were] applied under § 1981. Such variations, however, are commonplace in the law. In a run-of-the-mill automobile accident case, for example, identical facts could give nse to warranty claims sounding in contract and strict liability claims sounding in tort—each to be governed by a different statute of limitations. This is not thought to be a ‘bizarre result,’ and the possibility that the same or similar facts could support causes of action under different Civil Rights statutes is no more ‘bizarre.’ ” 777 F. 2d 113, 136 (CA3 1985). 678 OCTOBER TERM, 1986 Opinion of Brennan, J. 482 U. S. This Court has acknowledged the central theme of § 1981: “the Act was meant, by its broad terms, to proscribe discrimination in the making or enforcement of contracts against, or in favor of, any race.” McDonald n. Santa Fe Trail Transportation Co., 427 U. S. 273, 295 (1976). We should recognize this primary historical concern again today and, as Wilson requires, reflect it in our choice of the appropriate state statute of limitations for § 1981 claims. II Even aside from its inconsistency with the intent of the 39th Congress, the application of the state statute of limitations for personal injury to § 1981 actions is the wrong choice as a practical matter. An overwhelming number of § 1981 actions concern enforcement of economic rights. See Comment, Developments in the Law—Section 1981, 15 Harv. Civ. Rights-Civ. Lib. L. Rev. 29, 34 (1980) (“Plaintiffs in section 1981 suits have relied predominantly on the statute’s guarantee of the right to contract free from racial discrimination”); see also Brief for Petitioners in No. 85-1626, pp. 18-19. It is well known that States apply different, usually longer limitations periods to contractual claims than to those sounding in tort.11 Personal injury actions are often based upon a single, dramatic event, and depend upon evidence of physical injury or eyewitness testimony that becomes less accessible and less trustworthy with the passage of time. In contrast, contract actions or injury to economic relations may involve an extended relationship between parties and may be supported by documentary evidence. See, e. g., Comment, 15 Harv. Civ. 11A longer statute of limitations might actually reduce federal litigation. Cases arising under the Fair Housing Act of 1968, 42 U. S. C. § 3601 et seq. (1982 ed. and Supp. Ill), and Title VII of the Civil Rights Act of 1964, 42 U. S. C. § 2000e et seq., are likely to overlap with § 1981 claims. If a short limitations period is imposed, plaintiffs in such cases will be forced to file their suits before exhausting administrative remedies, for fear of running out of time. GOODMAN v. LUKENS STEEL CO. 679 656 Opinion of Brennan, J. Rights-Civ. Lib. L. Rev., supra, at 225, 228. Obviously, a breach of contract or a refusal to contract resulting from race discrimination can occur in one traumatic moment, but, as a general rule, state legislatures have concluded that contract actions frequently have an evidentiary foundation with a greater life expectancy, and thus warrant a longer limitations period.12 The Court has said that “the length of the period allowed for instituting suit inevitably reflects a value judgment concerning the point at which the interests in favor of protecting 12See Wilson v. Garcia, 471 U. S. 261, 282 (1985) (O’Connor, J., dissenting) (“[A] legislature’s selection of differing limitations periods for a claim sounding in defamation and one based on a written contract is grounded in its evaluation of the characteristics of those claims relevant to the realistic life expectancy of the evidence and the adversary’s reasonable expectations of repose”); 777 F. 2d, at 138 (statement of Judge Garth sur petition for rehearing) (“Most states have concluded that economically grounded causes of action will more frequently arise from patterned and well-documented courses of conduct than will claims for personal injury . . . . There is no reason we should not respect these policy choices, grounded as they are in real and substantial differences between and among causes of action, in applying civil rights statutes which reflect the same differences”); Meyers v. Pennypack Woods Horne Ownership Assn., 559 F. 2d 894, 903, n. 26 (CA3 1977) (quoting Dudley v. Textron, Inc., Burkart-Randall Division, 386 F. Supp. 602, 606 (ED Pa. 1975) (Section 1981 and 1982 cases normally involve “‘patterned-type behavior, frequently involving documentary proof’”; accordingly, “‘[t]he passage of time is less likely to impede the proof of facts’”)); Dupree v. Hertz Corp., 419 F. Supp. 764, 767 (ED Pa. 1976) (“[T]he passage of time is not as likely to interfere with the proof of an employment discrimination case as it would affect the memories of witnesses in a personal injury action”); Dudley v. Textron, Inc., supra, at 606 (“[Section] 1983 actions have typically involved tort claims arising from personal injury, in many cases involving physical conduct of an irregular or sudden nature. By contrast, claims made pursuant to § 1981 usually arise out of employment contract relationships which consist of more patterned-type behavior, frequently involving documentary proof in the form of employment records. Accordingly, the passage of time is less likely to impede the proof of facts in a § 1981, than in a § 1983, case and a longer statute of limitations under § 1981 is, therefore, more appropriate”). 680 OCTOBER TERM, 1986 Opinion of Powell, J. 482 U. S. valid claims are outweighed by the interests in prohibiting the prosecution of stale ones.” Johnson n. Railway Express Agency, Inc., 421 U. S. 454, 463-464 (1975). Today we have required States to apply in contract cases a “value judgment” reached with regard to torts. Inevitably, the statute of limitations henceforth used in § 1981 cases will be wrong most of the time.13 Ill It may well be that “it is the fate of contract to be swallowed up by tort (or for both of them to be swallowed up in a generalized theory of civil obligation),” G. Gilmore, The Death of Contract 94 (1974), but it has not happened yet. The general obligation to treat all persons with equal dignity undeniably prohibits discrimination based on race. Yet that obligation is still imposed in a legal system that classifies obligations, a system that distinguishes between obligations based on contract and those based on the reasonable person’s duty of care. Section 1981 actions were primarily intended to, and most often do, vindicate claims which related to contractual rights, and we should apply a state statute of limitations governing contractual relations to them. I respectfully dissent. Justice Powell, with whom Justice Scalia joins, and with whom Justice O’Connor joins as to Parts I through IV, concurring in part and dissenting in part. I concur in the Court’s holding that the state statute of limitations for personal injury actions should apply to claims arising under 42 U. S. C. § 1981. I also agree that the Court’s ruling on the statute of limitations question should 13 Pennsylvania formerly applied a 6-year statute of limitations to contract actions. Pa. Stat. Ann., Tit. 42, § 5527 (Purdon 1981). This statute has generally been applied to § 1981 actions arising in Pennsylvania “where the gist of the cause of action is economic rather than bodily injury caused by interference with the employment rights of black workers,” 777 F. 2d, at 131 (Garth, J., dissenting), and I would apply it here. GOODMAN v. LUKENS STEEL CO. 681 656 Opinion of Powell, J. apply to the parties in this case and therefore join Part I of the Court’s opinion. I dissent, however, from Part II of the Court’s opinion, that affirms the judgment against the Unions for violating § 1981 and Title VII of the Civil Rights Act of 1964. The ambiguous findings of the District Court, accepted by the Court of Appeals for the Third Circuit, do not provide adequate support for the Court’s conclusion that the Unions engaged in intentional discrimination against black members. Neither of the courts below specifically found that the Unions were motivated by racial animus, or that they are liable to black members under the alternative Title VII theory of disparate impact. Accordingly, I would remand to permit the District Court to clarify its findings of fact and to make additional findings if necessary. I Close examination of the findings of the District Court is essential to a proper understanding of this case. The plaintiffs, blacks employed by the Lukens Steel Company, sued the United Steelworkers of America and two of its local unions (Unions) for alleged violations of § 1981 and Title VII. The plaintiffs’ allegations were directed primarily at the Unions’ handling of grievances on behalf of black members. The District Court found that “[t]he steady increase in grievance filings each year has not produced a corresponding increase in the capacity of the grievance-processing system to handle complaints.” 580 F. Supp. 1114, 1158 (ED Pa. 1984). Consequently, the court found, the Unions gave priority to “[s]erious grievances”—that is, “those involving more than a four-day suspension, and those involving discharges.” Ibid. In an effort to reduce the backlog of grievances, the Unions disposed of many less serious grievances by simply withdrawing them and reserving the right to seek relief in a later grievance proceeding. The District Court found “no hard evidence to support an inference that these inadequacies disadvantage blacks to a greater extent than whites.” Ibid. 682 OCTOBER TERM, 1986 Opinion of Powell, J. 482 U. S. The incomplete evidence in the record suggests that the percentage of grievances filed on behalf of black employees was proportional to the number of blacks in the work force. Ibid. Of the relatively few grievances that proceeded all the way to arbitration, the District Court found that the number asserted on behalf of black members was proportional to the number of blacks in the work force. Ibid. Moreover, black members had a slightly higher rate of success in arbitration than white members. Id., at 1158-1159. In sum, the District Court found that “plaintiffs’ generalized evidence concerning perceptions about racial inequities in the handling of grievances does not, without more, establish a prima facie case ... J’1 Id., at 1159. The District Court concluded, however, that the plaintiffs were “on firmer ground” in challenging the Unions’ “repeated failures, during the limitations period, to include racial discrimination as a basis for grievances or other complaints against the company.” Ibid. Beginning in 1965, the Unions’ collective-bargaining agreements with the employer prohibited discrimination on the basis of race against any employee, permanent or probationary. It is undisputed that the Unions “were reluctant to assert racial discrimination as a basis for agrievance.” Ibid. The court found the Unions’ explanation for this reluctance facially reasonable. Ibid. The Unions observed that employees were more likely to obtain relief if a grievance based on racial discrimination was framed as a violation of another provision of the collectivebargaining agreement that did not require proof of racial animus. Moreover, when faced with an allegation of racial dis- :The District Court found that black union members “actively participated” in union meetings and affairs. 580 F. Supp., at 1157. A black member served as chairman of the grievance committee, and other black members served on the committee. Brief for Petitioners in No. 85-2010, p. 7; 2 App. 714-715. The percentage of black shop stewards, the Unions’ primary representatives in the grievance process, frequently exceeded the percentage of black members in the bargaining unit. Brief for Petitioners in No. 85-2010, p. 7; 2 App. 634-640. GOODMAN v. LUKENS STEEL CO. 683 656 Opinion of Powell, J. crimination, “the company tended ‘to get its back up’ and resist [the] charge.” Ibid. The court nevertheless rejected the Unions’ explanation, for two reasons. First, the court found that the Unions “virtually ignored” the “numerous instances of harassment, which were indisputably racial in nature, but which did not otherwise plainly violate a provision of the collective bargaining agreement.” Id., at 1160. Second, the court concluded that “vigorous pursuit of claims of racial discrimination would have focused attention upon racial issues and compelled some change in racial attitudes,” and that the Unions’ “unwillingness to assert racial discrimination claims as such rendered the non-discrimination clause in the collective bargaining agreement a dead letter.” Ibid. The District Court also found that the Unions had adopted a policy of refusing to process any grievances on behalf of probationary employees, despite the fact that the collectivebargaining agreement prohibited employers from discriminating against any employee, permanent or probationary, on the basis of race. The Unions adhered to this policy, the court found, even though they “knew that blacks were being discharged ... at a disproportionately higher rate than whites.” Id., at 1159. Finally, the court found that the Unions failed to object to written tests administered by the employer on the ground that it had a disparate impact on black members, even though they “were certainly chargeable with knowledge that many of the tests . . . were notorious in that regard.” Ibid. The court found, however, that the Unions objected to “tests of all kinds,” on the ground that they gave an unfair advantage to younger employees who had recently completed their formal education. Ibid. The Court of Appeals accepted each of the District Court’s findings of fact and affirmed the judgment against the Unions. 777 F. 2d 113 (CA3 1985). The appellate court concluded that the Unions’ “deliberate choice not to process grievances” violated Title VII “because it discriminated against the victims who were entitled to representation.” 684 OCTOBER TERM, 1986 Opinion of Powell, J. 482 U. S. Id., at 127. The Court of Appeals also concluded that “[t]he district court’s finding of intentional discrimination properly supports the claims under § 1981 as well.” Ibid. II A As the Court recognizes, plaintiffs can recover under § 1981 only for intentional discrimination. Ante, at 665-666, n. 10; General Building Contractors Assn., Inc. v. Pennsylvania, 458 U. S. 375, 382-391 (1982). The Court also recognizes that a valid claim under Title VII must be grounded on proof of disparate treatment or disparate impact. Ante, at 664. A disparate-treatment claim, like a § 1981 claim, requires proof of a discriminatory purpose. Teamsters n. United States, 431 U. S. 324, 335-336, n. 15 (1977). Of course, “‘[discriminatory purpose’ . . . implies more than intent as volition or intent as awareness of consequences.” Personnel Administrator of Mass. n. Feeney, 442 U. S. 256, 279 (1979) (citation omitted). It implies that the challenged action was taken “at least in part ‘because of,’ not merely ‘in spite of,’ its adverse effects upon an identifiable group.” Ibid, (footnote omitted). The Court concedes that “there was no suggestion below that the Unions held any racial animus against or denigrated blacks generally.” Ante, at 668. It nevertheless concludes that the Unions violated Title VII and § 1981 because they “refuse[d] to file any and all grievances presented by a black person on the ground that the employer looks with disfavor on and resents such grievances,” ante, at 669, and “pursue[d] a policy of rejecting disparatetreatment grievances presented by blacks solely because the claims assert racial bias and would be very troublesome to process,” ibid. In my view, this description of the Union’s conduct, and thus the Court’s legal conclusion, simply does not fit the facts found by the District Court. The Unions offered a nondiscriminatory reason for their practice of withdrawing grievances that did not involve a dis- GOODMAN v. LUKENS STEEL CO. 685 656 Opinion of Powell, J. charge or lengthy suspension. According to the Unions, this policy, that is racially neutral on its face, was motivated by the Unions’ nondiscriminatory interest in using the inadequate grievance system to assist members who faced the most serious economic harm. The District Court made no finding that the Unions’ explanation was a pretext for racial discrimination. The Unions’ policy against pursuing grievances on behalf of probationary employees also permitted the Unions to focus their attention on members with the most to lose. Similarly, the Unions’ stated purpose for processing racial grievances on nonracial grounds—to obtain the swiftest and most complete relief possible for the claimant, see 580 F. Supp., at 1159—was not racially invidious. The Unions opposed the use of tests that had a disparate impact on black members, although not on that ground. Their explanation was that more complete relief could be obtained by challenging the tests on nonracial grounds. 1 App. 237. The District Court made no finding that the Unions’ decision to base their opposition on nonracial grounds was motivated by racial animus.2 Absent a finding that the Unions intended to dis 2 Of course, an inference of discriminatory intent may arise from evidence of objective factors, including the inevitable or foreseeable consequences of the challenged policy or practice. Personnel Administrator of Mass. v. Feeney, 442 U. S. 256, 279, n. 25 (1979); Arlington Heights v. Metropolitan Housing Dev. Corp., 429 U. S. 252, 266 (1977). But when “the impact is essentially an unavoidable consequence of a . . . policy that has in itself always been deemed to be legitimate, . . . the inference simply fails to ripen into proof.” Personnel Administrator of Mass. v. Feeney, supra, at 279, n. 25. The District Court did not expressly rely on any inference of racial animus drawn from the consequences of the Unions’ grievance policies. Indeed, it appears that the District Court imposed liability for intentional discrimination without finding that the Unions acted, or failed to act, with the purpose of harming black members. The District Court’s primary justification for imposing liability was that “mere union passivity in the face of employer-discrimination renders the unions liable under Title VII and, if racial animus is properly inferrable, under § 1981 as well.” 580 F. Supp., at 1160 (citations omitted). It then stated: 686 OCTOBER TERM, 1986 Opinion of Powell, J. 482 U. S. criminate against black members, the conclusion that the Unions are liable under § 1981 or the disparate-treatment theory of Title VII is unjustified. B Although the District Court stated that the plaintiffs raised both disparate-treatment and disparate-impact claims, 580 F. Supp., at 1119, it did not make specific findings nor did it conclude that the plaintiffs are entitled to recover under a disparate-impact theory. Indeed, the limited amount of statistical evidence discussed by the District Court indicates that the Unions’ grievance procedures did not have a disparate impact on black members. See supra, at 682. Moreover, neither the District Court nor the Court of Appeals considered the validity of potential defenses to disparateimpact claims. For example, before the court properly could have held the Unions liable on a disparate-impact theory, the court should have considered whether the Unions’ practices were justified by the doctrine of business—or union—necessity. See Griggs v. Duke Power Co., 401 U. S. 424, 431 (1971). The court also should have considered arguments that some of the challenged practices, such as the Unions’ refusal to pursue grievances of probationary employees, were justifiable as part of a bona fide seniority system.3 See Ford “Moreover, the evidence in this case proves far more than mere passivity on the part of the unions. The distinction to be observed is between a union which, through lethargy or inefficiency simply fails to perceive problems or is inattentive to their possible solution (in which case, at least arguably, the union’s inaction has no connection with race) and a union which, aware of racial discrimination against some of its members, fails to protect their interests.” Ibid. Far from inferring racial animus from the foreseeable consequences of the Unions’ inaction, the District Court merely stated its view that union passivity—whether deliberate or inadvertent—is a basis for liability without regard to the Unions’ purpose or intent. 8 Although these defenses do not appear to have been raised by the Unions in the courts below, this is not surprising in view of the fact that the plaintiffs did not present evidence or legal arguments to support a disparate-impact theory. GOODMAN v. LUKENS STEEL CO. 687 656 Opinion of Powell, J. Motor Co. v. EEOC, 458 U. S. 219, 239-240 (1982). Because this Court is reluctant to consider alternative theories of liability not expressly passed upon by the lower courts, see Fumco Construction Corp. v. Waters, 438 U. S. 567, 580-581 (1978), I would remand to the District Court to permit it to consider whether the Unions are liable under a disparateimpact theory.4 Ill The Court does not reach the question whether a union may be held liable under Title VII for “mere passivity” in the face of discrimination by the employer, because it agrees with the courts below that the record shows more than mere passivity on the part of the Unions. Ante, at 665-666. I disagree with that conclusion, and so must consider whether the judgment can be affirmed on the ground that Title VII imposes an affirmative duty on unions to combat discrimination by the employer. The starting point for analysis of this statutory question is, as always, the language of the statute itself. Kelly v. Robinson, 479 U. S. 36, 43 (1986). Section 703(c), the provision of Title VII governing suits against unions, does not suggest that the union has a duty to take affirmative steps to remedy employer discrimination.5 Section 703(c)(1) makes it unlawful for a union “to exclude or to expel from its membership, or otherwise to discriminate against, any individual 4 An additional consideration supporting a remand is the Court’s determination that a 2-year statute of limitations applies rather than the 6-year statute of limitations applied by the District Court. It is not clear whether the District Court would impose liability on the Unions based solely on their conduct after 1971. The Court of Appeals vacated the District Court’s finding that racial harassment was a classwide problem because it could not determine from the record whether racial harassment after 1971 amounted to more than “a few isolated incidents.” 777 F. 2d 113, 121 (CAS 1985). Moreover, there is evidence in the record that the Unions filed grievances explicitly alleging racial discrimination after 1971. 2 App. 412, 422, 491, 657, 659, 684. 6 Section 703, 42 U. S. C. §2000e-2(c), is set out in full ante, at 658-659, n. 2. 688 OCTOBER TERM, 1986 Opinion of Powell, J. 482 U. S. because of his race, color, religion, sex, or national origin.” 42 U. S. C. § 2000e-2(c)(l). This subsection parallels §703 (a)(1), that applies to employers. See 42 U. S. C. §2000e-2(a)(1). This parallelism, and the reference to union membership, indicate that § 703(c)(1) prohibits direct discrimination by a union against its members; it does not impose upon a union an obligation to remedy discrimination by the employer. Moreover, § 703(c)(3) specifically addresses the union’s interaction with the employer, by outlawing efforts by the union “to cause or attempt to cause an employer to discriminate against an individual in violation of this section.” § 2000e-2(c)(3). If Congress had intended to impose on unions a duty to challenge discrimination by the employer, it hardly could have chosen language more ill suited to its purpose. First, “[t]o say that the union ‘causes’ employer discrimination simply by allowing it is to stretch the meaning of the word beyond its limits.” 1 A. Larson & L. Larson, Employment Discrimination, §44.50, p. 9-40 (1985). Moreover, the language of § 703(c)(3) is taken in haec verba from § 8(b)(2) of the National Labor Relations Act (NLRA), 29 U. S. C. § 158(b)(2). That provision of the NLRA has been held not to impose liability for passive acquiescence in wrongdoing by the employer. Indeed, well before the enactment of Title VII, the Court held that even encouraging or inducing employer discrimination is not sufficient to incur liability under § 8(b)(2). Electrical Workers v. NLRB, 341 U. S. 694, 703 (1951). In the absence of a clear statement of legislative intent, the Court has been reluctant to read Title VII to disrupt the basic policies of the labor laws. See Trans World Airlines, Inc. n. Hardison, 432 U. S. 63, 79 (1977). Unquestionably an affirmative duty to oppose employer discrimination could work such a disruption. A union, unlike an employer, is a democratically controlled institution directed by the will of its constituents, subject to the duty of fair representation. Like other representative entities, unions must balance the GOODMAN v. LUKENS STEEL CO. 689 656 Opinion of O’Connor, J. competing claims of its constituents. A union must make difficult choices among goals such as eliminating racial discrimination in the workplace, removing health and safety hazards, providing better insurance and pension benefits, and increasing wages. The Court has recognized that “[t]he complete satisfaction of all who are represented is hardly to be expected.” Ford Motor Co. v. Huffman, 345 U. S. 330, 338 (1953). For these reasons unions are afforded broad discretion in the handling of grievances. Electrical Workers v. Foust, 442 U. S. 42, 51 (1979); Vaca v. Sipes, 386 U. S. 171, 191-194 (1967). Union members’ suits against their unions may deplete union treasuries, and may induce unions to process frivolous claims and resist fair settlement offers. Electrical Workers v. Foust, supra, at 51-52; Vaca n. Sipes, supra, at 191-193. The employee is not without a remedy, because union members may file Title VII actions directly against their employers. Alexander v. Gardner-Denver Co., 415 U. S. 36 (1974). I therefore would hold that Title VII imposes on unions no affirmative duty to remedy discrimination by the employer. IV I agree that the judgment in No. 85-1626 should be affirmed. For the reasons stated above, I would vacate the judgment in No. 85-2010 and remand the case for further proceedings consistent with this opinion. Justice O’Connor, concurring in the judgment in No. 85-1626 and dissenting in No. 85-2010. In light of the Court’s decision to apply a uniform characterization for limitations purposes to actions arising under 42 u. S. C. § 1981, I agree that the most appropriate choice is each State’s limitations period for personal injury suits. But see Wilson v. Garcia, 471 U. S. 261, 280-287 (1985) (O’Connor, J., dissenting). Although I doubt whether the Court’s decision should be given general retroactive effect, I agree that the Court should adhere to its policy of applying the rule 690 OCTOBER TERM, 1986 Opinion of O’Connor, J. 482 U. S. it announces to the parties before the Court. See Stovall v. Denno, 388 U. S. 293, 301 (1967). I therefore concur in the judgment of the Court in No. 85-1626. I join Parts I through IV of Justice Powell’s opinion concurring in part and dissenting in part, as to No. 85-2010. NEW YORK v. BURGER 691 Syllabus NEW YORK v. BURGER CERTIORARI TO THE COURT OF APPEALS OF NEW YORK No. 86-80. Argued February 23, 1987—Decided June 19, 1987 Respondent junkyard owner’s business consists, in part, of dismantling automobiles and selling their parts. Pursuant to a New York statute authorizing warrantless inspections of automobile junkyards, police officers entered his junkyard and asked to see his license and records as to automobiles and vehicle parts in his possession. He replied that he did not have such documents, which are required by the statute. After announcing their intention to conduct an inspection of the junkyard pursuant to the statute, the officers, without objection by respondent, conducted the inspection and discovered stolen vehicles and parts. Respondent, who was charged with possession of stolen property and unregistered operation as a vehicle dismantler, moved in state court to suppress the evidence obtained as a result of the inspection, primarily on the ground that the administrative inspection statute was unconstitutional. The court denied the motion, and the Appellate Division affirmed. The New York Court of Appeals reversed, concluding that the statute violated the Fourth Amendment’s prohibition of unreasonable searches and seizures. Held: 1. A business owner’s expectation of privacy in commercial property is attenuated with respect to commercial property employed in a “closely regulated” industry. Where the owner’s privacy interests are weakened and the government interests in regulating particular businesses are concomitantly heightened, a warrantless inspection of commercial premises, if it meets certain criteria, is reasonable within the meaning of the Fourth Amendment. Pp. 699-703. 2. Searches made pursuant to the New York statute fall within the exception to the warrant requirement for administrative inspections of “closely regulated” businesses. Pp. 703-712. (a) The nature of the statute establishes that the operation of a junkyard, part of which is devoted to vehicle dismantling, is a “closely regulated” business. Although the duration of a’particular regulatory scheme has some relevancy, and New York’s scheme regulating vehicle dismantlers can be said to be of fairly recent vintage, nevertheless, because widespread use of the automobile is relatively new, automobile junkyards and vehicle dismantlers have not been in existence very long and thus do not have an ancient history of government oversight. 692 OCTOBER TERM, 1986 Syllabus 482 U. S. Moreover, the automobile-junkyard business is simply a new branch of an industry—general junkyards and secondhand shops—that has existed, and has been closely regulated in New York, for many years. Pp. 703-707. (b) New York’s regulatory scheme satisfies the criteria necessary to make reasonable the warrantless inspections conducted pursuant to the inspection statute. First, the State has a substantial interest in regulating the vehicle-dismantling and automobile-junkyard industry because motor vehicle theft has increased in the State and because the problem of theft is associated with such industry. Second, regulation of the industry reasonably serves the State’s substantial interest in eradicating automobile theft, and warrantless administrative inspections pursuant to the statute are necessary to further the regulatory scheme. Third, the statute provides a constitutionally adequate substitute for a warrant. It informs a business operator that regular inspections will be made, and also sets forth the scope of the inspection, notifying him as to how to comply with the statute and as to who is authorized to conduct an inspection. Moreover, the “time, place, and scope” of the inspection is limited to impose appropriate restraints upon the inspecting officers’ discretion. Pp. 708-712. 3. The New York inspection statute does not violate the Fourth Amendment on the ground that it was designed simply to give the police an expedient means of enforcing penal sanctions for possession of stolen property. A State can address a major social problem both by way of an administrative scheme—setting forth rules to guide an operator’s conduct of its business and allowing government officials to ensure that such rules are followed—and through penal sanctions. Cf. United States v. Biswell, 406 U. S. 311. New York’s statute was designed to contribute to the regulatory goals of ensuring that vehicle dismantlers are legitimate businesspersons and that stolen vehicles and vehicle parts passing through automobile junkyards can be identified. Nor is the administrative scheme unconstitutional simply because, in the course of enforcing it, an inspecting officer may discover evidence of crimes, besides violations of the scheme itself. Moreover, there is no constitutional significance in the fact that police officers, rather than “administrative” agents, are permitted to conduct the administrative inspection. So long as a regulatory scheme is properly administrative, it is not rendered illegal by the fact that the inspecting officer has the power to arrest individuals for violations other than those created by the scheme itself. Pp. 712-718. 67 N. Y. 2d 338, 493 N. E. 2d 926, reversed and remanded. NEW YORK v. BURGER 693 691 Opinion of the Court Blackmun, J., delivered the opinion of the Court, in which Rehnquist, C. J., and White, Powell, Stevens, and Scalia, JJ., joined. Brennan, J., filed a dissenting opinion, in which Marshall, J., joined, and in all but Part III of which O’Connor, J., joined, post, p. 718. Elizabeth Holtzman argued the cause for petitioner. With her on the briefs were Barbara D. Underwood and Leonard Joblove. Stephen R. Mahler argued the cause for respondent. With him on the brief was Perry S. Reich. * Justice Blackmun delivered the opinion of the Court. This case presents the question whether the warrantless search of an automobile junkyard, conducted pursuant to a statute authorizing such a search, falls within the exception to the warrant requirement for administrative inspections of pervasively regulated industries. The case also presents the question whether an otherwise proper administrative inspection is unconstitutional because the ultimate purpose of the regulatory statute pursuant to which the search is done—the deterrence of criminal behavior—is the same as that of penal laws, with the result that the inspection may disclose violations not only of the regulatory statute but also of the penal statutes. I Respondent Joseph Burger is the owner of a junkyard in Brooklyn, N. Y. His business consists, in part, of the dismantling of automobiles and the selling of their parts. His junkyard is an open lot with no buildings. A high metal fence surrounds it, wherein are located, among other things, vehicles and parts of vehicles. At approximately noon on November 17, 1982, Officer Joseph Vega and four other plainclothes officers, all members of the Auto Crimes Division of the New York City Police Department, entered re * Richard Emery, Gerard E. Lynch, and Alvin J. Bronstein filed a brief for the American Civil Liberties Union et al. as amici curiae urging affirmance. 694 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. spondent’s junkyard to conduct an inspection pursuant to N. Y. Veh. & Traf. Law §415-a5 (McKinney 1986).1 Tr. 6. On any given day, the Division conducts from 5 to 10 inspections of vehicle dismantlers, automobile junkyards, and related businesses.2 Id., at 26. Upon entering the junkyard, the officers asked to see Burger’s license3 and his “police book”—the record of the auto 1 This statute reads in pertinent part: “Records and identification, (a) Any records required by this section shall apply only to vehicles or parts of vehicles for which a certificate of title has been issued by the commissioner [of the Department of Motor Vehicles] or which would be eligible to have such a certificate of title issued. Every person required to be registered pursuant to this section shall maintain a record of all motor vehicles, trailers, and major component parts thereof, coming into his possession together with a record of the disposition of any such motor vehicle, trailer or part thereof and shall maintain proof of ownership for any motor vehicle, trailer or major component part thereof while in his possession. Such records shall be maintained in a manner and form prescribed by the commissioner. The commissioner may, by regulation, exempt vehicles or major component parts of vehicles from all or a portion of the record keeping requirements based upon the age of the vehicle if he deems that such record keeping requirements would serve no substantial value. Upon request of an agent of the commissioner or of any police officer and during his regular and usual business hours, a vehicle dismantler shall produce such records and permit said agent or police officer to examine them and any vehicles or parts of vehicles which are subject to the record keeping requirements of this section and which are on the premises. . . . The failure to produce such records or to permit such inspection on the part of any person required to be registered pursuant to this section as required by this paragraph shall be a class A misdemeanor.” 2 It was unclear from the record why, on that particular day, Burger’s junkyard was selected for inspection. Tr. 23-24. The junkyards designated for inspection apparently were selected from a list of such businesses compiled by New York City police detectives. Id., at 24. 3 An individual operating a vehicle-dismantling business in New York is required to have a license: “Definition and registration of vehicle dismantlers. A vehicle dismantler is any person who is engaged in the business of acquiring motor vehicles or trailers for the purpose of dismantling the same for parts or reselling such vehicles as scrap. No person shall engage in the business of or NEW YORK v. BURGER 695 691 Opinion of the Court mobiles and vehicle parts in his possession. Burger replied that he had neither a license nor a police book.4 The officers then announced their intention to conduct a § 415-a5 inspection. Burger did not object. Tr. 6, 47. In accordance with their practice, the officers copied down the Vehicle Identification Numbers (VINs) of several vehicles and parts of vehicles that were in the junkyard. Id., at 7, 20, 44, 46. After checking these numbers against a police computer, the officers determined that respondent was in possession of stolen vehicles and parts.5 Accordingly, Burger was arrested and charged with five counts of possession of stolen property6 operate as a vehicle dismantler unless there shall have been issued to him a registration in accordance with the provisions of this section. A violation of this subdivision shall be a class E felony.” N. Y. Veh. & Traf. Law § 415-al (McKinney 1986). 4 There appears to have been some initial confusion among the inspect- ing officers as to whether Burger had not compiled a police book or whether, at the moment of the inspection, it simply was not in his possession. See Tr. 6, 30, 46-47, 59-60. 6 The officers also determined that Burger possessed a wheelchair and a handicapped person’s walker that had been located in a stolen vehicle. See id., at 8-11, 13, 34-36. 6 Respondent was charged with two counts of criminal possession of stolen property in the second degree in violation of a New York statute that, at that time, read: “A person is guilty of criminal possession of stolen property in the second degree when he knowingly possesses stolen property, with intent to benefit himself or a person other than an owner thereof or to impede the recovery by an owner thereof, and when: “1. The value of the property exceeds two hundred fifty dollars; or “3. He is a pawnbroker or is in the business of buying, selling or otherwise dealing in property .... “Criminal possession of stolen property in the second degree is a class E felony.” N. Y. Penal Law § 165.45 (McKinney 1975). Burger also was charged with three counts of criminal possession of stolen property in the third degree pursuant to the following provision of a New York statute: 696 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. and one count of unregistered operation as a vehicle disman-tler, in violation of § 415-al. In the Kings County Supreme Court, Burger moved to suppress the evidence obtained as a result of the inspection, primarily on the ground that § 415-a5 was unconstitutional. After a hearing, the court denied the motion. It reasoned that the junkyard business was a “pervasively regulated” industry in which warrantless administrative inspections were appropriate, that the statute was properly limited in “time, place and scope,” and that, once the officers had reasonable cause to believe that certain vehicles and parts were stolen, they could arrest Burger and seize the property without a warrant. App. to Pet. for Cert. 18a-19a. When respondent moved for reconsideration in light of a recent decision of the Appellate Division, People v. Pace, 101 App. Div. 2d 336, 475 N. Y. S. 2d 443 (1984), aff’d, 65 N. Y. 2d 684, 481 N. E. 2d 250 (1985),7 the court granted reargument. Upon re- “A person is guilty of criminal possession of stolen property in the third degree when he knowingly possesses stolen property, with intent to benefit himself or a person other than an owner thereof or to impede the recovery by an owner thereof. “Criminal possession of stolen property in the third degree is a class A misdemeanor.” N. Y. Penal Law § 165.40 (McKinney 1975). 7 In People v. Pace, the Appellate Division was faced with a situation in which officers had conducted a warrantless search of an automobile salvage yard immediately after having their suspicions aroused about criminal activity there. The court did not find the exception for warrantless administrative inspections applicable in that situation, 101 App. Div. 2d, at 340, 475 N. Y. S. 2d, at 446, but made the following footnote remark: “Subdivision 5 of section 415-a of the Vehicle and Traffic Law, the statute under which the police officers said they were acting, has no application. While this section requires dismantlers to keep a police book, the book was missing when the officers entered and it would thus have been impossible for the officers to exercise the alleged implied authority to compare the book entries to the contents of the yard.” Id., at 339, n. 1, 475 N. Y. S. 2d, at 445, n. 1. Respondent construed this footnote to mean that police officers had to obtain a search warrant if a vehicle dismantler did not produce a police book NEW YORK v. BURGER 697 691 Opinion of the Court consideration, the court distinguished the situation in Pace from that in the instant case. It observed that the Appellate Division in Pace did not apply § 415-a5 to the search in question, 125 Mise. 2d 709, 711, 479 N. Y. S. 2d 936, 938 (1984), and that, in any event, the police officers in that case were not conducting an administrative inspection, but were acting on the basis of recently discovered evidence that criminal activity was taking place at the automobile salvage yard. Id., at 712-714, 479 N. Y. S. 2d, at 939-940. The court therefore reaffirmed its earlier determination in the instant case that § 415-a5 was constitutional.8 For the same reasons, the Appellate Division affirmed. 112 App. Div. 2d 1046, 493 N. Y. S. 2d 34 (1985). The New York Court of Appeals, however, reversed. 67 N. Y. 2d 338, 493 N. E. 2d 926 (1986). In its view, § 415-a5 violated the Fourth Amendment’s prohibition of unreasonable searches and seizures.9 According to the Court of Ap- and thus they could not conduct a warrantless inspection in the absence of this book. See 125 Mise. 2d 709, 711, 479 N. Y. S. 2d 936, 938 (Sup. 1984). 8 In addition, the court determined that the search was proper under New York City Charter and Admin. Code § 436 (Supp. 1985). 125 Mise. 2d, at 712-715, 479 N. Y. S. 2d, at 939-940. That section reads: “The commissioner [of the Police Department] shall possess powers of general supervision and inspection over all licensed and unlicensed pawnbrokers, vendors, junkshop keepers, junk boatmen, cartmen, dealers in second-hand merchandise and auctioneers within the city; and in connection with the performance of any police duties he shall have power to examine such persons, their clerks and employees and their books, business premises, and any articles of merchandise in their possession. A refusal or neglect' to comply in any respect with the provisions of this section on the part of any pawnbroker, vendor, junkshop keeper, junk boatman, cart-man, dealer in second-hand merchandise or auctioneer, or any clerk or employee of any thereof shall be triable by a judge of the criminal court and punishable by not more than thirty days’ imprisonment, or by a fine of not more than fifty dollars, or both.” 9 The Court of Appeals found that the question of the constitutionality of the statute and charter was squarely presented by this case, as it had not been in People v. Pace, because there was no dispute that the inspection was made pursuant to those provisions. 67 N. Y. 2d, at 342-343, 493 N. E. 2d, at 928. 698 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. peals, “[t]he fundamental defect [of § 415-a5] ... is that [it] authorize [s] searches undertaken solely to uncover evidence of criminality and not to enforce a comprehensive regulatory scheme. The asserted ‘administrative schem[e]’ here [is], in reality, designed simply to give the police an expedient means of enforcing penal sanctions for possession of stolen property.” Id., at 344, 493 N. E. 2d, at 929. In contrast to the statutes authorizing warrantless inspections whose constitutionality this Court has upheld, §415-a5, it was said, “do[es] little more than authorize general searches, including those conducted by the police, of certain commercial premises.” Ibid. To be sure, with its license and recordkeeping requirements, and with its authorization for inspections of records, §415-a appears to be administrative in character. “It fails to satisfy the constitutional requirements for a valid, comprehensive regulatory scheme, however, inasmuch as it permits searches, such as conducted here, of vehicles and vehicle parts notwithstanding the absence of any records against which the findings of such a search could be compared.” Id., at 344-345, 493 N. E. 2d, at 929-930. Accordingly, the only purpose of such searches is to determine whether a junkyard owner is storing stolen property on business premises.10 Because of the important state interest in administrative schemes designed to regulate the vehicle-dismantling or automobile-junkyard industry,11 we granted certiorari. 479 U. S. 812 (1986). 10 For similar reasons, the Court of Appeals concluded that Charter §436 also violated the Fourth Amendment’s prohibition on unreasonable searches and seizures. 67 N. Y. 2d, at 344-345, 493 N. E. 2d, at 929-930. “Numerous States have provisions for the warrantless inspections of vehicle dismantlers and automobile junkyards. See, e. g., Ala. Code §40-12-419 (1985); Ariz. Rev. Stat. Ann. §28-1307C (Supp. 1986); Ark. Stat. Ann. § 75-1803 (1979); Cal. Veh. Code Ann. §§ 2805(a) and (c) (West Supp. 1987); Conn. Gen. Stat. § 14-67m(a) (Supp. 1987); Del. Code Ann., Tit. 21, § 6717(a) (1985); Fla. Stat. §812.055 (Supp. 1987); Ga. Code Ann. §43-48-16 (1984); Ill. Rev. Stat., ch. 9572, 115-403 (Supp. 1986); Ind. NEW YORK v. BURGER 699 691 Opinion of the Court II A The Court long has recognized that the Fourth Amendment’s prohibition on unreasonable searches and seizures is applicable to commercial premises, as well as to private homes. See v. City of Seattle, 387 U. S. 541, 543, 546 (1967). An owner or operator of a business thus has an expectation of privacy in commercial property, which society is prepared to consider to be reasonable, see Katz v. United States, 389 U. S. 347, 361 (1967) (Harlan, J., concurring). This expecta- Code §§ 9-l-3.6-10(a) and (d) and 9-1-3.6-12 (1979 and Supp. 1986); Iowa Code §§ 321.90(3)(b) and 321.95 (1985); Kan. Stat. Ann. §8-2408(c) (1982); Ky. Rev. Stat. § 177.935(7) (1985); La. Rev. Stat. Ann. § 32:757 (West Supp. 1987); Me. Rev. Stat. Ann., Tit. 29, §2459 (Supp. 1986); Md. Transp. Code Ann. § 15-105 (Supp. 1986); Mich. Comp. Laws § 257.251 (Supp. 1987); Miss. Code Ann. §27-19-313 (1972); Mo. Rev. Stat. §301.225 (Supp. 1986); Mont. Code Ann. §§ 75-10-503 and 75-10-513 (1985); Nev. Rev. Stat. §482.3263 (1986); N. H. Rev. Stat. Ann. §261:132 (1982); N. J. Stat. Ann. §39.10B-2c (West Supp. 1987); N. M. Stat. Ann. § 66-2-12(A)(4) (1984); Okla. Stat., Tit. 47, §591.6 (Supp. 1987); Ore. Rev. Stat. §810.480 (1985); R. I. Gen. Laws §42-14.2-15 (Supp. 1986); S. C. Code § 56-5-5670(b) (1976); S. D. Codified Laws §§ 32-6B-38 to 32-6B-40 (Supp. 1987); Tenn. Code Ann. §55-14-106 (1980); Tex. Rev. Civ. Stat. Ann., Art. 6687-2(e) (Vernon Supp. 1987); Utah Code Ann. §§41-3-23(2) and (4) (Supp. 1987); Vt. Stat. Ann., Tit. 23, §466 (1978); Va. Code §46.1-550.12 (Supp. 1986); Wash. Rev. Code §§ 46.80.080(5) and 46.80.150 (1970); W. Va. Code § 17A-6-25 (1986); Wis. Stat. § 218.22(4)(c) (1982); Wyo. Stat. § 31-13-112(e)(iii) (1987). Courts have upheld such statutes against federal constitutional attack. See, e. g., Bionic Auto Parts & Sales, Inc. v. Fahner, 721 F. 2d 1072,1081 (CA71983); People v. Easley, 90 Cal. App. 3d 440, 445, 153 Cal. Rptr. 396, 399, cert, denied, 444 U. S. 899 (1979); Moore v. State, 442 So. 2d 215, 216 (Fla. 1983); People v. Barnes, 146 Mich. App. 37, 42, 379 N. W. 2d 464, 466 (1985); State v. Zinmeister, 27 Ohio App. 3d 313, 318, 501 N. E. 2d 59, 65 (1985); see also State v. Tindell, 272 Ind. 479, 483, -399 N. E. 2d 746, 748 (1980); Shirley v. Commonwealth, 218 Va. 49, 57-58, 235 S. E. 2d 432, 436-437 (1977). But see People v. Krull, 107 Ill. 2d 107, 116-117, 481 N. E. 2d 703, 707-708 (1985), rev’d, 480 U. S. 340 (1987); State v. Galio, 92 N. M. 266, 268-269, 587 P. 2d 44, 46-47 (1978). 700 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. tion exists not only with respect to traditional police searches conducted for the gathering of criminal evidence but also with respect to administrative inspections designed to enforce regulatory statutes. See Marshall n. Barlow's, Inc., 436 U. S. 307, 312-313 (1978). An expectation of privacy in commercial premises, however, is different from, and indeed less than, a similar expectation in an individual’s home. See Donovan n. Dewey, 452 U. S. 594, 598-599 (1981). This expectation is particularly attenuated in commercial property employed in “closely regulated” industries. The Court observed in Marshall v. Barlow's, Inc.: “Certain industries have such a history of government oversight that no reasonable expectation of privacy, see Katz n. United States, 389 U. S. 347, 351-352 (1967), could exist for a proprietor over the stock of such an enterprise.” 436 U. S., at 313. The Court first examined the “unique” problem of inspections of “closely regulated” businesses in two enterprises that had “a long tradition of close government supervision.” Ibid. In Colonnade Corp. v. United States, 397 U. S. 72 (1970), it considered a warrantless search of a catering business pursuant to several federal revenue statutes authorizing the inspection of the premises of liquor dealers. Although the Court disapproved the search because the statute provided that a sanction be imposed when entry was refused, and because it did not authorize entry without a warrant as an alternative in this situation, it recognized that “the liquor industry [was] long subject to close supervision and inspection.” Id., at 77. We returned to this issue in United States v. Biswell, 406 U. S. 311 (1972), which involved a warrantless inspection of the premises of a pawnshop operator, who was federally licensed to sell sporting weapons pursuant to the Gun Control Act of 1968, 18 U. S. C. §921 et seq. While noting that “[f]ederal regulation of the interstate traffic in firearms is not as deeply rooted in history as is governmental control of the liquor industry,” 406 U. S., at 315, we nonetheless concluded that the warrantless inspec NEW YORK v. BURGER 701 691 Opinion of the Court tions authorized by the Gun Control Act would “pose only limited threats to the dealer’s justifiable expectations of privacy.” Id., at 316. We observed: “When a dealer chooses to engage in this pervasively regulated business and to accept a federal license, he does so with the knowledge that his business records, firearms, and ammunition will be subject to effective inspection.” Ibid. The “Colonnade-Biswell” doctrine, stating the reduced expectation of privacy by an owner of commercial premises in a “closely regulated” industry, has received renewed emphasis in more recent decisions. In Marshall v. Barlow’s, Inc., we noted its continued vitality but declined to find that warrantless inspections, made pursuant to the Occupational Safety and Health Act of 1970, 84 Stat. 1598, 29 U. S. C. § 657(a), of all businesses engaged in interstate commerce fell within the narrow focus of this doctrine. 436 U. S., at 313-314. However, we found warrantless inspections made pursuant to the Federal Mine Safety and Health Act of 1977, 91 Stat. 1290, 30 U. S. C. §801 et seq., proper because they were of a “closely regulated” industry. Donovan v. Dewey, supra. Indeed, in Donovan v. Dewey, we declined to limit our consideration to the length of time during which the business in question—stone quarries—had been subject to federal regulation. 452 U. S., at 605-606. We pointed out that the doctrine is essentially defined by “the pervasiveness and regularity of the federal regulation” and the effect of such regulation upon an owner’s expectation of privacy. See id., at 600, 606. We observed, however, that “the duration of a particular regulatory scheme” would remain an “important factor” in deciding whether a warrantless inspection pursuant to the scheme is permissible. Id., at 606.12 12 We explained in Donovan n. Dewey: “If the length of regulation were the only criterion, absurd results would occur. Under appellees’ view, new or emerging industries, including ones such as the nuclear power industry that pose enormous potential safety and health problems, 702 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. B Because the owner or operator of commercial premises in a “closely regulated” industry has a reduced expectation of privacy, the warrant and probable-cause requirements, which fulfill the traditional Fourth Amendment standard of reasonableness for a government search, see O’Connor v. Ortega, 480 U. S. 709, 741 (1987) (dissenting opinion), have lessened application in this context. Rather, we conclude that, as in other situations of “special need,” see New Jersey v. T. L. 0., 469 U. S. 325, 353 (1985) (opinion concurring in judgment), where the privacy interests of the owner are weakened and the government interests in regulating particular businesses are concomitantly heightened, a warrantless inspection of commercial premises may well be reasonable within the meaning of the Fourth Amendment. This warrantless inspection, however, even in the context of a pervasively regulated business, will be deemed to be reasonable only so long as three criteria are met. First, there must be a “substantial” government interest that informs the regulatory scheme pursuant to which the inspection is made. See Donovan v. Dewey, 452 U. S., at 602 (“substantial federal interest in improving the health and safety conditions in the Nation’s underground and surface mines”); United States v. Biswell, 406 U. S., at 315 (regulation of firearms is “of central importance to federal efforts to prevent violent crime and to assist the States in regulating the firearms traffic within their borders”); Colonnade Corp. v. United States, 397 U. S., at 75 (federal interest “in protecting the revenue against various types of fraud”). Second, the warrantless inspections must be “necessary to further [the] regulatory scheme.” Donovan v. Dewey, 452 U. S., at 600. For example, in Dewey we recognized that forcing mine inspectors to obtain a warrant before every in could never be subject to warrantless searches even under the most carefully structured inspection program simply because of the recent vintage of regulation.” 452 U. S., at 606. NEW YORK v. BURGER 703 691 Opinion of the Court spection might alert mine owners or operators to the impending inspection, thereby frustrating the purposes of the Mine Safety and Health Act—to detect and thus to deter safety and health violations. Id., at 603. Finally, “the statute’s inspection program, in terms of the certainty and regularity of its application, [must] provid[e] a constitutionally adequate substitute for a warrant.” Ibid. In other words, the regulatory statute must perform the two basic functions of a warrant: it must advise the owner of the commercial premises that the search is being made pursuant to the law and has a properly defined scope, and it must limit the discretion of the inspecting officers. See Marshall v. Barlow's, Inc., 436 U. S., at 323; see also id., at 332 (Stevens, J., dissenting). To perform this first function, the statute must be “sufficiently comprehensive and defined that the owner of commercial property cannot help but be aware that his property will be subject to periodic inspections undertaken for specific purposes.” Donovan v. Dewey, 452 U. S., at 600. In addition, in defining how a statute limits the discretion of the inspectors, we have observed that it must be “carefully limited in time, place, and scope.” United States v. Biswell, 406 U. S., at 315. Ill A Searches made pursuant to § 415-a5, in our view, clearly fall within this established exception to the warrant requirement for administrative inspections in “closely regulated” businesses.13 First, the nature of the regulatory statute reveals that the operation of a junkyard, part of which is devoted to 13 Because we find the inspection at issue here constitutional under § 415-a5, we have no reason to reach the question of the constitutionality of §436 of the New York City Charter. Moreover, because the Court of Appeals addressed only the general question concerning the constitutionality of the administrative inspection, not the specific question whether the search and seizure of the wheelchair and walker were within the scope of the inspection, we do not reach here this latter issue. 704 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. vehicle dismantling, is a “closely regulated” business in the State of New York.14 The provisions regulating the activity of vehicle dismantling are extensive. An operator cannot engage in this industry without first obtaining a license, which means that he must meet the registration requirements and must pay a fee.15 Under § 415-a5(a), the operator must maintain a police book recording the acquisition and disposition of motor vehicles and vehicle parts, and make such records and inventory available for inspection by the police or any agent of the Department of Motor Vehicles. The operator also must display his registration number prominently at his place of business, on business documentation, and on vehicles and parts that pass through his business. § 415-a5(b). Moreover, the person engaged in this activity is subject to criminal penalties, as well as to loss of license or civil fines, 14 The New York Court of Appeals did not imply that automobile junk- yards were not a “closely regulated” business in that State. Rather, it found fault with one aspect of the administrative statutes regulating these junkyards. 67 N. Y. 2d, at 344-345, 493 N. E. 2d, at 929-930. In his brief in opposition to the petition for certiorari, respondent appears to con- cede that this industry in New York is “closely regulated” by his statement that the New York Legislature could enact a “ ‘comprehensive regulatory scheme’ ” directed at the industry. Brief in Opposition 3. 18 Under § 415-al, “[n]o person shall engage in the business of or operate as a vehicle dismantler unless there shall have been issued to him a registration in accordance with the provisions of this section.” In making an application for a registration, the operator must provide “a listing of all felony convictions and all other convictions relating to the illegal sale or possession of a motor vehicle or motor vehicle parts, and a listing of all arrests for any such violations by the applicant and any other person required to be named in such application.” § 415-a2. Section 415-a3 requires that the operator pay a registration fee, and § 415-a4 stipulates that “no registration shall be issued or renewed unless the applicant has a permanent place of business at which the activity requiring registration is performed which conforms to section one hundred thirty-six of the general municipal law as such section applies and to all local laws or ordinances and the applicant and all persons having a financial interest in the business have been determined by the commissioner to be fit persons to engage in such business.” NEW YORK v. BURGER 705 691 Opinion of the Court for failure to comply with these provisions. See §§ 415-al, 5, and 6.16 That other States besides New York have imposed similarly extensive regulations on automobile junkyards further supports the “closely regulated” status of this industry. See n. 11, supra. In determining whether vehicle dismantlers constitute a “closely regulated” industry, the “duration of [this] particular regulatory scheme,” Donovan v. Dewey, 452 U. S., at 606, has some relevancy. Section 415-a could be said to be of fairly recent vintage, see 1973 N. Y. Laws, ch. 225, § 1 (McKinney), and the inspection provision of § 415-a5 was added only in 1979, see 1979 N. Y. Laws, ch. 691, §2 (McKinney). But because the automobile is a relatively new phenomenon in our society and because its widespread use is even newer, automobile junkyards and vehicle dismantlers have not been in existence very long and thus do not have an ancient history of government oversight. Indeed, the indus 16 The broad extent of the regulation of the vehicle-dismantling industry further is shown by the fact that § 415-a regulates the activities not only of vehicle dismantlers but also of those in similar businesses, such as salvage pool operators, § 415-al-a, mobile car crushers, § 415-al-b, itinerant vehicle collectors, § 415-al-c, vehicle rebuilders, § 415-a8, scrap processors, § 415-a9, and scrap collectors and repair shops, § 415-alO. Moreover, the Commissioner of the Department of Motor Vehicles has promulgated regulations dealing specifically with this industry: e. g., N. Y. Comp. Codes, Rules & Regs., Tit. 15, § 81.2 (1986) (registration); § 81.8 (procedures upon acquisition of junk and salvage vehicles); §81.10 (vehicle identification numbers); § 81.12 (records). Amici argue that § 415-a does not create a truly administrative scheme, because its provisions are not sufficiently voluminous. See Brief for American Civil Liberties Union et al. as Amici Curiae 34-36. Although the number of regulations certainly is a factor in the determination whether a particular business is “closely regulated,” the sheer quantity of pages of statutory material is not dispositive of this question. Rather, the proper focus is on whether the “regulatory presence is sufficiently comprehensive and defined that the owner of commercial property cannot help but be aware that his property will be subject to periodic inspections undertaken for specific purposes.” Donovan v. Dewey, 452 U. S., at 600. Section 415-a plainly satisfies this criterion. 706 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. try did not attract government attention until the 1950’s, when all used automobiles were no longer easily reabsorbed into the steel industry and attention then focused on the environmental and aesthetic problems associated with abandoned vehicles. See Landscape 1970: National Conference on the Abandoned Automobile 11; see also Report to the President from the Panel on Automobile Junkyards, White House Conference on Natural Beauty 1 (1965) (statement of Charles M. Haar, Chairman: “There are junkyards and abandoned cars in the streets and along the countryside that are making America ugly, not beautiful”). The automobile-junkyard business, however, is simply a new branch of an industry that has existed, and has been closely regulated, for many years. The automobile junkyard is closely akin to the secondhand shop or the general junkyard. Both share the purpose of recycling salvageable articles and components of items no longer usable in their original form. As such, vehicle dismantlers represent a modem, specialized version of a traditional activity.17 In New York, general junkyards and secondhand shops long have been subject to regulation. One New York court has explained: 17 A member of the automobile-junkyard industry described it this way: “Webster says junk is old metal, rags, and rubbish. The word ‘junk’ can also be used as a verb, and as such would mean to discard. I represent an industry that buys vehicles which are no longer suitable for transportation. These vehicles have been wrecked, damaged, or have otherwise become inoperative. They are taken apart by members of our industry. The components that are still usable are made available to garages, body shops, and the general public as used parts for repair of other vehicles. The portion of the vehicle that is not suitable for parts is passed on to a scrap processor who then transforms the hulk, or the remnants, into a product suitable for resmelting purposes.” Junkyards & Solid Waste Disposal in the Highway Environment, Proceedings of National Seminar, June 10-11, 1975, p. 19 (1976) (statement of Donald J. Rouse, National Association of Auto and Truck Recyclers, now known as Automotive Dismantlers and Recyclers of America). NEW YORK v. BURGER 707 691 Opinion of the Court “Vehicle dismantlers are part of the junk industry as well as part of the auto industry. . . . Prior to the enactment of section 415-a of the Vehicle and Traffic Law, auto dismantlers were subject to regulatory provisions governing the licensing and operation of junkyards. These regulations included provisions mandating the keeping of detailed records of purchases and sales, and the making of such records available at reasonable times to designated officials including police officers, by junk dealers . . . and by dealers in secondhand articles .... “These regulatory, record keeping and warrantless inspection provisions for junk shops have been a part of the law of the City of New York and of Brooklyn for at least 140 years.” People v. Tinneny, 99 Mise. 2d 962, 969, 417 N. Y. S. 2d 840, 845 (Sup. 1979). See also N. Y. C. Charter and Admin. Code § B32-113.01 (1977) (“ ‘Junk dealer’. Any person engaged in the business of purchasing or selling junk”); §B32-126.0a (“‘dealer in second-hand articles’ shall mean any person who, in any way or as a principal broker or agent: 1. [d]eals in the purchase or sale of second-hand articles of whatever nature”).18 The history of government regulation of junk-related activities argues strongly in favor of the “closely regulated” status of the automobile junkyard. Accordingly, in light of the regulatory framework governing his business and the history of regulation of related industries, an operator of a junkyard engaging in vehicle dismantling has a reduced expectation of privacy in this “closely regulated” business. 18 In fact, by assuming that Charter § 436 with its use of the terms “junkshop keepers” and “dealers in second-hand merchandise,” see n. 8, supra, could be applied to respondent, the New York Court of Appeals understood that a vehicle dismantler fell within the scope of those terms. See also People v. Cusumano, 108 App. Div. 2d 752, 754, 484 N. Y. S. 2d 909, 912 (1985). 708 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. B The New York regulatory scheme satisfies the three criteria necessary to make reasonable warrantless inspections pursuant to §415-a5. First, the State has a substantial interest in regulating the vehicle-dismantling and automobilejunkyard industry because motor vehicle theft has increased in the State and because the problem of theft is associated with this industry. In this day, automobile theft has become a significant social problem, placing enormous economic and personal burdens upon the citizens of different States. For example, when approving the 1979 amendment to §415-a5, which added the provision for inspections of records and inventory of junkyards, the Governor of the State explained: “Motor vehicle theft in New York State has been rapidly increasing. It has become a multimillion dollar industry which has resulted in an intolerable economic burden on the citizens of New York. In 1976, over 130,000 automobiles were reported stolen in New York, resulting in losses in excess of $225 million. Because of the high rate of motor vehicle theft, the premiums for comprehensive motor vehicle insurance in New York are significantly above the national average. In addition, stolen automobiles are often used in the commission of other crimes and there is a high incidence of accidents resulting in property damage and bodily injury involving stolen automobiles.” Governor’s Message approving L. 1979, chs. 691 and 692, 1979 N. Y. Laws 1826,1826-1827 (McKinney). See also 25 Legislative Newsletter, New York State Automobile Assn., p. 1 (May 10, 1978), reprinted in Governor’s Bill Jacket, L. 1979, ch. 691 (1979 Bill Jacket) (“Auto theft in New York State has become a low-risk, high-profit, multi NEW YORK v. BURGER 709 691 Opinion of the Court million dollar growth industry that is imposing intolerable economic burdens on motorists”).19 Because contemporary automobiles are made from standardized parts, the nationwide extent of vehicle theft and concern about it are understandable. Second, regulation of the vehicle-dismantling industry reasonably serves the State’s substantial interest in eradicating automobile theft. It is well established that the theft problem can be addressed effectively by controlling the receiver of, or market in, stolen property. 2 W. LaFave & A. Scott, Substantive Criminal Law §8.10(a), p. 422 (1986) (“Without [professional receivers of stolen property], theft ceases to be profitable”); 2 Encyclopedia of Crime and Justice 789 (Kadish ed. 1983) (“[The criminal receiver] . . . inspires 95 per cent or more of the theft in America”). Automobile junkyards and vehicle dismantlers provide the major market for stolen vehicles and vehicle parts. See Memorandum from Paul Goldman, Counsel, State Consumer Protection Board, to Richard A. Brown, Counsel to the Governor (June 29, 1979), 1979 Bill Jacket (“It is believed that a major source of stolen vehicles, parts and registration documentation may involve vehicles which pass through the hands of [junk vehicle] dealers”). Thus, the State rationally may believe that it will reduce car theft by regulations that prevent automobile junkyards from becoming markets for stolen vehicles and that help trace the origin and destination of vehicle parts.20 19 A similar concern with stemming the social plague of automobile theft has motivated other States to pass legislation aimed at the vehicle-dismantling industry. See, e. g., Ill. Rev. Stat., ch. 9572, 5-100-1 (Supp. 1985) (legislative finding that “crimes involving the theft of motor vehicles and their parts have risen steadily over the past years, with a resulting loss of millions of dollars to the residents of this State”). 20 See Governor’s Message approving L. 1979, chs. 691 and 692, 1979 N. Y. Laws 1826, 1827 (McKinney) (“By making it difficult to traffic in stolen vehicles and parts, it can be anticipated that automobile theft problems will be decreased and the cost to insurance companies and the public 710 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. Moreover, the warrantless administrative inspections pursuant to § 415-a5 “are necessary to further [the] regulatory scheme.” Donovan n, Dewey, 452 U. S., at 600. In this respect, we see no difference between these inspections and those approved by the Court in United States v. Biswell and Donovan v. Dewey. We explained in Biswell: “[I]f inspection is to be effective and serve as a credible deterrent, unannounced, even frequent, inspections are essential. In this context, the prerequisite of a warrant could easily frustrate inspection; and if the necessary flexibility as to time, scope, and frequency is to be preserved, the protections afforded by a warrant would be negligible.” 406 U. S., at 316. See also Donovan v. Dewey, 452 U. S., at 603. Similarly, in the present case, a warrant requirement would interfere with the statute’s purpose of deterring automobile theft accomplished by identifying vehicles and parts as stolen and shutting down the market in such items. Because stolen cars and parts often pass quickly through an automobile junkyard, “frequent” and “unannounced” inspections are necessary in order to detect them. In sum, surprise is crucial if the regulatory scheme aimed at remedying this major social problem is to function at all. may be reduced”). As the Illinois Legislature found in passing regulations aimed at this industry, “(2) essential to the criminal enterprise of motor vehicle theft operations is the ability of thieves to transfer or sell stolen vehicles or their parts through legitimate commercial channels making them available for sale to the automotive industry; and (3) motor vehicle dealers, used parts dealers, scrap processors, automotive parts recyclers, and rebuilders are engaged in a type of business which often exposes them and their operations to pressures and influences from motor vehicle thieves; and (4) elements of organized crime are constantly attempting to take control of businesses engaged in the sale and repair of motor vehicles so as to further their own criminal interests.” Ill. Rev. Stat., ch. 95V2, 515-100-1 (1985). See also Kan. Stat. Ann. § 8-2402 (1982); Nev. Rev. Stat. § 482.318 (1985). NEW YORK v. BURGER 711 691 Opinion of the Court Third, § 415-a5 provides a “constitutionally adequate substitute for a warrant.” Donovan v. Dewey, 452 U. S., at 603. The statute informs the operator of a vehicle dismantling business that inspections will be made on a regular basis. Id., at 605. Thus, the vehicle dismantler knows that the inspections to which he is subject do not constitute discretionary acts by a government official but are conducted pursuant to statute. See Marshall n. Barlow’s, Inc., 436 U. S., at 332 (dissenting opinion). Section 415-a5 also sets forth the scope of the inspection and, accordingly, places the operator on notice as to how to comply with the statute. In addition, it notifies the operator as to who is authorized to conduct an inspection. Finally, the “time, place, and scope” of the inspection is limited, United States v. Biswell, 406 U. S., at 315, to place appropriate restraints upon the discretion of the inspecting officers. See Donovan v. Dewey, 452 U. S., at 605. The officers are allowed to conduct an inspection only “during [the] regular and usual business hours.” §415-a5.21 The inspections can be made only of vehicle-dismantling and related industries. And the permissible scope of these searches is narrowly defined: the inspectors may examine the records, as well as “any vehicles or parts of vehicles which are subject to 21 Respondent contends that § 415-a5 is unconstitutional because it fails to limit the number of searches that may be conducted of a particular business during any given period. Brief for Respondent 12. While such limitations, or the absence thereof, are a factor in an analysis of the adequacy of a particular statute, they are not determinative of the result so long as the statute, as a whole, places adequate limits upon the discretion of the inspecting officers. Indeed, we have approved statutes authorizing warrantless inspections even when such statutes did not establish a fixed number of inspections for a particular time period. See United States v. Bis-weU, 406 U. S. 311, 312, n. 1 (1972). And we have suggested that, in some situations, inspections must be conducted frequently to achieve the purposes of the statutory scheme. Id., at 316 (“Here, if inspection is to be effective and serve as a credible deterrent, unannounced, even frequent, inspections are essential”) (emphasis added). 712 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. the record keeping requirements of this section and which are on the premises.” Ibid.22 IV A search conducted pursuant to § 415-a5, therefore, clearly falls within the well-established exception to the warrant requirement for administrative inspections of “closely regulated” businesses. The Court of Appeals, nevertheless, struck down the statute as violative of the Fourth Amendment because, in its view, the statute had no truly administrative purpose but was “designed simply to give the police an expedient means of enforcing penal sanctions for possession of stolen property.” 67 N. Y. 2d, at 344, 493 N. E. 2d, at 929. The court rested its conclusion that the administrative goal of the statute was pre textual and that § 415-a5 really “authorize [d] searches undertaken solely to uncover evidence of criminality” particularly on the fact that, even if an operator failed to produce his police book, the inspecting officers could continue their inspection for stolen vehicles and parts. Id., at 344, 345, 493 N. E. 2d, at 929, 930. The court also suggested that the identity of the inspectors—police officers—was significant in revealing the true nature of the statutory scheme. Id., at 344, 493 N. E. 2d, at 929. In arriving at this conclusion, the Court of Appeals failed to recognize that a State can address a major social problem both by way of an administrative scheme and through penal sanctions. Administrative statutes and penal laws may have the same ultimate purpose of remedying the social problem, but they have different subsidiary purposes and prescribe different methods of addressing the problem. An administrative statute establishes how a particular business in a 22 With respect to the adequacy of the statutory procedures, this case is indistinguishable from United States v. Biswell. There, the regulatory provisions of the Gun Control Act permitted warrantless inspections of both records and inventory “at all reasonable times.” Id., at 312, n. 1. The Court held that the statute gave a firearms dealer adequate notice of “the purposes of the inspector [and] the limits of his task.” Id., at 316. NEW YORK v. BURGER 713 691 Opinion of the Court “closely regulated” industry should be operated, setting forth rules to guide an operator’s conduct of the business and allowing government officials to ensure that those rules are followed. Such a regulatory approach contrasts with that of the penal laws, a major emphasis of which is the punishment of individuals for specific acts of behavior. In United States v. Biswell, we recognized this fact that both administrative and penal schemes can serve the same purposes by observing that the ultimate purposes of the Gun Control Act were “to prevent violent crime and to assist the States in regulating the firearms traffic within their borders.” 406 U. S., at 315. It is beyond dispute that certain state penal laws had these same purposes. Yet the regulatory goals of the Gun Control Act were narrower: the Act ensured that “weapons [were] distributed through regular channels and in a traceable manner and [made] possible the prevention of sales to undesirable customers and the detection of the origin of particular firearms.” Id., at 315-316. The provisions of the Act, including those authorizing the warrantless inspections, served these immediate goals and also contributed to achieving the same ultimate purposes that the penal laws were intended to achieve. This case, too, reveals that an administrative scheme may have the same ultimate purpose as penal laws, even if its regulatory goals are narrower. As we have explained above, New York, like many States, faces a serious social problem in automobile theft and has a substantial interest in regulating the vehicle-dismantling industry because of this problem. The New York penal laws address automobile theft by punishing it or the possession of stolen property, including possession by individuals in the business of buying and selling property. See n. 6, supra.23 In accordance with its interest 23 The penal laws often are changed in response to the growth of a particular type of crime. For example, in 1986 New York amended its definition of grand larceny to include the following provision: 714 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. in regulating the automobile-junkyard industry, the State also has devised a regulatory manner of dealing with this problem. Section 415-a, as a whole, serves the regulatory goals of seeking to ensure that vehicle dismantlers are legitimate businesspersons and that stolen vehicles and vehicle parts passing through automobile junkyards can be identified.24 In particular, §415-a5 was designed to contribute to these goals, as explained at the time of its passage: “This bill attempts to provide enforcement not only through means of law enforcement but by making it unprofitable for persons to operate in the stolen car field. “A person is guilty of grand larceny in the fourth degree when he steals property and when: “8. The value of the property exceeds one hundred dollars and the property consists of a motor vehicle, as defined in section one hundred twenty-five of the vehicle and traffic law, other than a motorcycle, as defined in section one hundred twenty-three of such law.” 1986 N. Y. Laws, ch. 515, § 1 (McKinney), codified at N. Y. Penal Law § 155.30 (McKinney Supp. 1987). 24 See, e. g., Memorandum of State Department of Motor Vehicles in support of 1973 N. Y. Laws, ch. 225, 1973 N. Y. Laws 2166, 2167 (McKinney) (purpose of § 415-a “is to provide a system of record keeping so that vehicles can be traced through junk yards and to assure that such junk yards are run by legitimate business men rather than by auto theft rings”); Letter of John D. Caemmerer, Chairman of Senate Committee on Transportation, to Michael Whiteman, Counsel to the Governor (Apr. 12, 1973), reprinted in Governor’s Bill Jacket, L. 1973, ch. 225, p. 15 (1973 Bill Jacket) (“This bill establishes much needed safeguards for an industry which can be readily infiltrated by those wishing to dispose of stolen automobiles or automobile parts”); Letter of Peter M. Pryor, Chairman of New York State Consumer Protection Board, to Michael Whiteman, Counsel to the Governor (Apr. 18, 1973), 1973 Bill Jacket, p. 6 (“Organized crime has used the junk and salvage industry as a convenient staging ground for illicit activities concerning motor vehicles as well as for operations into other areas. The proposed legislation opens the junk and salvage business to the scrutiny of the police and the Department of Motor Vehicles thereby reducing the possibility of utilizing such dealerships as covers for covert businesses”). NEW YORK v. BURGER 715 691 Opinion of the Court “The various businesses which are engaged in this operation have been studied and the control and requirements on the businesses have been written in a manner which would permit the persons engaged in the business to legally operate in a manner conducive to good business practices while making it extremely difficult for a person to profitably transfer a stolen vehicle or stolen part. The general scheme is to identify every person who may legitimately be involved in the operation and to provide a record keeping system which will enable junk vehicles and parts to be traced back to the last legitimately registered or titled owner. Legitimate businessmen engaged in this field have complained with good cause that the lack of comprehensive coverage of the field has put them at a disadvantage with persons who currently are able to operate outside of statute and regulations. They have also legitimately complained that delays inherent in the present statutory regulation and onerous record keeping requirements have made profitable operation difficult. “The provisions of this bill have been drafted after consultation with respected members of the various industries and provides [sic] a more feasible system of controlling traffic in stolen vehicles and parts.” Letter of Stanley M. Gruss, Deputy Commissioner and Counsel, to Richard A. Brown, Counsel to the Governor (June 20, 1979), 1979 Bill Jacket. Accordingly, to state that §415-a5 is “really” designed to gather evidence to enable convictions under the penal laws is to ignore the plain administrative purposes of § 415-a, in general, and § 415-a5, in particular. If the administrative goals of § 415-a5 are recognized, the difficulty the Court of Appeals perceives in allowing inspecting officers to examine vehicles and vehicle parts even in the absence of records evaporates. The regulatory purposes of §415-a5 certainly are served by having the inspecting offi 716 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. cers compare the records of a particular vehicle dismantler with vehicles and vehicle parts in the junkyard. The purposes of maintaining junkyards in the hands of legitimate businesspersons and of tracing vehicles that pass through these businesses, however, also are served by having the officers examine the operator’s inventory even when the operator, for whatever reason, fails to produce the police book.25 Forbidding inspecting officers to examine the inventory in this situation would permit an illegitimate vehicle dismantler to thwart the purposes of the administrative scheme and would have the absurd result of subjecting his counterpart who maintained records to a more extensive search.26 Nor do we think that this administrative scheme is unconstitutional simply because, in the course of enforcing it, an inspecting officer may discover evidence of crimes, besides violations of the scheme itself. In United States v. Biswell, the pawnshop operator was charged not only with a violation of the recordkeeping provision, pursuant to which the inspection was made, but also with other violations detected during the inspection, see 406 U. S., at 313, n. 2, and convicted of a failure to pay an occupational tax for dealing in specific firearms, id., at 312-313. The discovery of evidence of crimes in the course of an otherwise proper administrative inspection does not render that search illegal or the administrative scheme suspect. Cf. United States v. Villamonte-Marquez, 462 U. S. 579, 583-584, and n. 3 (1983).27 25 Failure to produce a record is a misdemeanor, § 415-a5, which can be a ground for suspension of the operator’s license, § 415-a6. This suspension serves to remove illegitimate operators from the industry. 26 Indeed, in United States v. Biswell, we found no constitutional problem with a statute that authorized inspection both of records and inventory, 406 U. S., at 312, n. 1, and with an actual inspection of a dealer’s premises despite the fact that the dealer’s records were not properly maintained, id., at 313, n. 2. 27 The legislative history of § 415-a, in general, and § 415-a5, in particular, reveals that the New York Legislature had proper regulatory purposes for enacting the administrative scheme and was not using it as a NEW YORK v. BURGER 717 691 Opinion of the Court Finally, we fail to see any constitutional significance in the fact that police officers, rather than “administrative” agents, are permitted to conduct the §415-a5 inspection. The significance respondent alleges lies in the role of police officers as enforcers of the penal laws and in the officers’ power to arrest for offenses other than violations of the administrative scheme. It is, however, important to note that state police officers, like those in New York, have numerous duties in addition to those associated with traditional police work. See People v. De Bour, 40 N. Y. 2d 210, 218, 352 N. E. 2d 562, 568 (1976) (“To consider the actions of the police solely in terms of arrest and criminal process is an unnecessary distortion”); see also ABA Standards for Criminal Justice 1-1.1(b) and commentary (2d ed. 1980, Supp. 1982). As a practical matter, many States do not have the resources to assign the enforcement of a particular administrative scheme to a specialized agency. So long as a regulatory scheme is properly administrative, it is not rendered illegal by the fact that the inspecting officer has the power to arrest individuals for violations other than those created by the scheme itself.28 In “pretext” to enable law enforcement authorities to gather evidence of penal law violations. See supra, at 714-715 and n. 24; see also Illinois v. Krull, 480 U. S. 340, 351 (1987) (“[W]e are given no basis for believing that legislators are inclined to subvert their oaths and the Fourth Amendment”). There is, furthermore, no reason to believe that the instant inspection was actually a “pretext” for obtaining evidence of respondent’s violation of the penal laws. It is undisputed that the inspection was made solely pursuant to the administrative scheme. In fact, because the search here was truly a § 415-a5 inspection, the Court of Appeals was able to reach in this case, as it could not in People v. Pace, 65 N. Y. 2d 684, 481 N. E. 2d 250 (1985), the question of the constitutionality of the statute. See 67 N. Y. 2d, at 342-343, 493 N. E. 2d, at 928; see also n. 7, supra. In United States v. Biswell, the search in question was conducted by a city police officer and by a United States Treasury agent, 406 U. S., at 312, the latter being authorized to make arrests for federal crimes. See 27 CFR § 70.28 (1986). The Internal Revenue agents involved in the search in Colonnade Corp. v. United States, 397 U. S. 72, 73 (1970), had similar powers. See 26 U. S. C. § 7608(a). 718 OCTOBER TERM, 1986 Brennan, J., dissenting 482 U. S. sum, we decline to impose upon the States the burden of requiring the enforcement of their regulatory statutes to be carried out by specialized agents. V Accordingly, the judgment of the New York Court of Appeals is reversed, and the case is remanded to that court for further proceedings not inconsistent with this opinion. It is so ordered. Justice Brennan, with whom Justice Marshall joins, and with whom Justice O’Connor joins as to all but Part III, dissenting. Warrantless inspections of pervasively regulated businesses are valid if necessary to further an urgent state interest, and if authorized by a statute that carefully limits their time, place, and scope. I have no objection to this general rule. Today, however, the Court finds pervasive regulation in the barest of administrative schemes. Burger’s vehicledismantling business is not closely regulated (unless most New York City businesses are), and an administrative warrant therefore was required to search it. The Court also perceives careful guidance and control of police discretion in a statute that is patently insufficient to eliminate the need for a warrant. Finally, the Court characterizes as administrative a search for evidence of only criminal wrongdoing. As a result, the Court renders virtually meaningless the general rule that a warrant is required for administrative searches of commercial property.1 I In See v. City of Seattle, 387 U. S. 541, 543 (1967), we held that an administrative search of commercial property gener 1 The Court does not reach the question whether the search was lawful under New York City Charter and Admin. Code §436 (Supp. 1985). I agree with the analysis of the New York Court of Appeals, holding that this provision is plainly unconstitutional. NEW YORK v. BURGER 719 691 Brennan, J., dissenting ally must be supported by a warrant. We make an exception to this rule, and dispense with the warrant requirement, in cases involving “closely regulated” industries, where we believe that the commercial operator’s privacy interest is adequately protected by detailed regulatory schemes authorizing warrantless inspections. See Donovan v. Dewey, 452 U. S. 594, 599 (1981).2 The Court has previously made clear that “the closely regulated industry ... is the exception.” Marshall v. Barlow's, Inc., 436 U. S. 307, 313 (1978). Unfortunately, today’s holding makes it the rule. Initially, the Court excepted from the administrative-warrant requirement only industries which possessed a “‘long tradition of government regulation,’” Donovan v. Dewey, supra, at 605, quoting Marshall n. Dewey, 493 F. Supp. 963, 964 (1980), or which involved an “inherent and immediate danger to health or life.” Note, 48 Ind. L. J. 117, 120-121 (1972).3 The Court today places substantial reliance on the historical justification, and maintains that vehicle dismantling is part of the general junk and secondhand industry, which has a long history of regulation. In Dewey, however, we clarified that, although historical supervision may help to demonstrate that close regulation exists, it is “the pervasiveness and regularity of . . . regulation that ultimately determines whether a warrant is necessary to render 2 In only three industries have we invoked this exception. See Colonnade Catering Corp. x. United States, 397 U. S. 72 (1970) (liquor industry); United States v. Biswell, 406 U. S. 311 (1972) (firearm and ammunitions sales); Donovan v. Dewey, 452 U. S. 594 (1981) (coal mining). 3 Compare Biswell, supra, at 315 (permitting warrantless searches because, although regulation of firearms not as deeply rooted in history as control of the liquor industry, “close scrutiny of this traffic is undeniably of central importance to federal efforts to prevent violent crime”); Dewey, supra, at 602 (permitting warrantless searches in mining industry, which ranks “among the most hazardous in the country”), with Marshall v. Bar-low’s, Inc., 436 U. S. 307 (1978) (requiring warrant when statute authorizes agency to perform health and safety inspections of all businesses engaged in interstate commerce). 720 OCTOBER TERM, 1986 Brennan, J., dissenting 482 U. S. an inspection program reasonable under the Fourth Amendment.” 452 U. S., at 606.4 The provisions governing vehicle dismantling in New York simply are not extensive. A vehicle dismantler must register and pay a fee, display the registration in various circumstances, maintain a police book, and allow inspections. See N. Y. Veh. & Traf. Law §§415-al-6 (McKinney 1986). Of course, the inspections themselves cannot be cited as proof of pervasive regulation justifying elimination of the warrant requirement; that would be obvious bootstrapping. Nor can registration and recordkeeping requirements be characterized as close regulation. New York City, like many States and municipalities, imposes similar, and often more stringent licensing, recordkeeping, and other regulatory requirements on a myriad of trades and businesses.5 4 Moreover, it is “a long tradition of close government supervision” that is relevant to a finding that a business is closely regulated. Id., at 313 (emphasis added). Historically, government regulation of the general junk and secondhand industry was roughly equivalent to the modem regu- lation discussed infra. Neither the general junk industry, nor the vehicle- dismantling industry, is or ever has been pervasively regulated. 8 See licensing and regulatory requirements described in New York City Charter and Admin. Code § B32-1.0 (1977 and Supp. 1985) (exhibitors of public amusement or sport), §B32-22.0 (motion picture exhibitions), §B32-45.0 (billiard and pocket billiard tables), §B32-46.0 (bowling alleys), §B32-54.0 (sidewalk cafes), §B32-58.0 (sidewalk stands), §B32-76.0 (sight-seeing guides), §B32-93.0 (public carts and cartmen), §B32-98.0 (debt collection agencies), §B32-135.0 (pawnbrokers), §B32-138.0 (auctioneers), § B32-167.0 (laundries), § B32-183.0 (locksmiths and keymakers), § B32-206.0 (sales), § B32-251.0 (garages and parking lots), §B32-267.0 (commercial refuse removal), §B32-297.0 (public dance halls, cabarets, and catering establishments), § B32-311.0 (coffeehouses), § B32-324.0 (sight-seeing buses and drivers), § B32-352.0 (home improvement business), § B32-467.0 (television, radio, and audio equipment phonograph service and repairs), §B32-491.0 (general vendors), §B32-532.0 (storage warehouses). New York State has equally comprehensive licensing and permit requirements. See N. Y. Exec. Law § 875 (McKinney Supp. 1987): NEW YORK v. BURGER 721 691 Brennan, J., dissenting Few substantive qualifications are required of an aspiring vehicle dismantler; no regulation governs the condition of the premises, the method of operation, the hours of operation, the equipment utilized, etc. This scheme stands in marked contrast to, e. g., the mine safety regulations relevant in Donovan v. Dewey, supra.6 In sum, if New York City’s administrative scheme renders the vehicle-dismantling business closely regulated, few businesses will escape such a finding. Under these circumstances, the warrant requirement is the exception not the rule, and See has been constructively overruled.7 II Even if vehicle dismantling were a closely regulated industry, I would nonetheless conclude that this search violated the Fourth Amendment. The warrant requirement protects “More than thirty-five state agencies issue rules and permits affecting businesses, organizations and individuals. Permits number in the hundreds in statute with still more in rules and regulations. Those who are regulated move in a maze of rules, permits, licenses, and approvals.” 6 This is not an assertion that some minimal number of pages is a prerequisite to a finding of close regulation, see ante, at 705, n. 16; instead, it is an assertion about the minimal substantive scope of the regulations. The Mine Safety and Health Act at issue in Dewey, supra, mandated inspection of all mines, defined the frequency of inspection (at least twice annually for surface mines, four times annually for underground mines, and irregular 5-, 10-, or 15-day intervals for mines that generate explosive gases), mandated followup inspections where violations had been found, mandated immediate inspection upon notification by a miner or miner’s representative that a dangerous condition exists, required compliance with elaborate standards set forth in the Act and in Title 30 of the Code of Federal Regulations, and required individual notification to mine operators of all standards proposed pursuant to the Act. See Dewey, supra, at 604. 7 The Court further weakens limitations on the closely regulated industries category when it allows the government to proceed without a warrant upon a showing of a substantial state interest. See ante, at 702, 708. The Court should require a warrant for inspections in closely regulated industries unless the inspection scheme furthers an urgent governmental interest. See Dewey, supra, at 599-600, Biswell, supra, at 317. 722 OCTOBER TERM, 1986 Brennan, J., dissenting 482 U. S. the owner of a business from the “unbridled discretion [of] executive and administrative officers,” Marshall, supra, at 323, by ensuring that “reasonable legislative or administrative standards for conducting an . . . inspection are satisfied with respect to a particular [business],” Camara n. Municipal Court, 387 U. S. 523, 538 (1967). In order to serve as the equivalent of a warrant, an administrative statute must create “a predictable and guided [governmental] presence,” Dewey, 452 U. S., at 604. Section 415-a5 does not approach the level of “certainty and regularity of . . . application” necessary to provide “a constitutionally adequate substitute for a warrant.” Id., at 603.8 The statute does not inform the operator of a vehicledismantling business that inspections will be made on a regular basis; in fact, there is no assurance that any inspections at all will occur.9 There is neither an upper nor a lower limit on the number of searches that may be conducted at any given operator’s establishment in any given time period.10 81 also dispute the contention that warrantless searches are necessary to further the regulatory scheme, because of the need for unexpected and/or frequent searches. If surprise is essential (as it usually is in a criminal case), a warrant may be obtained ex parte. See W. LaFave, Search and Seizure § 10.2(e), p. 653 (1987). If the State seeks to conduct frequent inspections, then the statute (or some regulatory authority) should somewhere inform the industry of that fact. 9 See § 415-a5(a) (“Upon request of an agent of the commissioner or of any police officer and during his regular and usual business hours, a vehicle dismantler shall produce such records and permit said agent or police officer to examine them and any vehicles or parts of vehicles which are subject to the record keeping requirements of this section and which are on the premises”). 10 In Dewey, supra, of course, there was no upper limit on the number of mine inspections that could occur each year, but because the statute provided for the inspection of each mine every year, the chance that any particular mine would be singled out for repeated or intensive inspection was diminished. See 452 U. S., at 599 (inspections may not be so “random, infrequent, or unpredictable that the owner, for all practical purposes, has no real expectation that his property will from time to time be inspected by government officials”). NEW YORK v. BURGER 723 691 Brennan, J., dissenting Neither the statute, nor any regulations, nor any regulatory body, provides limits or guidance on the selection of vehicle dismantlers for inspection. In fact, the State could not explain why Burger’s operation was selected for inspection. 67 N. Y. 2d 338, 341, 493 N. E. 2d 926, 927 (1986). This is precisely what was objectionable about the inspection scheme invalidated in Marshall: It failed to “provide any standards to guide inspectors either in their selection of establishments to be searched or in the exercise of their authority to search.” Dewey, supra, at 601. The Court also maintains that this statute effectively limits the scope of the search. We have previously found significant that “the standards with which a [business] operator is required to comply are all specifically set forth,” 452 U. S., at 604, reasoning that a clear and complete definition of potential administrative violations constitutes an implied limitation on the scope of any inspection. Plainly, a statute authorizing a search which can uncover no administrative violations is not sufficiently limited in scope to avoid the warrant requirement. This statute fails to tailor the scope of administrative inspection to the particular concerns posed by the regulated business. I conclude that “the frequency and purpose of the inspections [are left] to the unchecked discretion of Government officers.” Ibid. The conduct of the police in this case underscores this point. The police removed identification numbers from a walker and a wheelchair, neither of which fell within the statutory scope of a permissible administrative search. The Court also finds significant that an operator is on notice as to who is authorized to search the premises; I do not find the statutory limitation—to “any police officer” or “agent of the commissioner”—significant. The sole limitation I see on a police search of the premises of a vehicle dismantler is that it must occur during business hours; otherwise it is open season. The unguided discretion afforded police in this scheme precludes its substitution for a warrant. 724 OCTOBER TERM, 1986 Brennan, J., dissenting 482 U. S. Ill The fundamental defect in §415-a5 is that it authorizes searches intended solely to uncover evidence of criminal acts. The New York Court of Appeals correctly found that § 415-a5 authorized a search of Burger’s business “solely to discover whether defendant was storing stolen property on his premises.” 67 N. Y. 2d, at 345, 493 N. E. 2d, at 930. In the law of administrative searches, one principle emerges with unusual clarity and unanimous acceptance: the government may not use an administrative inspection scheme to search for criminal violations. See Michigan v. Clifford, 464 U. S. 287, 292 (1984) (opinion of Powell, J.) (in fire investigation, the constitutionality of a postfire inspection depends upon “whether the object of the search is to determine the cause of the fire or to gather evidence of criminal activity”); Michigan n. Tyler, 436 U. S. 499, 508 (1978) (“ ‘if the authorities are seeking evidence to be used in a criminal prosecution, the usual standard of probable cause will apply’”) (citations omitted); Donovan n. Dewey, supra, at 598, n. 6 (“[Warrant and probable-cause requirements] pertain when commercial property is searched for contraband or evidence of crime”); Almeida-Sanchez n. United States, 413 U. S. 266, 278 (1973) (Powell, J., concurring) (traditional probable cause not required in border automobile searches because they are “undertaken primarily for administrative rather than prosecutorial purposes”); Camara v. Municipal Court, supra, at 539 (authorization of administrative searches on less than probable cause will not “endange[r] time-honored doctrines applicable to criminal investigations”); See v. City of Seattle, 387 U. S., at 549 (Clark, J., dissenting) (“[N]othing . . . suggests that the inspection was . . . designed as a basis for a criminal prosecution”); Abel v. United States, 362 U. S. 217, 226 (1960) (“The deliberate use by the Government of an administrative warrant for the purpose of gathering evidence in NEW YORK v. BURGER 725 691 Brennan, J., dissenting a criminal case must meet stem resistance by the courts”); id., at 248 (Douglas, J., dissenting) (Government cannot evade the Fourth Amendment “by the simple device of wearing the masks of [administrative] officials while in fact they are preparing a case for criminal prosecution”); Frank v. Maryland, 359 U. S. 360, 365 (1959) (“[E]vidence of criminal action may not ... be seized without a judicially issued search warrant”).11 Here the State has used an administrative scheme as a pretext to search without probable cause for evidence of criminal violations. It thus circumvented the requirements of the Fourth Amendment by altering the label placed on the search. This crucial point is most clearly illustrated by the fact that the police copied the serial numbers from a wheelchair and a handicapped person’s walker that were found on the premises, and determined that these items had been stolen. Obviously, these objects are not vehicles or parts of vehicles, and were in no way relevant to the State’s enforcement of its administrative scheme. The scope of the search alone reveals that it was undertaken solely to uncover evidence of criminal wrongdoing.12 Moreover, it is factually impossible that the search was intended to discover wrongdoing subject to administrative 11 In Camara v. Municipal Court, 387 U. S. 523 (1967), using the presently relevant example of a search for stolen goods, the Court stated that “public interest would hardly justify a sweeping search of an entire city conducted in the hope that these goods might be found. Consequently, a search for these goods ... is ‘reasonable’ only when there is ‘probable cause’ to believe that they will be uncovered in a particular dwelling. ” Id., at 535. 12 Thus, I respectfully disagree with the Court’s conclusion that there is “no reason to believe that the instant inspection was actually a ‘pretext’ for obtaining evidence of respondent’s violation of the penal laws.” Ante, at 717, n. 27. Inspection of the serial numbers on the wheelchair and walker demonstrates that the search went beyond any conceivable administrative purpose. At least the second and third counts of Burger’s indictment for possession of stolen property, which involve the wheelchair and the walker, must be dismissed. 726 OCTOBER TERM, 1986 Brennan, J., dissenting 482 U. S. sanction. Burger stated that he was not registered to dismantle vehicles as required by § 415-al, and that he did not have a police book, as required by §415-a5(a).13 At that point he had violated every requirement of the administrative scheme. There is no administrative provision forbidding possession of stolen automobiles or automobile parts.14 The inspection became a search for evidence of criminal acts when all possible administrative violations had been uncovered.15 The State contends that acceptance of this argument would allow a vehicle dismantler to thwart its administrative scheme simply by failing to register and keep records. This is false. 13 These omissions also subjected him to potential criminal liability; it is a class E felony to fail to register, § 415-al, and a class A misdemeanor to fail to produce a police book, § 415-a5(a). 14 Had Burger been registered as a vehicle dismantler, his registration could have been revoked for illegal possession of stolen vehicles or vehicle parts, and the examination of the vehicles and vehicle parts on his lot would have had an administrative purpose. But he was not registered. 15 In Michigan v. Clifford, 464 U. S. 287 (1984), a case involving an administrative inspection seeking the cause and origin of a fire, the Court was “unanimous in [the] opinion that after investigators have determined the cause of the fire and located the place it originated, a search of other portions of the premises may be conducted only pursuant to a warrant, issued upon probable cause that a crime has been committed.” Id., at 300 (Stevens, J., concurring); see also id., at 294 (“Circumstances that justify a warrantless search for the cause of a fire may not justify a search to gather evidence of criminal activity once that cause has been determined”); id., at 306 (Rehnquist, J., dissenting) (“[A]lthough the remaining parts of the house could not have been searched without the issuance of a warrant issued upon probable cause” the basement was properly searched for the cause and origin of the fire). Thus, “fire officials [could] not . . . rely on [evidence of criminal activity discovered during the course of a valid administrative search] to expand the scope of their administrative search without first making a showing of probable cause to an independent judicial officer.” Id., at 294. Likewise here, the administrative inspection ceased when all administrative purposes had been fulfilled. Further investigation was necessarily a search for evidence of criminal violations, and a warrant based on probable cause was required. NEW YORK v. BURGER 727 691 Brennan, J., dissenting A failure to register or keep required records violates the scheme and results in both administrative sanctions and criminal penalties. See n. 13, supra. Neither is the State’s further criminal investigation thwarted; the police need only obtain a warrant and then proceed to search the premises. If respondent’s failure to register and maintain records amounted to probable cause, then the inspecting police officers, who worked in the Auto Crimes Division of the New York City Police Department, possessed probable cause to obtain a criminal warrant authorizing a search of Burger’s premises.16 Several of the officers might have stayed on the premises to ensure that this unlicensed dismantler did no further business, while the others obtained a warrant. Any inconvenience to the police would be minimal, and in any event, “inconvenience alone has never been thought to be an adequate reason for abrogating the warrant requirement.” Almeida-Sanchez, 413 U. S., at 283 (Powell, J., concurring). The Court properly recognizes that “a State can address a major social problem both by way of an administrative scheme and through penal sanctions.” Ante, at 712. Ad 16 Although the fact that the police conducted the search is not dispositive as to its administrative or criminal nature, it should caution the Court to proceed with care, because “[s]earches by the police are inherently more intrusive than purely administrative inspections. Moreover, unlike administrative agents, the police have general criminal investigative duties which exceed the legitimate scope and purposes of purely administrative inspections.” Commonwealth v. Lipomi, 385 Mass. 370, 378, 432 N. E. 2d 86, 91 (1982). See also W. LaFave, Criminal Search and Seizure § 10.2(f), p. 661 (1987) (“[E]xisting scope limitations would be entitled to somewhat greater weight where by law the inspections may be conducted only by specialized inspectors who could be expected to understand and adhere to the stated scope limitations, rather than by any law enforcement officer”); United States ex rel. Terraciano v. Montanye, 493 F. 2d 682, 685 (CA2 1974) (Friendly, J.) (emphasizing the amendment of the New York statute on inspection of drug records “to restrict the right of inspection to representatives of the Health Department, . . . rather than ‘all peace officers within the state’ ”). 728 OCTOBER TERM, 1986 Brennan, J., dissenting 482 U. S. ministrative violations may also be crimes, and valid administrative inspections sometimes uncover evidence of crime; neither of these facts necessarily creates constitutional problems with an inspection scheme. In this case, the problem is entirely different. In no other administrative search case has this Court allowed the State to conduct an “administrative search” which violated no administrative provision and had no possible administrative consequences.17 The Court thus implicitly holds that if an administrative scheme has certain goals and if the search serves those goals, it may be upheld even if no concrete administrative consequences could follow from a particular search. This is a dangerous suggestion, for the goals of administrative schemes often overlap with the goals of the criminal law. Thus, on the Court’s reasoning, administrative inspections would evade the requirements of the Fourth Amendment so long as they served an abstract administrative goal, such as the prevention of automobile theft. A legislature cannot abrogate constitutional protections simply by saying that the purpose of an administrative search scheme is to prevent a certain type of crime. If the Fourth Amendment is to retain meaning in the commercial context, it must be applied to searches for evidence of criminal acts'even if those searches would also serve an administrative purpose, unless that administrative purpose takes the concrete form of seeking an administrative violation.18 17 This case thus does not present the more difficult question whether a State could take any criminal conduct, make it an administrative violation, and then search without probable cause for violations of the newly created administrative rule. The increasing overlap of administrative and criminal violations creates an obvious temptation for the State to do so, and plainly toleration of this type of pretextual search would allow an end run around the protections of the Fourth Amendment. 18 Today’s holding, of course, does not preclude consideration of the lawfulness of the search under the State Constitution. See People v. P. J-Video, Inc., 68 N. Y. 2d 296, 501 N. E. 2d 556 (1986); People v. Class, 67 N. Y. 2d 431, 494 N. E. 2d 444 (1986). NEW YORK v. BURGER 729 691 Brennan, J., dissenting IV The implications of the Court’s opinion, if realized, will virtually eliminate Fourth Amendment protection of commercial entities in the context of administrative searches. No State may require, as a condition of doing business, a blanket submission to warrantless searches for any purpose. I respectfully dissent. 730 OCTOBER TERM, 1986 Syllabus 482 U. S. KENTUCKY v. STINCER CERTIORARI TO THE SUPREME COURT OF KENTUCKY No. 86-572. Argued April 22, 1987—Decided June 19, 1987 After a jury was sworn at respondent’s Kentucky trial for committing sodomy with two minor girls, but before the presentation of evidence, the court conducted an in-chambers hearing to determine the girls’ competency to testify. Respondent, but not his counsel, was excluded from this hearing. Under Kentucky law, when a child’s competency to testify is raised, the judge is required to resolve whether the child is capable of observing, recollecting, and narrating the facts, and whether the child has a moral sense of the obligation to tell the truth. Thus, during the hearing, the judge and the attorneys limited themselves to questions designed to determine whether the girls were capable of remembering basic facts and of distinguishing between truth and falsehood. The judge ruled that both girls were competent to testify. Before each girl began her substantive testimony in open court, the prosecutor repeated some of the background questions asked at the hearing, while respondent’s counsel, on cross-examination, repeated other such questions, particularly those regarding the girls’ ability to distinguish truth from lies. After the girls’ testimony was complete, respondent’s counsel did not request that the court reconsider its competency rulings. Respondent was convicted, but the Kentucky Supreme Court reversed, holding that respondent’s exclusion from the competency hearing violated his right to confront the witnesses against him. Held: 1. Respondent’s rights under the Confrontation Clause of the Sixth Amendment were not violated by his exclusion from the competency hearing. Pp. 736-744. (a) The Confrontation Clause’s functional purpose is to promote reliability in criminal trials by ensuring a defendant an opportunity for cross-examination. Pp. 736-739. (b) Rather than attempting to determine whether a competency hearing is a “stage of trial” (as opposed to a pretrial proceeding) subject to the Confrontation Clause’s requirements, the more useful inquiry is whether excluding the defendant from the hearing interferes with his opportunity for cross-examination. No such interference occurred here, because the two girls were cross-examined in open court with respondent present and available to assist his counsel, and because any questions KENTUCKY v. STINGER 731 730 Syllabus asked during the hearing could have been repeated during direct and cross-examination. Moreover, the nature of the competency hearing militates against finding a Confrontation Clause violation, because questions at such hearings usually are limited to matters unrelated to basic trial issues. In addition, the judge’s responsibility to determine competency continues throughout the trial so that a competency determination may be reconsidered on motion after the substantive examination of the child. Pp. 739-744. 2. Respondent’s rights under the Due Process Clause of the Fourteenth Amendment were not violated by his exclusion from the competency hearing. The defendant’s due process right to be present at critical stages of a criminal proceeding if his presence would contribute to the fairness of the procedure is not implicated here in light of the particular nature of the competency hearing, whereby questioning was limited to competency issues and neither girl was asked about the substantive testimony she would give at trial. There is no indication that respondent’s presence at the hearing would have been useful in ensuring a more reliable competency determination. Pp. 745-747. 712 S. W. 2d 939, reversed. Blackmun, J., delivered the opinion of the Court, in which Rehnquist, C. J., and White, Powell, O’Connor, and Scalia, JJ., joined. Marshall, J., filed a dissenting opinion, in which Brennan and Stevens, JJ., joined, post, p. 748. Penny R. Warren, Assistant Attorney General of Kentucky, argued the cause for petitioner. With her on the briefs were David L. Armstrong, Attorney General, and John S. Gillig, Assistant Attorney General. Mark A. Posnansky, by appointment of the Court, 479 U. S. 1005, argued the cause and filed a brief for respondent.* *Briefs of amici curiae urging reversal were filed for the State of Arkansas et al. by Steve Clark, Attorney General of Arkansas, and Rodney A. Smolla, joined by the Attorneys General for their respective jurisdictions as follows: Don Siegelman of Alabama, Charles M. Oberly III of Delaware, Robert A. Butterworth of Florida, Charles Troutman of Guam, Jim Jones of Idaho, Linley E. Pearson of Indiana, Neil F. Hartigan of Illinois, Robert T. Stephan of Kansas, 'William J. Guste, Jr., of Louisiana, Frank J. Kelley of Michigan, Hubert H. Humphrey III of Minnesota, Edwin L. Pittman 732 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. Justice Blackmun delivered the opinion of the Court. The question presented in this case is whether the exclusion of a defendant from a hearing held to determine the competency of two child witnesses to testify violates the defendant’s rights under the Confrontation Clause of the Sixth Amendment or the Due Process Clause of the Fourteenth Amendment. I Respondent Sergio Stincer was indicted in the Circuit Court of Christian County, Ky., and charged with committing first-degree sodomy with T. G., an 8-year-old girl, N. G., a 7-year-old girl, and B. H., a 5-year-old boy, in violation of Ky. Rev. Stat. §510.070 (1985). After a jury was sworn, but before the presentation of evidence, the court conducted an in-chambers hearing to determine if the two young girls were competent to testify.1 Over his objection, of Mississippi, William L. Webster of Missouri, Mike Greely of Montana, Brian McKay of Nevada, Stephen E. Merrill of New Hampshire, W. Cary Edwards of New Jersey, Hal Stratton of New Mexico, Lacy H. Thornburg of North Carolina, Nicholas Spaeth of North Dakota, Dave Frohn-mayer of Oregon, LeRoy S. Zimmerman of Pennsylvania, James E. O’Neil of Rhode Island, T. Travis Medlock of South Carolina, Roger A. Tellinghuisen of South Dakota, W. J. Michael Cody of Tennessee, Jeffrey Amestoy of Vermont, J’Ada Finch-Sheen of the Virgin Islands, Charles G. Brown of West Virginia, and Donald J. Hanaway of Wisconsin; and for the Appellate Committee of the California District Attorneys Association by Ira Reiner and Harry B. Sondheim. Briefs of amicus curiae urging affirmance were filed for the American Civil Liberties Union by George Kannar; and for the National Association of Criminal Defense Lawyers by Nancy Hollander. Donald N. Bersoff filed a brief for the American Psychological Association as amicus curiae. ’Immediately prior to the competency hearing of the two girls, the prosecutor moved that the charge regarding B. H., the 5-year-old boy, be dismissed because the prosecution did not believe B. H. was competent to testify. Respondent did not object and the court granted the prosecutor’s motion. Tr. 13-14. KENTUCKY v. STINGER 733 730 Opinion of the Court respondent, but not his counsel (a public defender), was excluded from this hearing. Tr. 15. The two children were examined separately and the judge, the prosecutor, and respondent’s counsel asked questions of each girl to determine if she were capable of remembering basic facts and of distinguishing between telling the truth and telling a lie. Id., at 15-26. T. G., the 8-year-old, was asked her age, her date of birth, the name of her school, the names of her teachers, and the name of her Sunday school. She was also asked whether she knew what it meant to tell the truth, and whether she could keep a promise to God to tell the truth. Id., at 16-18.2 N. G., the 7-year-old girl, was asked similar questions. Id., at 20-25.3 The two children were not asked about the substance of the testimony they were to give at trial. The court ruled that the girls were competent to testify. Respondent’s counsel did not object to these rulings. Id., at 20, 25. Before each of the girls began her substantive testimony in open court, the prosecutor repeated some of the basic questions regarding the girl’s background that had been asked at the competency hearing. Id., at 31-33 (direct examination of T. G.) (questions regarding age, where the witness attended school and Sunday school, and the like); id., at 66 (direct examination of N. G.) (questions regarding age and where the witness attended school). T. G. then testified, on direct examination, that respondent had placed a sock over her eyes, had given her chocolate pudding to eat, and then had “put his d-i-c-k” in her mouth. Id., at 34. N. G., on direct examination, testified to a similar incident. Id., at 69.4 2 In response to these questions, T. G. stated that telling the truth meant “[d]on’t tell no stories.” Id., at 17. 3 N. G. replied that she would “get a whopping” if she told a lie. Id., at 24. 4 There is some confusion as to whether T. G. knew what a “d-i-c-k” was, although she spelled the word at trial. Id., at 55-58. It also 734 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. On cross-examination, respondent’s counsel asked each girl questions designed to determine if she could remember past events and if she knew the difference between the truth and a lie. Some of these questions were similar to those that had been asked at the competency hearing. See id., at 38-39, 44-47, 60-63 (cross-examination of T. G.); 71-72, 74-75, 78-83 (cross-examination of N. G.). After the testimony of the girls was concluded, counsel did not request that the trial court reconsider its ruling that the girls were competent to testify.5 The jury convicted respondent of first-degree sodomy for engaging in deviate sexual intercourse and fixed his sentence at 20 years’ imprisonment.6 appears that N. G. may have recanted her testimony somewhat on cross-examination. Id., at 77-78. These facts, however, relate to whether the evidence was sufficient to convict respondent of the crimes charged. The Kentucky Supreme Court concluded that the evidence was sufficient to withstand a motion for a directed verdict of acquittal. 712 S. W. 2d 939, 941 (1986). That ruling is not before us in this case. 6 After the two girls testified, the prosecution stated that it also wished to present the testimony of E. T., a 4-year-old boy who allegedly had witnessed the events in question. The court examined E. T. in the courtroom, without the jury present and, apparently, without respondent present. Tr. 87. No objection from respondent regarding his exclusion from this hearing appears on the record. The court ruled that the boy was competent to testify, a ruling to which respondent’s counsel apparently objected. Id., at 109-110. After direct and cross-examination of E. T., defense counsel moved that the court reconsider its previous ruling that the boy was competent to testify. The court declined to rule that he was incompetent. Id., at 126-127. Respondent’s exclusion from E. T.’s competency hearing is not before us because the validity of respondent’s absence from that hearing was never raised before the Kentucky Supreme Court. See Brief for Appellant in No. 84-SC-496-I (Ky. Sup. Ct.), pp. 14-17. Thus, not surprisingly, the majority opinion of the Kentucky Supreme Court refers solely to the competency hearing of the two girls. 6 Under Kentucky law, deviate sexual intercourse means “any act of sexual gratification between persons not married to each other involving the sex organs of one person and the mouth or anus of another.” Ky. Rev. KENTUCKY v. STINGER 735 730 Opinion of the Court On appeal to the Supreme Court of Kentucky, respondent argued, among other things, that his exclusion from the competency hearing of the two girls denied him due process and violated his Sixth Amendment right to confront the witnesses against him. The Kentucky Supreme Court, by a divided vote, agreed that, under the Sixth Amendment of the Federal Constitution and under § 11 of the Bill of Rights of the Kentucky Constitution (the right “to meet the witnesses face to face”), respondent had an absolute right to be present at the competency hearing because the hearing “was a crucial phase of the trial.” 712 S. W. 2d 939, 940 (1986). The court explained that respondent’s trial “might not have taken place had the trial court determined that the children were not competent to testify.” Id., at 941. Two justices, however, dissented, concluding that respondent’s right to confront the witnesses against him was not violated because respondent had the opportunity to assist counsel fully in cross-examining the two witnesses at trial. Id., at 942-944. We granted certiorari, 479 U. S. 1005 (1986), to determine whether respondent’s constitutional rights were violated by his exclusion from the competency hearing.7 Stat. § 510.010(1) (1985). First-degree sodomy with a child under 12 is a Class A felony and conviction carries a minimum sentence of 20 years’ imprisonment and a maximum sentence of life imprisonment. §§ 510.070(2) and 532.060. 7 As an initial matter, respondent asks us to vacate our grant of certiorari because, in his view, the decision of the Kentucky Supreme Court rests on “‘separate, adequate, and independent grounds.’” Brief for Respondent 50, quoting Michigan v. Long, 463 U. S. 1032, 1041 (1983). We decline to do so. In Michigan v. Long, we explained that “when ... a state court decision fairly appears ... to be interwoven with the federal law, and when the adequacy and independence of any possible state law ground is not clear from the face of the opinion,” we shall assume that the state court believed that federal law compelled its conclusion. Id., at 1040-1041. In this case, the Kentucky Supreme Court consistently referred to respondent’s rights under the Sixth Amendment to the Federal Constitution as supporting its ruling. The court gave no indication that 736 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. II A The Sixth Amendment’s Confrontation Clause provides: “In all criminal prosecutions, the accused shall enjoy the right... to be confronted with the witnesses against him.” This right is secured for defendants in state as well as in federal criminal proceedings. Pointer v. Texas, 380 U. S. 400 (1965). The Court has emphasized that “a primary interest secured by [the Confrontation Clause] is the right of cross-examination.” Douglas n. Alabama, 380 U. S. 415, 418 (1965). The opportunity for cross-examination, protected by the Confrontation Clause, is critical for ensuring the integrity of the factfinding process. Cross-examination is “the principal means by which the believability of a witness and the truth of his testimony are tested.” Davis n. Alaska, 415 U. S. 308, 316 (1974). Indeed, the Court has recognized that cross-examination is the “ ‘greatest legal engine ever invented for the discovery of truth.’” California v. Green, 399 U. S. 149, 158 (1970), quoting 5 J. Wigmore, Evidence § 1367, p. 29 (3d ed. 1940). The usefulness of cross-examination was emphasized by this Court in an early case explicating the Confrontation Clause: “The primary object of the constitutional provision in question was to prevent depositions or ex parte affidavits . . . being used against the prisoner in lieu of a personal examination and cross-examination of the witness in which the accused has an opportunity, not only of testing the recollection and sifting the conscience of the witness, but of compelling him to stand face to face with the jury in order that they may look at him, and judge by his demeanor upon the stand and the manner in which he respondent’s rights under § 11 of the Bill of Rights of the Kentucky Constitution were distinct from, or broader than, respondent’s rights under the Sixth Amendment. KENTUCKY v. STINGER 737 730 Opinion of the Court gives his testimony whether he is worthy of belief.” Mattox v. United States, 156 U. S. 237, 242-243 (1895).8 See also Kirby v. United States, 174 U. S. 47, 53 (1899). The right to cross-examination, protected by the Confrontation Clause, thus is essentially a “functional” right designed to promote reliability in the truth-finding functions of a criminal trial. The cases that have arisen under the Confrontation Clause reflect the application of this functional right. These cases fall into two broad, albeit not exclusive, categories: “cases involving the admission of out-of-court statements and cases involving restrictions imposed by law or by the trial court on the scope of cross-examination.” Delaware v. Fensterer, 474 U. S. 15, 18 (1985) (per curiam). In the first category of cases, the Confrontation Clause is violated when “hearsay evidence [is] admitted as substantive evidence against the defendan[t],” Tennessee v. Street, 471 U. S. 409, 413 (1985), with no opportunity to cross-examine the hearsay declarant at trial, or when an out-of-court statement of an unavailable witness does not bear adequate indications of trustworthiness. See Ohio v. Roberts, 448 U. S. 56, 65-66 (1980). For example, in Roberts, we held that an out-of-court statement by an unavailable witness was sufficiently reliable to be admitted at trial, consistent with the Confrontation Clause, because defense counsel had engaged in full cross-examination of the witness at the preliminary hearing where the statement was made. Id., at 70-73. In California v. Green, supra, the Court concluded that the Confrontation Clause was not violated by admitting a declarant’s inconsistent out-of-court statement “as long as the de 8 One noted commentator has pointed out that the main purpose of confrontation “is to secure for the opponent the opportunity of cross-examination” (emphasis omitted), 5 J. Wigmore, Evidence § 1395, p. 150 (Chadbourn rev. 1974) (Wigmore), with an additional advantage being that “the judge and the jury are enabled to obtain the elusive and incommunicable evidence of a witness’ deportment while testifying” (emphasis omitted). Id., at 153. 738 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. clarant is testifying as a witness and subject to full and effective cross-examination” at the trial itself. 399 U. S., at 158. The second category involves cases in which the opportunity for cross-examination has been restricted by law or by a trial court ruling. In Davis v. Alaska, supra, defense counsel was restricted by state confidentiality provisions from questioning a witness about his juvenile criminal record, although such evidence, might have affected the witness’ credibility. The Court held that the Confrontation Clause was violated because the defendant was denied the right “to expose to the jury the facts from which jurors . . . could appropriately draw inferences relating to the reliability of the witness.” 415 U. S., at 318. Similarly, in Delaware v. Van Arsdall, 475 U. S. 673 (1986), defense counsel was precluded by the trial court from questioning a witness about the State’s dismissal of a pending public drunkenness charge against him. The Court concluded: “By thus cutting off all questioning about an event. . . that a jury might reasonably have found furnished the witness a motive for favoring the prosecution in his testimony,” the trial court’s ruling violated the defendant’s rights under the Confrontation Clause. Id., at 679.9 9 The Court sometimes has referred to a defendant’s right of confrontation as a “trial right.” See Barber v. Page, 390 U. S. 719, 725 (1968); see also California v. Green, 399 U. S. 149, 157 (1970) (“right to ‘confront’ the witness at the time of trial”). In Pennsylvania v. Ritchie, 480 U. S. 39 (1987), a plurality of the Court interpreted the Clause to mean that the right of confrontation is designed simply “to prevent improper restrictions on the types of questions that defense counsel may ask during cross-examination.” Id., at 52. Thus, the plurality in Ritchie concluded that the constitutional error in Davis v. Alaska, 415 U. S. 308 (1974), was not that state law made certain juvenile criminal records confidential, but rather that the defense attorney had been precluded from asking questions about that criminal record at trial. 480 U. S., at 54. The personal view of the author of this opinion as to the Confrontation Clause is somewhat broader than that of the Ritchie plurality. Although he believes that “[t]here are cases, perhaps most of them, where simple questioning of a witness will satisfy the purposes of cross-examination,” id., at 62 (Black- KENTUCKY v. STINGER 739 730 Opinion of the Court Although daims arising under the Confrontation Clause may not always fall neatly into one of these two categories, these cases reflect the Confrontation Clause’s functional purpose in ensuring a defendant an opportunity for cross-examination. See Lee v. Illinois, 476 U. S. 530 (1986). Of course, the Confrontation Clause guarantees only “an opportunity for effective cross-examination, not cross-examination that is effective in whatever way, and to whatever extent, the defense might wish.” Delaware v. Fensterer, 474 U. S., at 20 (emphasis in original). This limitation is consistent with the concept that the right to confrontation is a functional one for the purpose of promoting reliability in a criminal trial. B The Commonwealth argues that respondent’s exclusion from the competency hearing of the two children did not violate the Confrontation Clause because a competency hearing is not “a stage of trial where evidence or witnesses are being presented to the trier of fact.” Brief for Petitioner 22. Cf. Gannett Co. v. DePasquale, 443 U. S. 368, 394 (1979) (Burger, C. J., concurring). Distinguishing between a “trial” and a “pretrial proceeding” is not particularly helpful here, however, because a competency hearing may well be a “stage of trial.” In this case, for instance, the competency hearing was held after the jury was sworn, in the judge’s chambers, and in the presence of opposing counsel who asked questions mun, J., concurring), he also believes that there are cases in which a state rule that precludes a defendant from access to information before trial may hinder that defendant’s opportunity for effective cross-examination at trial, and thus that such a rule equally may violate the Confrontation Clause. Id., at 63-65. His differences with the plurality in Ritchie, however, are not implicated in this case. As is demonstrated below, respondent’s ability to engage in full cross-examination at trial was not affected by his exclusion from the competency hearing, nor was his opportunity to engage in effective cross-examination interfered with by his exclusion. Thus, under either the author’s view or that of the plurality in Ritchie, there was no Confrontation Clause violation in this case. 740 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. of the witnesses.10 Moreover, although questions regarding the guilt or innocence of the defendant usually are not asked at a competency hearing, the hearing retains a direct relationship with the trial because it determines whether a key witness will testify. Further, although the preliminary determination of a witness’ competency to testify is made at this hearing, the determination of competency is an ongoing one for the judge to make based on the witness’ actual testimony at trial. Instead of attempting to characterize a competency hearing as a trial or pretrial proceeding, it is more useful to consider whether excluding the defendant from the hearing interferes with his opportunity for effective cross-examination. No such interference occurred when respondent was excluded from the competency hearing of the two young girls in this case. After the trial court determined that the two children were competent to testify, they appeared and testified in open court. At that point, the two witnesses were subject to full and complete cross-examination, and were so examined. Tr. 38-58 (cross-examination of T. G.); id., at 71-84 (cross-examination of N. G.). Respondent was present throughout this cross-examination and was available to assist his counsel as necessary. There was no Kentucky rule of law, nor any ruling by the trial court, that restricted respondent’s ability to cross-examine the witnesses at trial. Any questions asked during the competency hearing, which respondent’s counsel attended and in which he participated, could have been repeated during direct examination and cross-examination of the witnesses in respondent’s presence. See California v. Green, 399 U. S., at 159 (“[T]he inability to cross-examine the witness at the time he made his prior statement cannot easily be shown to be of crucial significance as long as the defendant is assured of full and effective cross-examination at the time of trial”). 10 Indeed, a competency hearing may take place in the middle of a trial, as did the hearing of E. T. See n. 5, supra. KENTUCKY v. STINGER 741 730 Opinion of the Court Moreover, the type of questions that were asked at the competency hearing in this case were easy to repeat on cross-examination at trial. Under Kentucky law, when a child’s competency to testify is raised, the judge is required to resolve three basic issues: whether the child is capable of observing and recollecting facts, whether the child is capable of narrating those facts to a court or jury, and whether the child has a moral sense of the obligation to tell the truth. See Moore v. Commonwealth, 384 S. W. 2d 498, 500 (Ky. 1964) (“When the competency of an infant to testify is properly raised it is then the duty of the trial court to carefully examine the witness to ascertain whether she (or he) is sufficiently intelligent to observe, recollect and narrate the facts and has a moral sense of obligation to speak the truth”); Capps v. Commonwealth, 560 S. W. 2d 559, 560 (Ky. 1977); Hendricks n. Commonwealth, 550 S. W. 2d 551, 554 (Ky. 1977); see also Thomas v. Commonwealth, 300 Ky. 480, 481-482, 189 S. W. 2d 686, 686-687 (1945); Comment, An Overview of the Competency of Child Testimony, 13 No. Ky. L. Rev. 181, 184 (1986).11 Thus, questions at a competency hearing usually are limited to matters that are unrelated to the basic issues of the trial. Children often are asked their names, where they go to school, how old they are, whether they know who the judge is, whether they know what a lie is, and whether they know what happens when one tells a lie. See Comment, The Competency Requirement for the Child Victim of Sexual Abuse: Must We Abandon It?, 40 U. Miami L. Rev. 245, 263, and n. 78 (1985); Comment, Defendants’ Rights in Child Witness Competency Hearings: Establishing 11 Similar requirements for establishing competency to testify were set forth in Wheeler v. United States, 159 U. S. 523 (1895): “[T]here is no precise age which determines the question of competency. This depends on the capacity and intelligence of the child, his appreciation of the difference between truth and falsehood, as well as of his duty to tell the former.” Id., at 524. See generally 2 Wigmore §§ 505-507. 742 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. Constitutional Procedures for Sexual Abuse Cases, 69 Minn. L. Rev. 1377, 1381-1383, and nn. 9-11 (1985).12 In Kentucky, as in certain other States, it is the responsibility of the judge, not the jury, to decide whether a witness is competent to testify based on the witness’ answers to such questions. Whitehead v. Stith, 268 Ky. 703, 709, 105 S. W. 2d 834, 837 (1937) (question of competency is one for court, not jury, and if court finds witness lacks qualification, “it commits a palpable abuse of its discretion” should it then permit witness to testify); Payne n. Commonwealth, 623 S. W. 2d 867, 878 (Ky. 1981); Capps n. Commonwealth, 560 S. W. 12 Some States explicitly allow children to testify without requiring a prior competency qualification, while others simply provide that all persons, including children, are deemed competent unless otherwise limited by statute. See B. Battman & J. Bulkley, National Legal Resource Center for Child Advocacy and Protection, Protecting Child Victim/Witnesses: Sample Laws and Materials 43-44 (1986) (listing statutes) (Protecting Child Victim/Witnesses); Bulkley, Evidentiary and Procedural Trends in State Legislation and Other Emerging Legal Issues in Child Sexual Abuse Cases, 89 Dick. L. Rev. 645, 645 (1985). Some commentators have urged that children be allowed to testify without undergoing a prior competency qualification. See Protecting Child Victim/Witnesses, at 38 (proposing sample competency statute according children same rebuttable presumption of competency granted other witnesses); 2 Wigmore § 509, p. 719 (“it must be concluded that the sensible way is to put the child upon the stand to give testimony for what it may seem to be worth”). A number of States, however, mandate by statute that a trial judge assess a child’s competency to testify on the basis of specified requirements. These usually include a determination that the child is capable of expression, is capable of understanding the duty to tell the truth, and is capable of receiving just impressions of the facts about which he or she is called to testify. See, e. g., Ariz. Rev. Stat. Ann. § 12-2202 (1982); Ga. Code Ann. §24-9-5 (1982); Idaho Code §9-202 (Supp. 1987); Ind. Code §34-1-14-5 (1986); Mich. Comp. Laws § 600.2163 (1986); Minn. Stat. §595.02 Subd. 1(f) (Supp. 1987); N. Y. Crim. Proc. Law § 60.20 (McKinney 1981); Ohio Rev. Code Ann. § 2317.01 (1981); see Protecting Child Victim/Witnesses, at 45 (listing statutes). The recent reforms in some States of presuming the competency of young children and allowing juries to assess credibility at trial is not called into question by this opinion. We are concerned solely with those States that retain competency qualification requirements. KENTUCKY v. STINGER 743 730 Opinion of the Court 2d, at 560. See 2 Wigmore § 507, p. 714 (citing cases). In those States where the judge has the responsibility for determining competency, that responsibility usually continues throughout the trial.13 A motion by defense counsel that the court reconsider its earlier decision that a child is competent may be raised after the child testifies on direct examination, see, e. g., In re R. R., 79 N. J. 97, 106, 398 A. 2d 76, 80 (1979) (at close of State’s case, defense attorney moved that 4-year-old boy be declared incompetent on basis of actual testimony given by boy),14 or after direct and cross-examination of the witness. See, e. g., Reply Brief for Petitioner 12 (“If, during trial, there arises some basis for challenging the judge’s competency determination, the judge may be asked to reconsider,” referring to respondent’s motion to that effect, Tr. 126-127). Moreover, appellate courts reviewing a trial judge’s determination of competency also often will look at the full testimony at trial.15 13 See, e. g., Litzkuhn v. Clark, 85 Ariz. 355, 360, 339 P. 2d 389, 392 (1959) (“[I]t is the duty of the trial judge who has permitted a child to be sworn as a witness, at any time to change his mind upon due occasion therefor, to remove the child from the stand and to instruct the jury to disregard his testimony”); Davis v. Weber, 93 Ariz. 312, 317, 380 P. 2d 608, 611 (1963) (“The right of a trial judge to change his mind [regarding a child’s competency] can hardly be denied”). 14 California recently amended its statute governing the disqualification of incompetent witnesses to provide explicitly: “In any proceeding held outside the presence of a jury, a court may reserve challenges to the competency of a witness until the conclusion of the direct examination of that witness.” Cal. Evid. Code Ann. § 701(b) (West Supp. 1987). 15 See, e. g., Payne v. Commonwealth, 623 S. W. 2d 867, 878 (Ky. 1981) (review of children’s testimony at trial reveals that trial court’s ruling of competency was appropriate); Hendricks v. Commonwealth, 550 S. W. 2d 551, 554 (Ky. 1977) (“Not only did the trial judge determine that the children were competent to testify, but the transcript of the testimony of these children clearly demonstrates their intellectual ability to observe, recollect and narrate the facts and to recognize their moral obligation to tell the truth”); see also In re R. R., 79 N. J. 97, 113, 398 A. 2d 76, 84 (1979) (“[I]n determining the propriety of the trial judge’s determination, an appellate court need not limit its view to the responses given by the witness during 744 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. In this case both T. G. and N. G. were asked several background questions during the competency hearing, as well as several questions directed at what it meant to tell the truth. Some of the questions regarding the witnesses’ backgrounds were repeated by the prosecutor on direct examination, while others—particularly those regarding the witnesses’ ability to tell the difference between truth and falsehood—were repeated by respondent’s counsel on cross-examination. At the close of the children’s testimony, respondent’s counsel, had he thought it appropriate, was in a position to move that the court reconsider its competency rulings on the ground that the direct and cross-examination had elicited evidence that the young girls lacked the basic requisites for serving as competent witnesses.16 Thus, the critical tool of cross-examination was available to counsel as a means of establishing that the witnesses were not competent to testify, as well as a means of undermining the credibility of their testimony. Because respondent had the opportunity for full and effective cross-examination of the two witnesses during trial, and because of the nature of the competency hearing at issue in this case, we conclude that respondent’s rights under the Confrontation Clause were not violated by his exclusion from the competency hearing of the two girls.17 the voir dire examination; instead, it can consider the entire record— including the testimony in fact given by the witness under oath—in order to arrive at its decision”). 16 Respondent’s counsel, in fact, did move for reconsideration of the court’s ruling on the competency of E. T. after that young boy had testified and had been subjected to cross-examination. See n. 5, supra. 17 We note once again that the Kentucky Supreme Court held that respondent’s confrontation rights were violated because the competency hearing was a “crucial phase of the trial.” 712 S. W. 2d, at 940. It is true that the hearing was crucial in the sense that respondent may not have been convicted had the two girls been found incompetent to testify. Nevertheless, the question whether a particular proceeding is critical to the outcome of a trial is not the proper inquiry in determining whether the Confrontation Clause has been violated. The appropriate question is whether there has been any interference with the defendant’s opportunity KENTUCKY v. STINGER 745 730 Opinion of the Court III Respondent argues that his rights under the Due Process Clause of the Fourteenth Amendment were violated by his exclusion from the competency hearing.18 The Court has assumed that, even in situations where the defendant is not actually confronting witnesses or evidence against him, he has a due process right “to be present in his own person whenever his presence has a relation, reasonably substantial, to the fulness of his opportunity to defend against the charge.” Snyder v. Massachusetts, 291 U. S. 97, 105-106 (1934). Although the Court has emphasized that this privilege of presence is not guaranteed “when presence would be useless, or the benefit but a shadow,” id., at 106-107, due process clearly requires that a defendant be allowed to be present “to the extent that a fair and just hearing would be thwarted by his absence,” id., at 108. Thus, a defendant is guaranteed the right to be present at any stage of the criminal proceeding that is critical to its outcome if his presence would contribute to the fairness of the procedure. We conclude that respondent’s due process rights were not violated by his exclusion from the competency hearing in this case. We emphasize, again, the particular nature of the competency hearing. No question regarding the substantive testimony that the two girls would have given during trial for effective cross-examination. No such interference occurred in this case. Of course, the fact that a stage in the proceeding is critical to the outcome of a trial may be relevant to due process concerns. Even in that context, however, the question is not simply whether, “but for” the outcome of the proceeding, the defendant would have avoided conviction, but whether the defendant’s presence at the proceeding would have contributed to the defendant’s opportunity to defend himself against the charges. See infra, Part III. 18 Although respondent perhaps could have been more artful in presenting his due process claim to the Kentucky Supreme Court as clearly founded on the Fourteenth Amendment, he did raise a due process claim to that court, see Brief for Appellant in No. 84-SC-496-I (Ky. Sup. Ct.), pp. 14-17, and the claim therefore is properly before us. 746 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. was asked at that hearing. All the questions, instead, were directed solely to each child’s ability to recollect and narrate facts, to her ability to distinguish between truth and falsehood, and to her sense of moral obligation to tell the truth.19 Thus, although a competency hearing in which a witness is asked to discuss upcoming substantive testimony might bear a substantial relationship to a defendant’s opportunity better to defend himself at trial, that kind of inquiry is not before us in this case.20 19 During the competency hearing of E. T., the judge, the prosecutor, and respondent’s counsel asked the boy several questions regarding the substance of his testimony. Tr. 91-101. As noted above, however, see n. 5, supra, respondent’s exclusion from E. T.’s competency hearing is not before us. 20 Counsel for the Commonwealth acknowledged that if a competency hearing “were to exceed its normal scope,” that would “begi[n] to bear a substantial relation to [a defendant’s] opportunity to defend.” Tr. of Oral Arg. 9; see also State v. Howard, 57 Ohio App. 2d 1, 4-5, 385 N. E. 2d 308, 312-313 (1978) (defendant’s presence bears reasonably substantial relation to defense when witnesses give testimony in in-camera hearing identifying defendant as assailant). Although, as noted above, most competency hearings do not focus on substantive testimony, it is not impossible that questions related to substantive testimony could be asked. See Comment, Defendants’ Rights in Child Witness Competency Hearings: Establishing Constitutional Procedures for Sexual Abuse Cases, 69 Minn. L. Rev. 1377, 1384 (1985) (emphasis on testing a child’s memory “suggests that a judge may inquire about the actual sexual assault”); see n. 19, supra. But see Moll n. State, 351 N. W. 2d 639, 643 (Minn. App. 1984) (“[T]he trial court has broad discretion as to the type of question to be put to the child during this preliminary examination, but should not elicit from the child the anticipated testimony concerning the alleged offense, recognizing the suggestibility of young children”). Where the competency hearing bears a substantial relationship to the defendant’s opportunity to defend, a court must then balance the defendant’s role in assisting in his defense against the risk of identifiable and substantial injury to the specific child witness. See Brief for American Psychological Association as Amicus Curiae 15-26 (noting that intuitive view that child victims of sexual abuse are particularly vulnerable in legal proceedings may not be correct for all children). KENTUCKY v. STINGER 747 730 Opinion of the Court Respondent has given no indication that his presence at the competency hearing in this case would have been useful in ensuring a more reliable determination as to whether the witnesses were competent to testify. He has presented no evidence that his relationship with the children, or his knowledge of facts regarding their background, could have assisted either his counsel or the judge in asking questions that would have resulted in a more assured determination of competency. On the record of this case, therefore, we cannot say that respondent’s rights under the Due Process Clause of the Fourteenth Amendment were violated by his exclusion from the competency hearing.21 As was said in United States v. Gagnon, 470 U. S. 522, 527 (1985) (per curiam), there is no indication that respondent “could have done [anything] had [he] been at the [hearing] nor would [he] have gained anything by attending.”22 21 Contrary to the dissent’s charge, see post, at 754, we do not address the question whether harmless-error analysis applies in the situation where a defendant is excluded from a critical stage of the proceedings in which his presence would contribute to the fairness of the proceeding. In this case, respondent simply has failed to establish that his presence at the competency hearing would have contributed to the fairness of the proceeding. He thus fails to establish, as an initial matter, the presence of a constitutional deprivation. 22 Respondent also argues that his Sixth Amendment right to effective assistance of counsel was violated by his inability to consult with counsel during the competency hearing. Brief for Respondent 28-32. Respondent acknowledges that this argument was not raised below, id., at 28, n. 25, but he argues that, as the prevailing party, he may assert any ground in support of his judgment. See Dandridge v. Williams, 397 U. S. 471,475, n. 6 (1970). The Court has noted, however, that it is “the settled practice of this Court, in the exercise of its appellate jurisdiction, that it is only in exceptional cases, and then only in cases coming from the federal courts, that it considers questions urged by a petitioner or appellant not pressed or passed upon in the courts below.” McGoldrick v. Compagnie Generate Transatlantique, 309 U. S. 430, 434 (1940); see also Heckler v. Campbell, 461 U. S. 458, 468-469, n. 12 (1983) (Court will consider ground not presented to federal court below only “in exceptional cases”). Because 748 OCTOBER TERM, 1986 Marshall, J., dissenting 482 U. S. The judgment of the Supreme Court of Kentucky is reversed. It is so ordered. Justice Marshall, with whom Justice Brennan and Justice Stevens join, dissenting. The Court today defines respondent’s Sixth Amendment right to be confronted with the witnesses against him as guaranteeing nothing more than an opportunity to cross-examine these witnesses at some point during his trial. The Confrontation Clause protects much more. In this case, it secures at a minimum respondent’s right of presence to assist his lawyer at the in-chambers hearing to determine the competency of the key prosecution witnesses. Respondent’s claim under the Due Process Clause of the Fourteenth Amendment, though similar in this testimonial context to his claim under the Confrontation Clause, was not addressed by the court below and should not be decided here. Were this issue properly before the Court, however, I would again dissent. Due process requires that respondent be allowed to attend every critical stage of his trial. I The Sixth Amendment guarantees the criminal defendant “the right ... to be confronted with the witnesses against him.” The text plainly envisions that witnesses against the accused shall, as a rule, testify in his presence. I can only marvel at the manner in which the Court avoids this manifest import of the Confrontation Clause. Without explanation, the Court narrows its analysis to address exclusively what is accurately identified as simply a primary interest the Clause was intended to secure: the right of cross-examination. See ante, at 736 (citing Douglas n. Alabama, 380 U. S. 415, 418 the judgment here is that of a state court, and because we do not believe that respondent’s claim of deprival of the effective assistance of counsel qualifies as an exceptional case, we decline to review this claim. KENTUCKY v. STINGER 749 730 Marshall, J., dissenting (1965)). This use of analytical blinders is undoubtedly convenient. Since respondent ultimately did receive an opportunity for full cross-examination of the witnesses in his presence, the narrowly drawn standard enables the Court to conclude with relative ease that respondent’s confrontation rights were not violated, see ante, at 740 and 744, even though the in-chambers competency hearing admittedly was, in this case, a “crucial” phase of respondent’s trial from which he was physically excluded. Ante, at 744-745, n. 17. Although cross-examination may be a primary means for ensuring the reliability of testimony from adverse witnesses, we have never held that standing alone it will suffice in every case. It is true that we have addressed in some detail the Confrontation Clause as it pertains to the admission of out-of-court statements, e. g., Ohio v. Roberts, 448 U. S. 56 (1980); California v. Green, 399 U. S. 149 (1970); and restrictions on the scope of cross-examination, e. g., Davis v. Alaska, 415 U. S. 308 (1974). But these cases have arisen in contexts in which the defendants’ right to be present during the testimony was never doubted, thus making the Court’s categorical analysis, see ante, at 737-738, largely beside the point. Not until today has this Court gone so far as to substitute a defendant’s subsequent opportunity for cross-examination for his right to confront adverse witnesses in a prior testimonial proceeding. Rather, the Court has taken care not to identify the right of cross-examination as the exclusive interest protected by the Confrontation Clause. That right is simply among those “included in” the defendant’s broad right to confront the witnesses against him. Pointer v. Texas, 380 U. S. 400, 404 (1965). Though “[c]on-frontation means more than being allowed to confront the witness physically,” Davis v. Alaska, supra, at 315, it must by implication encompass the right of physical presence at any testimonial proceeding. As this Court has previously recognized, “it is this literal right to ‘confront’ the witness at the time of trial that forms the core of the values furthered 750 OCTOBER TERM, 1986 Marshall, J., dissenting 482 U. S. by the Confrontation Clause,” California v. Green, supra, at 157, guaranteeing the accused an opportunity to compel the witness to meet him “face to face” before the trier of fact. Mattox v. United States, 156 U. S. 237, 242 (1895); see also Ohio v. Roberts, supra, at 63, and nn. 5, 6. Physical presence of the defendant enhances the reliability of the factfinding process. Under Kentucky law, in a witness competency proceeding the trial judge must assess the witness’ ability to observe and recollect facts with accuracy and with committed truthfulness. See ante, at 741. This determination necessarily requires the judge to make independent factual findings against which can be measured the accuracy of the witness’ testimony at the competency proceeding, whether addressing facts such as the witness’ name, age, and relation to the defendant, or events concerning the alleged offense itself. These findings are critical to the trial judge’s assessment of the witness’ competency to testify, and they often concern matters about which the defendant, and not his counsel, possesses the knowledge needed to expose inaccuracies in the witness’ answers. Having the defendant present ensures that these inaccuracies are called to the judge’s attention immediately—before the witness takes the stand with the trial court’s imprimatur of competency and testifies in front of the jury as to the defendant’s commission of the alleged offense. It is both functionally inefficient and fundamentally unfair to attribute to the defendant’s attorney complete knowledge of the facts which the trial judge, in the defendant’s involuntary absence, deems relevant to the competency determination. That determination, which turns entirely on the trial court’s evaluation of the witness’ statements, cannot be made out of the physical presence of the defendant without violating the basic guarantee of the Confrontation Clause: “[A] fact which can be primarily established only by witnesses cannot be proved against an accused . . . except by witnesses who confront him at the trial, upon whom KENTUCKY v. STINGER 751 730 Marshall, J., dissenting he can look while being tried, whom he is entitled to cross-examine, and whose testimony he may impeach in every mode authorized by the established rules governing the trial or conduct of criminal cases. The presumption of innocence of an accused attends him throughout the trial and has relation to every fact that must be established in order to prove his guilt beyond reasonable doubt.” Kirby v. United States, 174 U. S. 47, 55 (1899). But more than the reliability of the competency determination is at stake in this case. As we recently observed in Lee v. Illinois, 476 U. S. 530 (1986), the constitutional guarantee of the right of confrontation serves certain “symbolic goals” as well: “[T]he right to confront and cross-examine adverse witnesses contributes to the establishment of a system of criminal justice in which the perception as well as the reality of fairness prevails. To foster such a system, the Constitution provides certain safeguards to promote to the greatest possible degree society’s interest in having the accused and accuser engage in open and even contest in a public trial. The Confrontation Clause advances these goals by ensuring that convictions will not be based on the charges of unseen and unknown—and hence unchallengable—individuals.” Id., at 540. This appearance of fairness is woefully lacking in the present case. The Commonwealth did not request that respondent be excluded from the competency hearing. The trial judge raised this issue sua sponte, and only the personal protestations of respondent, a recent Cuban immigrant whose fluency in the English language was limited, preserved the issue for appeal.1 Neither the prosecuting attorney nor the trial 1 The relevant portion of the transcript of the in-chambers hearing reads in its entirety: “Mr. Rogers [the prosecutor]: We’re dealing here with seven and eightyear-old children and I think as a preliminary matter maybe the Court 752 OCTOBER TERM, 1986 Marshall, J., dissenting 482 U. S. judge articulated any reason for excluding him. From this defendant’s perspective, the specter of the judge, prosecutor, and court-appointed attorney conferring privately with the key prosecution witnesses was understandably upsetting. From a constitutional perspective, the unrequested and unjustified exclusion constitutes an intolerable subversion of the symbolic functions of the Confrontation Clause.2 should inquire of them to determine whether or not you believe that they’re competent to testify. Of course, that would still be up to the jury to determine, I understand, but I think you do have to make a preliminary decision. “The Court: Okay. Let’s bring them in one at a time. I think we need to get Mr. Stincer back in the courtroom while we’re interviewing these children in chambers. “Mr. Embry [respondent’s attorney]: We don’t have any problem with that, Judge. Sergio, we’re going to talk to the children, not about the case really but just to see if they’re old enough to understand the difference between telling a lie and telling the truth, that sort of thing and I think they’ll have you set [sic] outside. I will tell you what happens in a little bit. “Mr. Stincer: (phonetic). “Mr. Embry: I guess what he’s saying is, Judge, he wishes to be here. Of course, I think you’d probably have the right to handle it. “The Court: I think they’re going to have to be interviewed with counsel present only. I think I can exclude everyone. “Mr. Embry: Right, Judge. I just— “The Court: I’ll let counsel be present. “Mr. Embry: To protect my client, I’ll ask that he be allowed to stay. “The Court: Fine. Overruled. Let’s bring one of them in.” App. 1-2. 2 The reality and appearance of fairness are fully protected by the succinct holding of the Kentucky Supreme Court below: “A criminal defendant has the right to attend hearings to determine the competency of witnesses. The trial court’s determination of whether the prosecuting witnesses could testify was pivotal. Because the children’s testimony was sine qua non to the prosecution’s case, appellant’s trial might not have taken place had the trial court determined that the children were not competent to testify. “Although this court recognizes the problems and pressures encountered when dealing with child witnesses, when a defendant is placed on trial by the state for criminal conduct he is entitled to be present and to assist his counsel at hearings to determine the competency of witnesses against him.” 712 S. W. 2d 939, 941 (1986). KENTUCKY v. STINGER 753 730 Marshall, J., dissenting Had respondent invoked his Sixth Amendment right of self-representation and appeared pro se, there would be little doubt that he would have been entitled to attend the competency hearing and cross-examine the child witnesses. “The Sixth Amendment . . . grants to the accused personally the right to make his defense. It is the accused, not counsel, who must be ‘informed of the nature and cause of the accusation,’ who must be ‘confronted with the witnesses against him,’ and who must be accorded ‘compulsory process for obtaining witnesses in his favor.’” Faretta v. California, 422 U. S. 806, 819 (1975). A defendant who represents himself is “entitled to as much latitude in conducting his defense as we have held is enjoyed by counsel vigorously espousing a client’s cause.” In re Little, 404 U. S. 553, 555 (1972). Given these well-founded constitutional pronouncements, today’s decision may create for the criminal defendant a difficult dilemma: a choice between continuing to exercise his right to assistance of counsel, thereby being excluded from the competency hearing, and appearing pro se so that he may be in attendance at this critical stage of his trial. This Court has on occasion held that a forced choice between two fundamental constitutional guarantees is untenable, see Simmons n. United States, 390 U. S. 377, 394 (1968) (defendant’s testimony in support of motion to suppress evidence under the Fourth Amendment may not, under the Fifth Amendment, be admitted over objection at trial as evidence of defendant’s guilt). Today’s decision neglects the serious question whether this choice is constitutionally defensible. II Respondent’s right to be present at the competency hearing does not flow exclusively from the Sixth Amendment. The confrontation right attaches in this context because the competency proceeding was testimonial in nature. As the 754 OCTOBER TERM, 1986 Marshall, J., dissenting 482 U. S. Court acknowledges, however, respondent also claims a right independently grounded in the Fourteenth Amendment’s Due Process Clause to attend any trial proceeding in which his presence “has a relation, reasonably substantial, to the fulness of his opportunity to defend against the charge,” Snyder v. Massachusetts, 291 U. S. 97, 105-106 (1934). Ante, at 745; see also Faretta, supra, at 819, n. 15. That the competency hearing in this case bore a reasonably substantial relation to respondent’s defense can hardly be doubted. As the Court correctly acknowledges, “although questions regarding the guilt or innocence of the defendant usually are not asked at a competency hearing, the hearing retains a direct relationship with the trial because it determines whether a key witness will testify.” Ante, at 740. Reviewing the transcript of the competency hearing, the Court concludes that respondent’s due process rights were not violated because no question regarding the substantive testimony of the witnesses was asked and respondent has given no indication that his presence would have assisted in achieving more reliable competency determinations. Ante, at 745-747. But the propriety of the decision to exclude respondent from this critical stage of his trial should not be evaluated in light of what transpired in his absence. To do so transforms the issue from whether a due process violation has occurred into whether the violation was harmless. Neither issue was addressed by the court below. More importantly, however, the Court, citing a single per curiam decision, United States v. Gagnon, 470 U. S. 522 (1985), unfairly shifts the burden of proving harm from this constitutional deprivation to the excluded criminal defendant, who was in no way responsible for the error and is least able to demonstrate what would have occurred had he been allowed to attend. The Fourteenth Amendment does not permit this presumption that the involuntary exclusion of a defendant from a critical stage of his trial is harmless. I respectfully dissent. HEWITT v. HELMS 755 Syllabus HEWITT et AL. v. HELMS CERTIORARI TO THE UNITED STATES COURT OF APPEALS FOR THE THIRD CIRCUIT No. 85-1630. Argued March 4, 1987—Decided June 19, 1987 After more than seven weeks in administrative segregation pending an investigation into his possible involvement in a state prison riot, respondent inmate was found guilty of misconduct by a prison hearing committee and sentenced to six months of disciplinary confinement solely on the basis of an officer’s report of the statements of an undisclosed informant. Respondent filed suit against petitioner prison officials for damages and injunctive relief under 42 U. S. C. § 1983, but was released on parole before any decision was rendered. Subsequently, the Court of Appeals reversed the District Court’s entry of summary judgment against respondent, finding, inter alia, that his misconduct conviction constituted a denial of due process since it was based solely on hearsay. The District Court was instructed to enter summary judgment for respondent unless petitioners could establish an immunity defense, and was given authority to determine the appropriateness and availability of the relief respondent requested. On remand, respondent pursued only his damages claim. The District Court granted summary judgment for petitioners on the basis of qualified immunity, and the Court of Appeals affirmed. While the appeal was pending, the State Corrections Bureau (Bureau) revised its regulations to include procedures for the use of confidential source information in inmate disciplinary proceedings. The District Court then denied respondent’s claim for attorney’s fees on the ground that he was not a “prevailing party” as required by 42 U. S. C. § 1988, but the Court of Appeals reversed, concluding that its prior holding that his constitutional rights had been violated was “a form of judicial relief.” In the alternative, the court directed the District Court to reconsider whether respondent’s suit was a “catalyst” for the amendment of the Bureau’s regulations. Held: Respondent is not a “prevailing party” eligible for attorney’s fees under § 1988. A plaintiff must receive at least some relief on the merits of his claim before he can be said to “prevail. ” Respondent obtained neither a damages award, injunction, or declaratory judgment, nor a consent decree, settlement, or other relief without benefit of a formal judgment. Pp. 759-764. (a) A favorable judicial statement of law in the course of litigation that results in judgment against the plaintiff does not suffice to render him a 756 OCTOBER TERM, 1986 Syllabus 482 U. S. “prevailing party.” The Court of Appeals’ treatment of its initial constitutional holding as “a form of judicial relief”—presumably a form of declaratory judgment—was in error, since the court neither granted nor ordered relief of any kind. Even if respondent’s nonmonetary claims were not rendered moot by his release from prison, and it could be said that those claims were kept alive by his interest in expunging his misconduct conviction from his prison record, his counsel never took the steps necessary to have a declaratory judgment or expungement order properly entered. The argument that the Court of Appeals’ initial holding is a “vindication of rights” that is at least the equivalent of declaratory relief ignores the fact that a judicial decree is not the end of the judicial process but is rather the means of prompting some action (or cessation of action) by the defendant. Here, respondent obtained nothing from petitioners. Moreover, equating statements of law (even legal holdings en route to a final judgment for the defendant) with declaratory judgments has the practical effect of depriving the defendant of any valid defenses that a court might take into account in deciding whether to enter a declaratory judgment. Furthermore, the same considerations that influence courts to issue declaratory judgments may not enter into the decision whether to include statements of law in opinions. However, if they do, the court’s decision is not appealable in the same manner as its entry of a declaratory judgment. Pp. 759-763. (b) The alternative argument that a hearing is required to determine whether respondent’s suit prompted the Bureau to amend its regulations also fails. Even if respondent can demonstrate a clear causal link between his lawsuit and the amendment, and can “prevail” by having the State take action that his complaint did not in terms request, he did not obtain redress from that amendment since he had long since been released from prison at the time it was issued. Pp. 763-764. 780 F. 2d 367, reversed. Scalia, J., delivered the opinion of the Court, in which Rehnquist, C. J., and White, Powell, and O’Connor, JJ., joined. Marshall, J., filed a dissenting opinion, in which Brennan, Blackmun, and Stevens, JJ., joined, post, p. 764. Thomas G. Saylor, Jr., First Deputy Attorney General of Pennsylvania, argued the cause for petitioners. With him on the briefs were LeRoy S. Zimmerman, Attorney General, Allen C. Warshaw, Executive Deputy Attorney General, Andrew S. Gordon, Chief Deputy Attorney General, and Gregory R. Neuhauser, Senior Deputy Attorney General. HEWITT v. HELMS 757 755 Opinion of the Court Deputy Solicitor General Wallace argued the cause for the United States as amicus curiae urging reversal. With him on the brief were Solicitor General Fried, Assistant Attorney General Willard, Harriet S. Shapiro, and William Kanter. Robert H. Vesely argued the cause for respondent. With him on the brief was John M. Humphrey.* Justice Scalia delivered the opinion of the Court. This case presents the peculiar-sounding question whether a party who litigates to judgment and loses on all of his claims can nonetheless be a “prevailing party” for purposes of an award of attorney’s fees. Following a prison riot at the Pennsylvania State Correctional Institution at Huntingdon, inmate Aaron Helms was placed in administrative segregation, a form of restrictive custody, pending an investigation into his possible involvement in the disturbance. More than seven weeks later, a prison hearing committee, relying solely on an officer’s report of the testimony of an undisclosed informant, found Helms guilty of misconduct for striking a corrections officer during the riot. Helms was sentenced to six months of disciplinary restrictive confinement. While still incarcerated, Helms brought suit under 42 U. S. C. § 1983 against a number of prison officials, alleging that the lack of a prompt hearing on his misconduct charges and his conviction for misconduct on the basis of uncorroborated hearsay testimony violated his rights to due process. The prison officials asserted qualified immunity from suit and contested the constitutional claims on the merits. Before any decision was rendered, Helms was released from prison on parole. Nearly six months after Helms’ release, the District Court rendered summary judgment against him on his constitu *Benna Ruth Solomon, Joyce Holmes Benjamin, and Peter J. Kalis filed a brief for the National Governors’ Association et al. as amici curiae urging reversal. 758 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. tional claims without passing on the defendants’ assertions of immunity. The Court of Appeals for the Third Circuit reversed, finding that “Helms was denied due process unless he was afforded a hearing, within a reasonable time of his initial [segregative] confinement, to determine whether he represented the type of ‘risk’ warranting administrative detention,” Helms n. Hewitt, 655 F. 2d 487, 500 (1981) (Helms I), and that he “suffered a denial of due process by being convicted on a misconduct charge when the only evidence offered against him was a hearsay recital, by the charging officer, of an uncorroborated report of an unidentified informant.” Id., at 502. The District Court was instructed to enter summary judgment for Helms on the latter claim unless the defendants could establish an immunity defense. Before the proceedings on remand could take place, we granted certiorari to determine whether Helms’ administrative segregation violated the Due Process Clause. We concluded that the prison’s informal, nonadversarial procedures for determining the need for restrictive custody provided all the process that is due when prisoners are removed from the general prison population. Hewitt n. Helms, 459 U. S. 460 (1983). Certiorari was not sought on, and we did not decide, the question whether Helms’ misconduct conviction violated his constitutional rights. When the case was returned to the Court of Appeals, it therefore reaffirmed its instruction to the District Court to enter judgment for Helms on this claim unless the defendants established a defense of official immunity. Helms v. Hewitt, 712 F. 2d 48 (1983) (Helms II). In the District Court, Helms pursued only his claims for damages. The District Court granted summary judgment for all the defendants on the basis of qualified immunity, because the constitutional right at issue was not “clearly established,” Harlow v. Fitzgerald, 457 U. S. 800, 818 (1982), at the time of Helms’ misconduct hearing. See App. 22a-47a. Helms appealed, seeking both damages and expungement of his misconduct conviction. The defendants argued to the HEWITT v. HELMS 759 755 Opinion of the Court Court of Appeals that all claims for injunctive and declaratory relief had been waived by the failure to pursue them in the District Court, and in any event were moot because Helms was no longer in prison. While that appeal was pending, the Pennsylvania Bureau of Corrections revised its regulations to include for the first time procedures for the use of confidential-source information in inmate disciplinary proceedings. See BC-ADM 801 Administrative Directive: Inmate Disciplinary Procedures §V(F) (1984), App. 101a-102a (Directive 801). The District Court’s decision was affirmed without opinion. Helms n. Hewitt, 745 F. 2d 46 (1984) (Helms III). Helms then sought attorney’s fees under 42 U. S. C. §1988, which provides in relevant part: “In any action or proceeding to enforce a provision of [§ 1983], the court, in its discretion, may allow the prevailing party, other than the United States, a reasonable attorney’s fee as part of the costs.” The District Court denied the claim on the ground that Helms was not a “prevailing party”: the defendants’ official immunity precluded a damages award, Helms’ release from prison made his claims for injunctive relief moot, and he could not claim that his suit was a “catalyst” for the amendment of Directive 801 because he neither sought nor benefited from that action. App. to Pet. for Cert. 27a-39a. The Court of Appeals reversed, concluding that its prior holding that Helms’ constitutional rights were violated was “a form of judicial relief which serves to affirm the plaintiff’s assertion that the defendants’ actions were unconstitutional and which will serve as a standard of conduct to guide prison officials in the future.” 780 F. 2d 367, 370 (1986) (Helms IV). The court also directed the District Court to reconsider whether Helms’ suit was a “catalyst” for the amendment of Directive 801. We granted certiorari. 476 U. S. 1181 (1986). In order to be eligible for attorney’s fees under § 1988, a litigant must be a “prevailing party.” Whatever the outer boundaries of that term may be, Helms does not fit within 760 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. them. Respect for ordinary language requires that a plaintiff receive at least some relief on the merits of his claim before he can be said to prevail. See Hanrahan n. Hampton, 446 U. S. 754, 757 (1980). Helms obtained no relief. Because of the defendants’ official immunity he received no damages award. No injunction or declaratory judgment was entered in his favor. Nor did Helms obtain relief without benefit of a formal judgment—for example, through a consent decree or settlement. See Maher v. Gagne, 448 U. S. 122, 129 (1980). The most that he obtained was an interlocutory ruling that his complaint should not have been dismissed for failure to state a constitutional claim. That is not the stuff of which legal victories are made. Cf. Hanrahan, supra, at 758-759. The Court of Appeals treated its 1981 holding that Helms’ misconduct conviction was unconstitutional as “a form of judicial relief”—presumably (since nothing else is even conceivable) a form of declaratory judgment. It was not that. Helms I explicitly left it to the District Court “to determine the appropriateness and availability of the requested relief,” 655 F. 2d, at 503; the Court of Appeals granted no relief of its own, declaratory or otherwise. The petitioners contend that the court in fact could not have granted declaratory or injunctive relief at that point, since all of Helms’ nonmonetary claims were moot as a result of his release from prison. Even if that is not correct, and Helms’ interest in expungement of the misconduct conviction from his prison record was enough to keep those claims alive, the fact is that Helms’ counsel never took the steps necessary to have a declaratory judgment or expungement order properly entered. Consequently, Helms received no judicial relief. It is settled law, of course, that relief need not be judicially decreed in order to justify a fee award under § 1988. A lawsuit sometimes produces voluntary action by the defendant that affords the plaintiff all or some of the relief he sought through a judgment—e. g., a monetary settlement or a HEWITT u HELMS 761 755 Opinion of the Court change in conduct that redresses the plaintiff’s grievances. When that occurs, the plaintiff is deemed to have prevailed despite the absence of a formal judgment in his favor. See Maher, supra, at 129. The Court of Appeals held, and Helms argues here, that the statement of law in Helms I that Helms’ disciplinary proceeding was unconstitutional is a “vindication of . . . rights,” Brief for Respondent 19, that is at least the equivalent of declaratory relief, just as a monetary settlement is the informal equivalent of relief by way of damages. To suggest such an equivalency is to lose sight of the nature of the judicial process. In all civil litigation, the judicial decree is not the end but the means. At the end of the rainbow lies not a judgment, but some action (or cessation of action) by the defendant that the judgment produces—the payment of damages, or some specific performance, or the termination of some conduct. Redress is sought through the court, but from the defendant. This is no less true of a declaratory judgment suit than of any other action. The real value of the judicial pronouncement—what makes it a proper judicial resolution of a “case or controversy” rather than an advisory opinion—is in the settling of some dispute which affects the behavior of the defendant towards the plaintiff The “equivalency” doctrine is simply an acknowledgment of the primacy of the redress over the means by which it is obtained. If the defendant, under the pressure of the lawsuit, pays over a money claim before the judicial judgment is pronounced, the plaintiff has “prevailed” in his suit, because he has obtained the substance of what he sought. Likewise in a declaratory judgment action: if the defendant, under pressure of the lawsuit, alters his conduct (or threatened conduct) towards the plaintiff that was the basis for the suit, the plaintiff will have prevailed. That is the proper equivalent of a judicial judgment which would produce the same effect; a judicial statement that does not affect the relationship between the plaintiff and the defendant is not an equivalent. As a consequence of the present lawsuit, Helms obtained nothing 762 OCTOBER TERM, 1986 Opinion of the Court 482 U. S. from the defendants. The only “relief” he received was the moral satisfaction of knowing that a federal court concluded that his rights had been violated. The same moral satisfaction presumably results from any favorable statement of law in an otherwise unfavorable opinion. There would be no conceivable claim that the plaintiff had “prevailed,” for instance, if the District Court in this case had first decided the question of immunity, and the Court of Appeals affirmed in a published opinion which said: “The defendants are immune from suit for damages, and the claim for expungement is either moot or has been waived, but if not for that we would reverse because Helms’ constitutional rights were violated.” That is in essence what happened here, except that the Court of Appeals expressed its view on the constitutional rights before, rather than after, it had become apparent that the issue was irrelevant to the case. There is no warrant for having status as a “prevailing party” depend upon the essentially arbitrary order in which district courts or courts of appeals choose to address issues. Besides the incompatibility in principle, there is a very practical objection to equating statements of law (even legal holdings en route to a final judgment for the defendant) with declaratory judgments: The equation deprives the defendant of valid defenses to a declaratory judgment to which he is entitled. Imagine that following Helms I, Helms’ counsel, armed with the holding that his client’s constitutional rights had been violated, pressed the District Court for entry of a declaratory judgment. The defendants would then have had the opportunity to contest its entry not only on the ground that the case was moot but also on equitable grounds. The fact that a court can enter a declaratory judgment does not mean that it should. See 28 U. S. C. § 2201 (a court “may declare the rights and other legal relations of any interested party seeking such declaration”) (emphasis added); Public Affairs Associates, Inc. v. Rickover, 369 U. S. Ill, 112 (1962); Eccles n. Peoples Bank of Lakewood, 333 U. S. 426, HEWITT v. HELMS 763 755 Opinion of the Court 431 (1948). If, for example, Helms I had unambiguously involved only a claim for damages, the requested declaratory judgment would not definitively “settle the controversy between the parties,” 10A C. Wright, A. Miller, & M. Kane, Federal Practice and Procedure §2759, p. 648 (2d ed. 1983), because immunity might still preclude liability. See generally E. Borchard, Declaratory Judgments 299 (2d ed. 1941). If the only effect of a declaratory judgment in those circumstances would be to provide a possible predicate for a fee award against defendants who may ultimately be found immune, and thus to undermine the doctrine of official immunity, it is conceivable that the court might take that into account in deciding whether to enter a judgment. The same considerations may not enter into the decision whether to include statements of law in opinions—or if they do, the court’s decision is not appealable in the same manner as its entry of a declaratory judgment. We conclude that a favorable judicial statement of law in the course of litigation that results in judgment against the plaintiff does not suffice to render him a “prevailing party.” Any other result strains both the statutory language and common sense. The Court of Appeals held in the alternative, and Helms argues in the alternative here, that a hearing is needed to determine whether Helms’ lawsuit prompted the Pennsylvania Bureau of Corrections to amend its regulations in 1984 to provide standards for the use of informant testimony at disciplinary hearings. We need not decide the circumstances, if any, under which this “catalyst” theory could justify a fee award under § 1988, because even if Helms can demonstrate a clear causal link between his lawsuit and the State’s amendment of its regulations, and can “prevail” by having the State take action that his complaint did not in terms request, he did not and could not get redress from promulgation of the informant-testimony regulations. When Directive 801 was amended, Helms had long since been released from prison. 764 OCTOBER TERM, 1986 Marshall, J., dissenting 482 U. S. Although he has subsequently been returned to prison, and is presumably now benefiting from the new procedures (to the extent that they influence prison administration even when not directly being applied), that fortuity can hardly render him, retroactively, a “prevailing party” in this lawsuit, even though he was not such when the final judgment was entered. For the reasons stated, the judgment of the court of appeals is Reversed. Justice Marshall, with whom Justice Brennan, Justice Blackmun, and Justice Stevens join, dissenting. The Court makes a number of sweeping statements in its opinion, most of which are of no help in resolving the present case. In my view, the application of settled law to the facts of this case, tangled as they are, leads to conclusions other than those reached by the Court. I The Court’s account of the history of this litigation is complete, but a summary may be helpful. Respondent originally claimed in the District Court both procedural and substantive violations of due process in connection with his prison misconduct conviction, and raised in addition a pendent state claim. He sought declaratory and injunctive relief, damages, and the expungement of his prison disciplinary record. App. 19a-21a. Petitioners alleged immunity defenses, as well as contesting the merits of the federal and state claims. The District Court initially dismissed both the procedural and substantive due process causes of action. The Court of Appeals reversed as to both claims. Helms n. Hewitt, 655 F. 2d 487 (CA3 1981). We granted certiorari only as to procedural due process and reversed, reinstating the District Court’s grant of summary judgment for petitioners. Hewitt v. Helms, 459 U. S. 460 (1983). On remand from this Court, the Court of Appeals noted that its substantive due process HEWITT v. HELMS 765 755 Marshall, J., dissenting holding concerning the use of anonymous informant evidence was unaffected by our decision, and remanded for entry of judgment in favor of respondent unless petitioners were immune. Helms v. Hewitt, 712 F. 2d 48, 49 (1983); see 655 F. 2d, at 502-503 (“[O]n remand, if the defendants do not establish official immunity. . . the district court should enter summary judgment for Helms”). The District Court, on remand from the Court of Appeals, concluded that petitioners were immune from the payment of damages because the law concerning the use of anonymous informant evidence in prison disciplinary proceedings “was not so clear and well established” at the time of respondent’s disciplinary proceeding as to overcome petitioners’ qualified official immunity. App. 47a. Respondent appealed from this second order granting summary judgment for petitioners. During the pendency of this appeal the Commonwealth of Pennsylvania issued Administrative Directive 801, App. 85a-116a, which incorporated policies with respect to the use of anonymous informant evidence in prison misconduct proceedings consistent with the earlier holding of the Court of Appeals. Id., at 101a-102a. The Court of Appeals subsequently affirmed the District Court’s judgment in a summary order. Helms v. Hewitt, 745 F. 2d 46 (CA3 1984). Respondent then moved for fees in the District Court pursuant to 42 U. S. C. §1988. II Some aspects of the procedural development of this case may be difficult to fathom, but at the very least the case does not present, as the Court declares, a fee application by “a party who litigates to judgment and loses on all of his claims.” Ante, at 757. Respondent’s complaint alleged two federal causes of action. We held that respondent had not stated a viable cause of action for violation of his right to procedural due process. The final word on the substantive due process claim, however, was spoken by the Court of Appeals, which directed the District Court to enter summary judg 766 OCTOBER TERM, 1986 Marshall, J., dissenting 482 U. S. ment for respondent on that claim unless the petitioners were immune. The Court devotes much of its opinion to demonstrating on theoretical grounds that this statement by the Court of Appeals was not a declaratory judgment. I think that effort unnecessary; it is plain from the language of the first opinion of the Court of Appeals that it was not entering judgment for respondent. Instead, consistent with the ordinary practice of appellate courts, it simply found respondent’s cause of action good as a matter of law, and remanded with instructions to enter judgment for respondent insofar as such a judgment was not incompatible with petitioners’ immunity, if any. 655 F. 2d, at 502-503. The District Court then found that petitioners were entitled to qualified immunity. This precluded any remedy in damages against petitioners, but by no means prevented the ordering of delaratory or injunctive relief or a grant of attorney’s fees. See Pulliam v. Allen, 466 U. S. 522, 543-544 (1984). Respondent’s complaint sought relief in the form of a declaratory judgment and an injunction expunging his prison disciplinary record.1 Under the Court of Appeals’ remand order, the District Court could, and probably should, have entered judgment granting the requested declaratory and injunctive relief. Instead, the District Court 1 Petitioners have taken the position that these requests for declaratory and injunctive relief were somehow mooted by respondent’s release on parole in the early stages of this litigation. Brief for Petitioners 24. Indeed, petitioners represent to the Court that the District Court found that “[respondent’s] claim for injunctive relief had been rendered moot by his release from prison in 1980.” Id., at 10. This statement is flagrantly inaccurate. The District Court in fact held that “plaintiff did not seek any relief which became mooted during the controversy.” App. to Pet. for Cert 38a. Petitioners have offered no authority, nor can they, for the remarkable proposition that the request for expungement of respondent’s record became moot upon his parole. Nor, since the expungement would have depended upon the finding that respondent’s due process rights were violated, have they explained how the request for declaratory relief supposedly became moot. HEWITT v. HELMS 767 755 Marshall, J., dissenting first took up the question of immunity, and upon finding qualified immunity precipitately issued an order closing the case. App. 48a. No order was entered disposing of respondent’s pending claims for equitable relief.2 Respondent contends, and the Court of Appeals agreed, that the issuance of Administrative Directive 801 during the pendency of the subsequent appeal might be the sort of informal relief justifying a fee award, if the Commonwealth’s change of policy was “catalyzed” by respondent’s lawsuit. There is no dispute that informal relief may be sufficient to support a fee award under § 1988. See Maher n. Gagne, 448 U. S. 122, 129-130 (1980). The Court wisely leaves for another day any discussion of the general circumstances under which action “catalyzed” by a lawsuit may be characterized as informal relief for purposes of § 1988. Ante, at 763. But the Court errs in holding that Administrative Directive 801 cannot constitute relief, even under a “catalyst” theory, because respondent derived no benefit from it. The Directive does not, of course, provide an informal equivalent to respondent’s request for injunctive relief, because it did not effect an expungement of his disciplinary record. But the Directive may be, in substance, functionally equivalent to respondent’s requested declaratory relief. As the Court correctly states, in a declaratory judgment action “if the defendant, under pressure of the lawsuit, alters his conduct (or threatened 2 The record does not contain the briefs, if any, filed with the Court of Appeals on respondent’s appeal from the District Court’s order. Accordingly it is not clear whether respondent challenged only the District Court’s holding on immunity, or also its failure to award equitable relief. Nor is it clear, since the District Court’s order did not dispose of all respondent’s outstanding claims, whether respondent might not even now move in the District Court for the equitable relief requested in the complaint. The issuance of such equitable relief would, of course, support a fee award under 42 U. S. C. § 1988. The remand ordered by the Court of Appeals in the judgment presently before us would give the District Court an opportunity to rectify the substantial confusion engendered by its earlier proceedings. 768 OCTOBER TERM, 1986 Marshall, J., dissenting 482 U. S. conduct) towards the plaintiff . . . the plaintiff will have prevailed.” Ante, at 761. The Court observes that respondent is once again an inmate of the Commonwealth’s prisons. Ante, at 764. The behavior of the Commonwealth’s officials toward respondent is as effectively constrained by Directive 801 today as it would have been by a formal declaratory judgment.3 In sum, respondent’s claim for fees is based upon the following premises: that the Court of Appeals held his civil rights cause of action good as a matter of law; that at the time of the District Court’s judgment on the issue of immunity, respondent had outstanding meritorious claims for equitable relief; that the judgment as to petitioners’ immunity did not foreclose the granting of equitable relief or an award of attorney’s fees; and that the issuance of Directive 801 during the pendency of litigation provided respondent, by the voluntary action of petitioners and those in privity with them, informal relief substantially equivalent to the relief sought in respondent’s prayer for a declaratory judgment. None of these propositions is subject to serious dispute, and none is rejected by the Court today. The question remains, of course, whether there is any causal connection between the litigation instituted by respondent and the Commonwealth’s promulgation of Directive 801. This is an issue of fact which can only be resolved in the District Court. Should the District Court find that the promulgation of Directive 801 was not “catalyzed” by this litigation, then the error of respondent’s counsel in failing to move in the District Court for formal entry of a declaratory judgment, to which respondent was clearly en- 8 The Court characterizes respondent’s renewed incarceration as a “fortuity,” evidently implying that it has no relevance to this case. But the record does not disclose whether respondent was imprisoned after parole revocation proceedings, or instead as the result of a subsequent criminal conviction. If respondent’s parole was revoked, then it is his temporary release during the course of the litigation, rather than his reincarceration, which is a “fortuity” in determining respondent’s entitlement to attorney’s fees. HEWITT v. HELMS 769 755 Marshall, J., dissenting titled under the Court of Appeals’ two remand orders, would probably foreclose any fee award. But any such conclusion must await further factfinding. Ill The disposition of this chaotic case depends upon the procedural accidents of extended litigation conducted with less than exemplary precision by the parties and the District Court. While the Court sensibly declines to establish any broadly applicable doctrine upon a basis as unreliable as the present record, it nonetheless indulges in a theoretical exposition which varies substantially from the few ascertainable facts. If further review of this litigation was a prudent exercise of our certiorari jurisdiction, which I doubt, it should have occurred after the necessary facts had been found, and the general fog of confusion dispelled, by the District Court. I would affirm the judgment of the Court of Appeals insofar as it remanded to the District Court for factual findings on respondent’s “catalyst” theory. Reporter’s Note The next page is purposely numbered 901. The numbers between 769 and 901 were intentionally omitted, in order to make it possible to publish the orders with permanent page numbers, thus making the official citations available upon publication of the preliminary prints of the United States Reports. ORDERS FROM JUNE 1 THROUGH JUNE 18, 1987 June 1, 1987 Appeals Dismissed No. 86-1652. Liquid Transporters, Inc. v. Revenue Cabinet of Kentucky. Appeal from Ct. App. Ky. dismissed for want of substantial federal question. Reported below: 721 S. W. 2d 722. No. 86-1666. Cross v. Ohio. Appeal from Ct. App. Ohio, Greene County, dismissed for want of substantial federal question. Reported below: 31 Ohio App. 3d 28, 508 N. E. 2d 172. Certiorari Granted—Vacated and Remanded. (See also No. 86-298, ante, at 68, n. 2.) No. 86-908. United Steelworkers of America, AFL-CIO-CLC v. Rawson, Individually and as Guardian ad Litem for Rawson, et al. Sup. Ct. Idaho. Certiorari granted, judgment vacated, and case remanded for further consideration in light of Electrical Workers v. Hechler, 481 U. S. 851 (1987). Reported below: 111 Idaho 630, 726 P. 2d 742. No. 86-5533. Chirino v. Jordan Marsh Co. C. A. 11th Cir. Motion of petitioner for leave to proceed in forma pauperis and certiorari granted. Judgment vacated and case remanded for further consideration in light of Shaare Tefila Congregation v. Cobb, 481 U. S. 615 (1987), and Saint Francis College v. Al-Khazraji, 481 U. S. 604 (1987). Reported below: 795 F. 2d 87. Certiorari Granted—Reversed. (See No. 86-1053, ante, p. 117.) Vacated and Remanded After Certiorari Granted No. 86-1408. Haynie et al. v. Ross Gear Division of TRW, Inc., et al. C. A. 6th Cir. [Certiorari granted, 481 U. S. 1003.] On the basis of the representations made in the parties’ joint stipulation of dismissal and suggestion of mootness, the judgment of the United States Court of Appeals for the Sixth Circuit is vacated and the case is remanded to that court with a direction to dismiss. See United States v. Munsing wear, Inc., 340 901 902 OCTOBER TERM, 1986 June 1, 1987 482 U. S. U. S. 36 (1950). Justice Marshall took no part in the consideration or decision of this case. Certiorari Dismissed No. 86-6452. Patterson v. South Carolina. Sup. Ct. S. C. Certiorari dismissed for want of jurisdiction. Reported below: 290 S. C. 523, 351 S. E. 2d 853. Miscellaneous Orders No.------------. Granviel v. Texas. Motion to direct the Clerk to file the petition for writ of certiorari without an affidavit of indigency executed by the petitioner granted. No. A-778 (86-1646). Texas v. Williams. Application to continue the stay of mandate of the Court of Criminal Appeals of Texas, presented to Justice White, and by him referred to the Court, is granted pending final disposition by this Court of the petition for writ of certiorari. No. A-807. Levy v. United States. Application for release pending appeal, addressed to Justice Brennan and referred to the Court, denied. No. A-811. Millsap v. Federal National Mortgage Association et al. C. A. 7th Cir. Application for stay, addressed to Justice Brennan and referred to the Court, denied. No. A-846. Mississippi Power & Light Co. v. Mississippi et AL. Application for stay of judgment of the Supreme Court of Mississippi, presented to Justice White, and by him referred to the Court, granted pending the timely filing and disposition of the appeal by this Court. This order is further conditioned upon the posting of a good and sufficient bond, in manner and amount to be determined by the Supreme Court of Mississippi. No. A-847 (86-1879). National Treasury Employees Union et al. v. Von Raab, Commissioner, United States Customs Service. C. A. 5th Cir. Application for stay, presented to Justice White, and by him referred to the Court, denied. Justice Brennan would grant the application. No. D-608. In re Disbarment of Share. Disbarment entered. [For earlier order herein, see 480 U. S. 902.] No. D-611. In re Disbarment of Brickle. Disbarment entered. [For earlier order herein, see 480 U. S. 902.] ORDERS 903 482 U. S. June 1, 1987 No. D-623. In re Disbarment of Connell. Disbarment entered. [For earlier order herein, see 480 U. S. 944.] No. D-625. In re Disbarment of Ernst. Thomas J. Ernst, of Clayton, Mo., having requested to resign as a member of the Bar of this Court, it is ordered that his name be stricken from the roll of attorneys admitted to practice before the Bar of this Court. The rule to show cause, heretofore issued on April 6, 1987 [481 U. S. 1002], is hereby discharged. No. D-627. In re Disbarment of Gerns. Disbarment entered. [For earlier order herein, see 481 U. S. 1002.] No. D-635. In re Disbarment of Fleischer. It is ordered that Edward Leo Fleischer, of Morganville, N. J., be suspended from the practice of law in this Court and that a rule issue, returnable within 40 days, requiring him to show cause why he should not be disbarred from the practice of law in this Court. No. D-636. In re Disbarment of Brown. It is ordered that Arnold E. Brown, of Englewood, N. J., be suspended from the practice of law in this Court and that a rule issue, returnable within 40 days, requiring him to show cause why he should not be disbarred from the practice of law in this Court. No. D-637. In re Disbarment of Hoagland. It is ordered that Robert D. Hoagland, of Charlotte, N. C., be suspended from the practice of law in this Court and that a rule issue, returnable within 40 days, requiring him to show cause why he should not be disbarred from the practice of law in this Court. No. D-638. In re Disbarment of Haeberle. It is ordered that W. Gene Haeberle, of Camden, N. J., be suspended from the practice of law in this Court and that a rule issue, returnable within 40 days, requiring him to show cause why he should not be disbarred from the practice of law in this Court. No. D-639. In re Disbarment of Goldberg. It is ordered that Gerald Mark Goldberg, of Rockaway, N. J., be suspended from the practice of law in this Court and that a rule issue, returnable within 40 days, requiring him to show cause why he should not be disbarred from the practice of law in this Court. No. D-640. In re Disbarment of Oxfeld. It is ordered that Emil Oxfeld, of South Orange, N. J., be suspended from the practice of law in this Court and that a rule issue, returnable 904 OCTOBER TERM, 1986 June 1, 1987 482 U. S. within 40 days, requiring him to show cause why he should not be disbarred from the practice of law in this Court. No. 85-2079. Laborers Health and Welfare Trust Fund for Northern California et al. v. Advanced Lightweight Concrete Co., Inc. C. A. 9th Cir. [Certiorari granted, 479 U. S. 1083.] Motion of the Solicitor General for leave to participate in oral argument as amicus curiae and for divided argument granted. No. 86-279. Basic Inc. et al. v. Levinson et al. C. A. 6th Cir. [Certiorari granted, 479 U. S. 1083.] Motion of the Solicitor General for leave to participate in oral argument as amicus curiae and for divided argument denied. The Chief Justice took no part in the consideration or decision of this motion. No. 86-492. Boyle, Personal Representative of the Heirs and Estate of Boyle v. United Technologies Corp. C. A. 4th Cir. [Certiorari granted, 479 U. S. 1029.] Motion of Bell Helicopter Textron Inc. for leave to file a brief as amicus curiae granted. Justice Powell took no part in the consideration or decision of this motion. No. 86-890. Deakins et al. v. Monaghan et al. C. A. 3d Cir. [Certiorari granted, 479 U. S. 1063.] Motion of American Civil Liberties Union Foundation et al. for leave to file a brief as amici curiae granted. No. 86-1108. Vermont v. Cox. Sup. Ct. Vt. [Certiorari granted, 479 U. S. 1083.] Motion of the Solicitor General for leave to participate in oral argument as amicus curiae and for divided argument granted. No. 86-1593. Shell Oil Co. v. Department of Revenue of Florida. Appeal from Sup. Ct. Fla. The Solicitor General is invited to file a brief in this case expressing the views of the United States. Certiorari Granted No. 86-637. Communications Workers of America et al. v. Beck et al. C. A. 4th Cir. Certiorari granted. In addition to the questions presented by the petition, the parties are directed to brief and argue the applicability of San Diego Building Trades Council v. Garmon, 359 U. S. 236 (1959). Reported below: 800 F. 2d 1280. ORDERS 905 482 U. S. June 1, 1987 No. 86-6284. Satterwhite v. Texas. Ct. Crim. App. Tex. Motion of petitioner for leave to proceed in forma pauperis and certiorari granted limited to Question 1 presented by the petition. Reported below: 726 S. W. 2d 81. Certiorari Denied. (See also No. 86-1054, ante, p. 117.) No. 85-455. Polo Fashions, Inc. v. Stock Buyers International, Inc., et al. C. A. 6th Cir. Certiorari denied. Reported below: 760 F. 2d 698. No. 86-940. Pacific First Federal Savings Bank et al. v. Rembold et al. C. A. 9th Cir. Certiorari denied. Reported below: 798 F. 2d 1307. No. 86-1304. Caifano v. United States. C. A. 11th Cir. Certiorari denied. Reported below: 807 F. 2d 997. No. 86-1440. Monsour Medical Center v. Bowen, Secretary of Health and Human Services, et al. C. A. 3d Cir. Certiorari denied. Reported below: 806 F. 2d 1185. No. 86-1464. Corbett v. Tisch, Postmaster General of the United States, et al. C. A. 9th Cir. Certiorari denied. Reported below: 804 F. 2d 145. No. 86-1616. Plumbers & Steamfitters Union Local 598 et al. v. Washington Public Power Supply System. Ct. App. Wash. Certiorari denied. Reported below: 44 Wash. App. 906, 724 P. 2d 1030. No. 86-1618. Harrison School District No. 1 et al. v. Lewis. C. A. 8th Cir. Certiorari denied. Reported below: 805 F. 2d 310. No. 86-1622. United States v. Fuccillo. C. A. 1st Cir. Certiorari denied. Reported below: 808 F. 2d 173. No. 86-1630. Mirabole v. Florida Bar. Sup. Ct. Fla. Certiorari denied. Reported below: 498 So. 2d 428. No. 86-1631. Margoles v. Johns et al. C. A. 7th Cir. Certiorari denied. Reported below: 798 F. 2d 1069. No. 86-1634. Butcher v. City of Detroit. Ct. App. Mich. Certiorari denied. Reported below: 156 Mich. App. 165, 401 N. W. 2d 260. No. 86-1635. Lyne v. Coxon et al. Sup. Ct. N. M. Certiorari denied. Reported below: 105 N. M. 57, 728 P. 2d 467. 906 OCTOBER TERM, 1986 June 1, 1987 482 U. S. No. 86-1639. Vinson v. Ford Motor Co. C. A. 6th Cir. Certiorari denied. Reported below: 806 F. 2d 686. No. 86-1640. Pace Resources, Inc. v. Shrewsbury Township et al. C. A. 3d Cir. Certiorari denied. Reported below: 808 F. 2d 1023. No. 86-1644. Yamhill Yamaha, Inc. v. United States District Court for the District of Oregon (Yamaha Motor Corp, et al., Real Parties in Interest). C. A. 9th Cir. Certiorari denied. No. 86-1648. West et al. v. Multibanco Comermex, S. A., et al. C. A. 9th Cir. Certiorari denied. Reported below: 807 F. 2d 820. No. 86-1651. Shedrick v. Donnelly et al. C. A. 5th Cir. Certiorari denied. Reported below: 811 F. 2d 601. No. 86-1653. Bourke v. Schuman. App. Dept., Super. Ct. Cal., Alameda County. Certiorari denied. No. 86-1668. Haney v. Louisiana Training Institute. C. A. 5th Cir. Certiorari denied. No. 86-1671. Coker v. Gielow, Chairman, Railroad Retirement Board, et al. C. A. 6th Cir. Certiorari denied. Reported below: 806 F. 2d 689. No. 86-1691. Rayburn v. United States. C. A. 6th Cir. Certiorari denied. Reported below: 813 F. 2d 105. No. 86-1716. Moore v. State Bank of Burden et al. Sup. Ct. Kan. Certiorari denied. Reported below: 240 Kan. 382, 729 P. 2d 1205. No. 86-1737. Flaugh v. Michigan. Ct. App. Mich. Certiorari denied. No. 86-6289. Minotti v. Lensink, Commissioner, Connecticut State Department of Mental Retardation. C. A. 2d Cir. Certiorari denied. Reported below: 798 F. 2d 607. No. 86-6425. Watson v. Hensiek. Ct. App. Mo., Eastern Dist. Certiorari denied. Reported below: 716 S. W. 2d 284. No. 86-6453. Williams v. United States. C. A. 1st Cir. Certiorari denied. Reported below: 809 F. 2d 75. ORDERS 907 482 U. S. June 1, 1987 No. 86-6466. Beasley v. United States. Ct. App. D. C. Certiorari denied. Reported below: 512 A. 2d 1007. No. 86-6482. Martir v. United States. C. A. 2d Cir. Certiorari denied. Reported below: 810 F. 2d 1161. No. 86-6669. Sweetman v. Township of Pennsauken, New Jersey. Super. Ct. N. J., App. Div. Certiorari denied. No. 86-6676. Annonson v. Grover, State Superintendent of Public Instruction. Sup. Ct. Wis. Certiorari denied. Reported below: 133 Wis. 2d 482, 400 N. W. 2d 470. No. 86-6682. Standifer v. New York. App. Div., Sup. Ct. N. Y., 4th Jud. Dept. Certiorari denied. Reported below: 124 App. Div. 2d 1077, 508 N. Y. S. 2d 130. No. 86-6689. Washington v. Scully, Superintendent, Green Haven Correctional Facility. C. A. 2d Cir. Certiorari denied. Reported below: 812 F. 2d 713. No. 86-6690. Phillips v. Illinois. App. Ct. Ill., 3d Dist. Certiorari denied. Reported below: 147 Ill. App. 3d 1163, 512 N. E. 2d 139. No. 86-6691. Rhoden v. Florida et al. C. A. 11th Cir. Certiorari denied. No. 86-6693. Soto v. LeFevre, Superintendent, Clinton Correctional Facility. C. A. 2d Cir. Certiorari denied. Reported below: 812 F. 2d 713. No. 86-6694. Morris v. Kemp, Warden. C. A. 11th Cir. Certiorari denied. Reported below: 809 F. 2d 1499. No. 86-6714. Monnin v. Consolidated Rail Corporation. C. A. 6th Cir. Certiorari denied. Reported below: 812 F. 2d 1407. No. 86-6715. Owens v. Adams et al. App. Ct. Ill., 1st Dist. Certiorari denied. Reported below: 144 Ill. App. 3d 1173, 510 N. E. 2d 1330. No. 86-6720. Lewis v. Illinois. App. Ct. Ill., 1st Dist. Certiorari denied. Reported below: 147 Ill. App. 3d 249, 498 N. E. 2d 1169. No. 86-6744. Gordon v. Lynaugh, Director, Texas Department of Corrections. C. A. 5th Cir. Certiorari denied. 908 OCTOBER TERM, 1986 482 U. S. June 1, 1987 No. 86-6753. Attwell v. Metropolitan Atlanta Rapid Transit Authority et al. C. A. 11th Cir. Certiorari denied. Reported below: 811 F. 2d 610. No. 86-6754. Lewis v. Lynaugh, Director, Texas Department of Corrections. C. A. 5th Cir. Certiorari denied. No. 86-6764. Thomas v. Edwards, Warden, et al. C. A. 3d Cir. Certiorari denied. Reported below: 808 F. 2d 1518. No. 86-6774. Raffoul v. United States. C. A. 3d Cir. Certiorari denied. Reported below: 815 F. 2d 696. No. 86-6779. Gomez v. United States. C. A. 10th Cir. Certiorari denied. Reported below: 810 F. 2d 947. No. 86-6785. Marquez v. United States. C. A. 2d Cir. Certiorari denied. Reported below: 816 F. 2d 670. No. 86-6787. Rodrigues v. United States. C. A. 5th Cir. Certiorari denied. Reported below: 810 F. 2d 1338. No. 86-6790. Loney v. United States. C. A. 4th Cir. Certiorari denied. Reported below: 811 F. 2d 1506. No. 86-6801. Woodcock v. United States. C. A. 5th Cir. Certiorari denied. Reported below: 812 F. 2d 965. No. 86-6804. Brown-Bey v. United States Marshal. C. A. 8th Cir. Certiorari denied. Reported below: 815 F. 2d 711. No. 86-6817. Portal v. United States. C. A. 11th Cir. Certiorari denied. Reported below: 755 F. 2d 830. No. 86-6818. Vaden v. Village of Maywood, Illinois, et al. C. A. 7th Cir. Certiorari denied. Reported below: 809 F. 2d 361. No. 86-6819. Roque-Pino v. United States. C. A. 11th Cir. Certiorari denied. Reported below: 816 F. 2d 686. No. 86-6820. Bailey v. United States. C. A. 4th Cir. Certiorari denied. Reported below: 817 F. 2d 102. No. 86-6822. Bascos v. United States. C. A. 9th Cir. Certiorari denied. Reported below: 812 F. 2d 715. ORDERS 909 482 U. S. June 1, 1987 No. 86-6824. Klein v. United States. C. A. 6th Cir. Certiorari denied. Reported below: 815 F. 2d 706. No. 86-6833. Rohrbach v. United States. C. A. 8th Cir. Certiorari denied. Reported below: 813 F. 2d 142. No. 86-6837. Watkins v. United States. C. A. 11th Cir. Certiorari denied. Reported below: 811 F. 2d 1408. No. 86-6838. Rawwad v. United States. C. A. 1st Cir. Certiorari denied. Reported below: 807 F. 2d 294. No. 86-1519. Akzo N. V. et al. v. United States International Trade Commission et al. C. A. Fed. Cir. Motion of Kingdom of the Netherlands for leave to file a brief as amicus curiae granted. Certiorari denied. Reported below: 808 F. 2d 1471. No. 86-6638. Harris v. Indiana. Sup. Ct. Ind.; No. 86-6677. Gaskins v. South Carolina. Ct. Common Pleas of Richland County, S. C.; No. 86-6700. Adkins v. Tennessee. Sup. Ct. Tenn.; and No. 86-6750. Coleman v. Brown, Warden, et al. C. A. 10th Cir. Certiorari denied. Reported below: No. 86-6638, 499 N. E. 2d 723; No. 86-6700, 725 S. W. 2d 660; No. 86-6750, 802 F. 2d 1227. Justice Brennan and Justice Marshall, dissenting. Adhering to our views that the death penalty is in all circumstances cruel and unusual punishment prohibited by the Eighth and Fourteenth Amendments, Gregg v. Georgia, 428 U. S. 153, 227, 231 (1976), we would grant certiorari and vacate the death sentences in these cases. No. 86-6706. Heislup et al. v. Town of Colonial Beach, Virginia, et al. C. A. 4th Cir. Certiorari denied. The Chief Justice took no part in the consideration or decision of this petition. Reported below: 813 F. 2d 401. Rehearing Denied No. 86-1155. International Primate Protection League et al. v. Institute for Behavioral Research, Inc., et al., 481 U. S. 1004; No. 86-6312. Hymen v. Merit Systems Protection Board, 481 U. S. 1019; 910 OCTOBER TERM, 1986 482 U. S. June 1, 5, 8, 1987 No. 86-6401. U. S. 1020; No. 86-6417. 481 U. S. 1020; No. 86-6450. No. 86-6607. Skurdal v. City of Billings, Montana, 481 Lay v. Horan, Commonwealth Attorney, Marshall v. Bauer, 481 U. S. 1021; and In re Fixel, 481 U. S. 1012. Petitions for rehearing denied. June 5, 1987 Miscellaneous Order No. A-884. Berry v. Phelps, Secretary, Louisiana Department of Corrections, et al. Application for stay of execution of sentence of death, presented to Justice White, and by him referred to the Court, denied. Justice Brennan and Justice Marshall, dissenting. Adhering to our views that the death penalty is in all circumstances cruel and unusual punishment prohibited by the Eighth and Fourteenth Amendments, Gregg v. Georgia, 428 U. S. 153, 227, 231 (1976), we would grant the application for stay of execution in order to give the applicant time to file a petition for writ of certiorari and would grant the petition and vacate the death sentence in this case. June 8, 1987 Affirmed on Appeal No. 86-1559. Burrell et al. v. Allain, Governor of Mississippi, et al. Affirmed on appeal from D. C. S. D. Miss. Appeals Dismissed No. 86-1096. Griffith v. Illinois. Appeal from App. Ct. Ill., 5th Dist., dismissed for want of jurisdiction. Treating the papers whereon the appeal was taken as a petition for writ of certiorari, certiorari denied. Reported below: 143 Ill. App. 3d 683, 493 N. E. 2d 413. Justice White, dissenting. I agree with the majority that this case is not a proper appeal but, for the reason stated in my dissent from denial of certiorari in Nyflot n. Minnesota Commissioner of Public Safety, 474 U. S. 1027 (1985), I would grant certiorari. ORDERS 911 482 U. S. June 8, 1987 No. 86-1502. Spring Realty Co. et al. v. New York City Loft Board et al. Appeal from Ct. App. N. Y. dismissed for want of substantial federal question. Reported below: 69 N. Y. 2d 657, 503 N. E. 2d 1367. No. 86-6748. Riley v. City of Junction City, Kansas. Appeal from Sup. Ct. Kan. dismissed for want of properly presented federal question. Reported below: 240 Kan. 614, 731 P. 2d 310. No. 86-6758. Hatton v. Minnesota. Appeal from Sup. Ct. Minn, dismissed for want of jurisdiction. Treating the papers whereon the appeal was taken as a petition for writ of certiorari, certiorari denied. Miscellaneous Orders No. A-619. Effinger v. Greene et al. C. A. 11th Cir. Motion to direct the Clerk to file the petition for writ of certiorari out of time denied. No. D-604. In re Disbarment of Kornowski. Disbarment entered. [For earlier order herein, see 479 U. S. 1051.] No. D-641. In re Disbarment of Wechsler. It is ordered that Benjamin B. Wechsler II, of Pittsburgh, Pa., be suspended from the practice of law in this Court and that a rule issue, returnable within 40 days, requiring him to show cause why he should not be disbarred from the practice of law in this Court. No. D-642. In re Disbarment of Palomo. It is ordered that Raul Palomo, Jr., of Los Angeles, Cal., be suspended from the practice of law in this Court and that a rule issue, returnable within 40 days, requiring him to show cause why he should not be disbarred from the practice of law in this Court. No. 86-327. Mullins Coal Co., Inc. of Virginia, et al. v. Director, Office of Workers’ Compensation Programs, United States Department of Labor, et al. C. A. 4th Cir. [Certiorari granted, 479 U. S. 1029.] Motion of United Mine Workers of America for leave to participate in oral argument as amicus curiae, for additional time for oral argument, and for divided argument denied. No. 86-492. Boyle, Personal Representative of the Heirs and Estate of Boyle v. United Technologies Corp. 912 OCTOBER TERM, 1986 June 8, 1987 482 U. S. C. A. 4th Cir. [Certiorari granted, 479 U. S. 1029.] Motions of Chamber of Commerce of the United States, UNR Industries, Inc., Defense Research Institute, Inc., and Product Liability Advisory Council, Inc., et al. for leave to file briefs as amici curiae granted. Justice Powell took no part in the consideration or decision of these motions. No. 86-728. Honig, California Superintendent of Public Instruction v. Doe et al. C. A. 9th Cir. [Certiorari granted, 479 U. S. 1084.] Motion of respondents for leave to proceed further herein in forma pauperis granted. No. 86-836. Hazelwood School District et al. v. Kuhl-meier et al. C. A. 8th Cir. [Certiorari granted, 479 Ü. S. 1053.] Motions of People for the American Way, American Society of Newspaper Editors et al., American Civil Liberties Union et al., and NOW Legal Defense & Educational Fund et al. for leave to file briefs as amici curiae granted. No. 86-890. Deakins et al. v. Monaghan et al. C. A. 3d Cir. [Certiorari granted, 479 U. S. 1063.] Respondents’ suggestion of mootness rejected. No. 86-1415. Marino et al. v. Ortiz et al.; and Costello et al. v. New York City Police Department et al. C. A. 2d Cir. The order entered May 18, 1987 [481 U. S. 1047], granting the petition for writ of certiorari is amended to read as follows: Certiorari granted limited to Questions 2 and 3 presented by the petition. No. 86-1879. National Treasury Employees Union et al. v. Von Raab, Commissioner, United States Customs Service. C. A. 5th Cir. Motion of petitioners to expedite consideration of the petition for writ of certiorari denied. No. 86-6717. In re Tyler et al.; No. 86-6742. In re Kennedy; and No. 86-6772. In re Roy. Petitions for writs of mandamus denied. Certiorari Granted No. 85-1910. Business Electronics Corp. v. Sharp Electronics Corp. C. A. 5th Cir. Certiorari granted. Reported below: 780 F. 2d 1212. ORDERS 913 482 U. S. June 8, 1987 No. 86-1294. Webster, Director of Central Intelligence v. Doe. C. A. D. C. Cir. Certiorari granted. Reported below: 254 U. S. App. D. C. 282, 796 F. 2d 1508. No. 86-1461. Edward J. DeBartolo Corp. v. Florida Gulf Coast Building & Construction Trades Council et al. C. A. 11th Cir. Certiorari granted. Reported below: 796 F. 2d 1328. No. 86-1650. Trans World Airlines, Inc. v. Independent Federation of Flight Attendants. C. A. 8th Cir. Certiorari granted. Reported below: 809 F. 2d 483. No. 86-1672. Commissioner of Internal Revenue v. Bollinger et AL. C. A. 6th Cir. Certiorari granted. Reported below: 807 F. 2d 65. Certiorari Denied. (See also Nos. 86-1096 and 86-6758, supra.) No. 86-129. Pulido-Zorrilla v. United States. C. A. 3d Cir. Certiorari denied. No. 86-1240. Taylor v. Bowen, Secretary of Health and Human Services. C. A. 3d Cir. Certiorari denied. Reported below: 787 F. 2d 584. No. 86-1263. Henrich v. Texas. Ct. App. Tex., 5th Dist. Certiorari denied. Reported below: 697 S. W. 2d 841. No. 86-1287. Ahlberg et al. v. United States Department of Health and Human Services; and Sumida v. United States Department of Health and Human Services. C. A. Fed. Cir. Certiorari denied. Reported below: 804 F. 2d 1238 (first case). No. 86-1345. Aragon v. California. Ct. App. Cal., 4th App. Dist. Certiorari denied. No. 86-1399. Young v. United States. C. A. 8th Cir. Certiorari denied. Reported below: 804 F. 2d 116. No. 86-1451. Oglala Sioux Tribe of the Pine Ridge Indian Reservation v. United States et al. C. A. Fed. Cir. Certiorari denied. Reported below: 806 F. 2d 1046. No. 86-1454. Mays et al. v. United States. C. A. 10th Cir. Certiorari denied. Reported below: 806 F. 2d 976. 914 OCTOBER TERM, 1986 482 U. S. June 8, 1987 No. 86-1488. Di Bella v. United States. C. A. 2d Cir. Certiorari denied. Reported below: 833 F. 2d 1003. No. 86-1526. Griffin v. Hilke et al.; and No. 86-1684. Hilke v. Griffin. C. A. 8th Cir. Certiorari denied. Reported below: 804 F. 2d 1052. No. 86-1533. Iannelli v. New York. Ct. App. N. Y. Certiorari denied. Reported below: 69 N. Y. 2d 684, 504 N. E. 2d 383. No. 86-1538. New York v. Mosley. Ct. App. N. Y. Certiorari denied. Reported below: 68 N. Y. 2d 881, 501 N. E. 2d 580. No. 86-1591. City of Twin Falls, Idaho v. Envirotech Corp., dba Envirotech Systems, Inc. C. A. 9th Cir. Certiorari denied. Reported below: 806 F. 2d 862. No. 86-1598. Kaiser Engineers, Division of Henry J. Kaiser Co., et al. v. Allendale Mutual Insurance Co.; and No. 86-1615. AZL Engineering, Inc., Formerly Sergent Hauskins & Beckwith v. Allendale Mutual Insurance Co. C. A. 10th Cir. Certiorari denied. Reported below: 804 F. 2d 592. No. 86-1632. Wetzel v. Arizona State Real Estate Department. Ct. App. Ariz. Certiorari denied. Reported below: 151 Ariz. 330, 727 P. 2d 825. No. 86-1645. Batayola v. Municipality of Metropolitan Seattle. C. A. 9th Cir. Certiorari denied. Reported below: 798 F. 2d 355. No. 86-1657. Woolery v. Nevada. Sup. Ct. Nev. Certiorari denied. Reported below: 103 Nev. 826. No. 86-1660. Coulter v. Tennessee et al. C. A. 6th Cir. Certiorari denied. Reported below: 805 F. 2d 146. No. 86-1663. Via v. Williams, Commissioner, Virginia Division of Motor Vehicles. Sup. Ct. Va. Certiorari denied. No. 86-1670. Kile v. Pacific Telephone & Telegraph et al. Ct. App. Cal., 3d App. Dist. Certiorari denied. ORDERS 915 482 U. S. June 8, 1987 No. 86-1676. Morrison v. Illinois. App. Ct. Ill., 3d Dist. Certiorari denied. Reported below: 149 Ill. App. 3d 282, 500 N. E. 2d 442. No. 86-1677. Huiskamp v. New York Connecting Railroad Co. et al. C. A. 3d Cir. Certiorari denied. Reported below: 806 F. 2d 252. No. 86-1679. Averbach v. Rival Manufacturing Co. C. A. 3d Cir. Certiorari denied. Reported below: 809 F. 2d 1016. No. 86-1680. Nicolet Instrument Corp. v. Bio-Rad Laboratories, Inc. C. A. Fed. Cir. Certiorari denied. Reported below: 807 F. 2d 964. No. 86-1682. Ferrino v. Cunard Line Ltd. C. A. 3d Cir. Certiorari denied. Reported below: 802 F. 2d 446. No. 86-1687. Hampton v. Brewer, Appeal Officer for Department of Administration, State of Nevada. Sup. Ct. Nev. Certiorari denied. Reported below: 103 Nev. 73, 733 P. 2d 852. No. 86-1690. Meadows v. Kuhlmann, Superintendent, Sullivan Correctional Facility, et al. C. A. 2d Cir. Certiorari denied. Reported below: 812 F. 2d 72. No. 86-1698. International Distribution Centers, Inc. v. Walsh Trucking Co. et al. C. A. 2d Cir. Certiorari denied. Reported below: 812 F. 2d 786. No. 86-1726. Globe Life Insurance Co. et al. v. Ed Miniat, Inc., et al. C. A. 7th Cir. Certiorari denied. Reported below: 805 F. 2d 732. No. 86-1757. Cloutier v. United States. C. A. 2d Cir. Certiorari denied. Reported below: 816 F. 2d 670. No. 86-1775. Darnell v. City of Jasper, Alabama. C. A. Uth Cir. Certiorari denied. Reported below: 806 F. 2d 1068. No. 86-1792. Walsh et al. v. Ford Motor Co. C. A. D. C. Cir. Certiorari denied. Reported below: 257 U. S. App. D. C. 85, 807 F. 2d 1000. 916 OCTOBER TERM, 1986 June 8, 1987 482 U. S. No. 86-1831. Camer v. Seattle Post-Intelligencer et al. Ct. App. Wash. Certiorari denied. Reported below: 45 Wash. App. 29, 723 P. 2d 1195. No. 86-5173. Vester v. Rogers, Warden, et al. C. A. 4th Cir. Certiorari denied. Reported below: 795 F. 2d 1179. No. 86-6016. Davis v. United States. C. A. 6th Cir. Certiorari denied. Reported below: 805 F. 2d 1036. No. 86-6255. Davis v. Oklahoma. Ct. Crim. App. Okla. Certiorari denied. Reported below: 728 P. 2d 846. No. 86-6299. Drane v. Mississippi. Sup. Ct. Miss. Certiorari denied. Reported below: 493 So. 2d 294. No. 86-6305. Ramirez v. United States. C. A. 11th Cir. Certiorari denied. Reported below: 805 F. 2d 1447. No. 86-6330. Ayodeji v. United States. C. A. 4th Cir. Certiorari denied. Reported below: 802 F. 2d 740. No. 86-6392. Guillory v. St. Landry Parish Police Jury et al. C. A. 5th Cir. Certiorari denied. Reported below: 802 F. 2d 822. No. 86-6431. Melton v. United States. C. A. 11th Cir. Certiorari denied. Reported below: 808 F. 2d 1522. No. 86-6433. Bates v. Blackburn, Warden. C. A. 5th Cir. Certiorari denied. Reported below: 805 F. 2d 569. No. 86-6514. Glenn v. Sowders, Warden. C. A. 6th Cir. Certiorari denied. Reported below: 811 F. 2d 605. No. 86-6515. Neuroth v. United States. C. A. 6th Cir. Certiorari denied. Reported below: 809 F. 2d 339. No. 86-6519. Cofield v. Merit Systems Protection Board. C. A. Fed. Cir. Certiorari denied. No. 86-6526. Misenko v. Wisconsin. Ct. App. Wis. Certiorari denied. Reported below: 135 Wis. 2d 543, 401 N. W. 2d 27. No. 86-6532. Trullo v. United States. C. A. 1st Cir. Certiorari denied. Reported below: 809 F. 2d 108. ORDERS 917 482 U. S. June 8, 1987 No. 86-6554. Carwile v. Kentucky. Sup. Ct. Ky. Certiorari denied. No. 86-6557. Riley v. City of Junction City, Kansas. Sup. Ct. Kan. Certiorari denied. Reported below: 240 Kan. 614, 731 P. 2d 310. No. 86-6576. Gopman v. United States. C. A. 11th Cir. Certiorari denied. Reported below: 806 F. 2d 1487. No. 86-6701. Koenig v. Iowa. Sup. Ct. Iowa. Certiorari denied. No. 86-6713. Williams v. Blackburn, Warden. C. A. 5th Cir. Certiorari denied. Reported below: 813 F. 2d 405. No. 86-6721. Brown v. Dugger, Secretary, Florida Department of Corrections. C. A. 11th Cir. Certiorari denied. No. 86-6722. Jones v. United States. C. A. 10th Cir. Certiorari denied. No. 86-6725. Bryant v. Kansas. Sup. Ct. Kan. Certiorari denied. Reported below: 240 Kan. 803. No. 86-6728. Evans v. Oberle^Jordre Co., Inc. C. A. 7th Cir. Certiorari denied. Reported below: 804 F. 2d 143. No. 86-6729. Dorrill v. United States. C. A. 11th Cir. Certiorari denied. Reported below: 812 F. 2d 1413. No. 86-6730. Turner v. Fuerst. C. A. 6th Cir. Certiorari denied. Reported below: 815 F. 2d 706. No. 86-6736. Fixel v. Slansky et al. C. A. 9th Cir. Certiorari denied. No. 86-6738. Hammond v. Alabama. Ct. Crim. App. Ala. Certiorari denied. Reported below: 502 So. 2d 843. No. 86-6739. Battles v. Oklahoma. Ct. Crim. App. Okla. Certiorari denied. Reported below: 732 P. 2d 480. No. 86-6740. Charrette v. California. Ct. App. Cal., 2d App. Dist. Certiorari denied. 918 OCTOBER TERM, 1986 June 8, 1987 482 U. S. No. 86-6746. Max v. Lane, Director, Illinois Department of Corrections. C. A. 7th Cir. Certiorari denied. Reported below: 810 F. 2d 1167. No. 86-6756. Cooper v. Simon. Ct. App. Mo., Western Dist. Certiorari denied. Reported below: 719 S. W. 2d 463. No. 86-6760. Sears v. Maryland. Ct. Sp. App. Md. Certiorari denied. Reported below: 68 Md. App. 749. No. 86-6761. Wilson v. McKenna et al. C. A. 6th Cir. Certiorari denied. Reported below: 815 F. 2d 707. No. 86-6776. Butler v. Welschmeyer. C. A. 8th Cir. Certiorari denied. Reported below: 815 F. 2d 711. No. 86-6789. Moore v. Ohio. Ct. App. Ohio, Butler County. Certiorari denied. No. 86-6791. Hernandez-Cano v. United States. C. A. Uth Cir. Certiorari denied. Reported below: 808 F. 2d 779. No. 86-6796. Easley v. Petsock, Superintendent, State Correctional Institution and Diagnostic and Classification Center at Pittsburgh, et al. C. A. 3d Cir. Certiorari denied. No. 86-6811. Porter v. Dugger, Secretary, Florida Department of Corrections. C. A. 11th Cir. Certiorari denied. Reported below: 805 F. 2d 930. No. 86-6814. Montgomery v. United States. Ct. App. D. C. Certiorari denied. No. 86-6832. Thompson v. United States. C. A. 6th Cir. Certiorari denied. Reported below: 815 F. 2d 706. No. 86-6834. Moore v. United States. C. A. 7th Cir. Certiorari denied. Reported below: 804 F. 2d 144. No. 86-6848. Page v. United States. C. A. 10th Cir. Certiorari denied. Reported below: 808 F. 2d 723. No. 86-1353. Foltz, Warden v. Thigpen. C. A. 6th Cir. Motion of respondent for leave to proceed in forma pauperis granted. Certiorari denied. Reported below: 804 F. 2d 893. ORDERS 919 482 U. S. June 8, 1987 No. 86-1507. Dugger, Secretary, Florida Department of Corrections v. Porter. C. A. 11th Cir. Motion of respondent for leave to proceed in forma pauperis granted. Certiorari denied. Reported below: 805 F. 2d 930. No. 86-1371. Telecommunications Research & Action Center et al. v. Federal Communications Commission et al. C. A. D. C. Cir. Certiorari denied. Justice Scalia took no part in the consideration or decision of this petition. Reported below: 255 U. S. App. D. C. 287, 801 F. 2d 501. No. 86-1373. New York et al. v. Thomas, Administrator, United States Environmental Protection Agency, et al.; and No. 86-1374. Her Majesty the Queen in Right of Ontario et al. v. Thomas, Administrator, United States Environmental Protection Agency, et al. C. A. D. C. Cir. Certiorari denied. Justice Powell and Justice Scalia took no part in the consideration or decision of these petitions. Reported below: 256 U. S. App. D. C. 49, 802 F. 2d 1443. No. 86-1475. Burlington Northern Railroad Co. v. Bell et al. Ct. App. Okla. Certiorari denied. Reported below: 738 P. 2d 949. Justice White, dissenting. This case presents the question whether a state court may proceed to adjudicate a case after a removal petition has been filed in federal court. Respondents in this case brought a negligence action against petitioner and an individual defendant in state court. After seven days of trial, the trial court dismissed the individual defendant. Petitioner then filed a removal petition in federal court, alleging diversity of citizenship. The state trial court proceeded despite the removal petition and the jury rendered a verdict against petitioner. Subsequently, the District Court held that the case was not properly removed because the individual defendant was dismissed on the merits and not, as petitioner contended, with the consent of the plaintiffs. The District Court remanded the case and the state trial court entered judgment against petitioner on the basis of the previously rendered jury verdict. The trial court and the Oklahoma Court of Appeals rejected petitioner’s argument that the verdict against it is void under 28 920 OCTOBER TERM, 1986 482 U. S. June 8, 1987 U. S. C. § 1446(e), which provides that the proper filing of a removal petition “shall effect the removal [to federal court] and the State court shall proceed no further unless and until the case is remanded.” The decision below conflicts with cases holding that when a case has been removed to federal court the state court lacks jurisdiction to act until the case is remanded. See, e. g., South Carolina v. Moore, 447 F. 2d 1067, 1072-1074 (CA4 1971); Mississippi Power Co. v. Luter, 336 So. 2d 753, 755 (Miss. 1976). I would grant certiorari to resolve this conflict. No. 86-6216. No. 86-6711. No. 86-6731. No. 86-6747. No. 86-6795. Demouchette v. Texas. Ct. Crim. App. Tex.; Tarver v. Alabama. Sup. Ct. Ala.; Schlup v. Missouri. Sup. Ct. Mo.; Terry v. Pennsylvania. Sup. Ct. Pa.; and Jackson v. Florida. Sup. Ct. Fla. Certiorari denied. Reported below: No. 86-6216, 731 S. W. 2d 75; No. 86-6711, 500 So. 2d 1256; No. 86-6731, 724 S. W. 2d 236; No. 86- 6747, 513 Pa. 381, 521 A. 2d 398; No. 86-6795, 502 So. 2d 409. Justice Brennan and Justice Marshall, dissenting. Adhering to our views that the death penalty is in all circumstances cruel and unusual punishment prohibited by the Eighth and Fourteenth Amendments, Gregg v. Georgia, 428 U. S. 153, 227, 231 (1976), we would grant certiorari and vacate the death sentences in these cases. No. 86-7013 (A-888). Moore v. Butler, Warden. C. A. 5th Cir. Application for stay of execution of sentence of death, presented to Justice White, and by him referred to the Court, denied. Certiorari denied. Reported below: 819 F. 2d 517. Justice Brennan and Justice Marshall, dissenting. Adhering to our views that the death penalty is in all circumstances cruel and unusual punishment prohibited by the Eighth and Fourteenth Amendments, Gregg v. Georgia, 428 U. S. 153, 227, 231 (1976), we would grant the application for stay of execution and the petition for writ of certiorari and would vacate the death sentence in this case. Rehearing Denied No. 84-6811. McCleskey v. Kemp, Superintendent, Georgia Diagnostic and Classification Center, 481 U. S. 279; ORDERS 921 482 U. S. June 8, 11, 15, 1987 No. 86-1158. SCHELLONG V. IMMIGRATION AND NATURALIZA- tion Service, 481 U. S. 1004; No. 86-1209. Polur v. New York et al., 480 U. S. 932; No. 86-1467. Stemer v. Wayne County Department of Health et al., 481 U. S. 1017; No. 86-1479. Ratcliff v. McKeever et al., 481 U. S. 1017; No. 86-1545. In re Kowalik, 481 U. S. 1012; No. 86-5884. Irving v. Mississippi, 481 U. S. 1042; No. 86-6323. Velilla v. United Technologies Corp., Hamilton Standard Division, et al., 480 U. S. 948; No. 86-6325. No. 86-6333. No. 86-6377. No. 86-6460. 1021; No. 86-6474. Perry v. Gresk et al., 480 U. S. 949; Perry v. Astrike et al., 480 U. S. 949; Lusk v. Florida, 481 U. S. 1024; Cardelle v. Delta Air Lines, Inc., 481 U. S. Robinson, by his Mother and Next Friend, Robinson v. United States, 481 U. S. 1026; and No. 86-6606. Baker v. United States, 481 U. S. 1040. Pe titions for rehearing denied. No. 84-6075. Tison v. Arizona (two cases), 481 U. S. 137. Motion of California Appellate Project for leave to file a memorandum as amicus curiae denied. Petition for rehearing denied. June 11, 1987 Miscellaneous Order No. A-899. Glass v. Butler, Warden. Application for stay of execution of sentence of death, presented to Justice White, and by him referred to the Court, denied. Justice Brennan and Justice Marshall, dissenting. Adhering to our views that the death penalty is in all circumstances cruel and unusual punishment prohibited by the Eighth and Fourteenth Amendments, Gregg v. Georgia, 428 U. S. 153, 227, 231 (1976), we would grant the application for stay of execution in order to give the applicant time to file a petition for writ of certiorari and would grant the petition and vacate the death sentence in this case. June 15, 1987 Appeals Dismissed No. 86-1707. Blinn v. Massachusetts. Appeal from Sup. Jud. Ct. Mass, dismissed for want of substantial federal question. 922 OCTOBER TERM, 1986 June 15, 1987 482 U. S. Justice Brennan and Justice O’Connor would note probable jurisdiction and set case for oral argument. Reported below: 399 Mass. 126, 503 N. E. 2d 25. No. 86-1711. Edwards v. Turner. Appeal from Ct. App. Mich, dismissed for want of jurisdiction. Treating the papers whereon the appeal was taken as a petition for writ of certiorari, certiorari denied. No. 86-1728. Cahoon v. Alton Packaging Corp, et al. Appeal from App. Ct. Ill., 5th Dist., dismissed for want of properly presented federal question. Justice Brennan and Justice Stevens would note probable jurisdiction and set case for oral argument. Reported below: 148 Ill. App. 3d 480, 499 N. E. 2d 522. Certiorari Granted—Vacated and Remanded No. 85-1442. Bowen, Secretary of Health and Human Services v. Johnson et al. C. A. 7th Cir. Motion of respondents for leave to proceed in forma pauperis and certiorari granted. Judgment vacated and case remanded for further consideration in light of Bowen v. Yuckert, ante, p. 137. Reported below: 769 F. 2d 1202 and 776 F. 2d 166. No. 85-7216. Sueiro v. Florida. Sup. Ct. Fla. Motion of petitioner for leave to proceed in forma pauperis and certiorari granted. Judgment vacated and case remanded for further consideration in light of Miller n. Florida, ante, p. 423. Reported below: 487 So. 2d 1071. No. 86-2. Bowen, Secretary of Health and Human Services v. Dixon et al. C. A. 2d Cir. Certiorari granted, judgment vacated, and case remanded for further consideration in light of Bowen v. Yuckert, ante, p. 137. Reported below: 785 F. 2d 1102. No. 86-40. Kidder, Peabody & Co. Inc. v. Intre Sport, Ltd. C. A. 2d Cir. Certiorari granted, judgment vacated, and case remanded for further consideration in light of Shearson/American Express Inc. v. McMahon, ante, p. 220. Reported below: 795 F. 2d 1004. No. 86-282. Drexel Burnham Lambert, Inc., et al. v. King. C. A. 5th Cir. Certiorari granted, judgment vacated, and case remanded for further consideration in light of Shearson!Amer- ORDERS 923 482 U. S. June 15, 1987 ican Express Inc. v. McMahon, ante, p. 220. Reported below: 796 F. 2d 59. No. 86-321. Dean Witter Reynolds Inc. et al. v. Conover. C. A. 9th Cir. Certiorari granted, judgment vacated, and case remanded for further consideration in light of Shearson!American Express Inc. v. McMahon, ante, p. 220. Reported below: 794 F. 2d 520. No. 86-487. Merrill Lynch, Pierce, Fenner & Smith, Inc., et al. v. Jacobson et al.; No. 86-543. Paine, Webber, Jackson & Curtis, Inc., et al. v. Burris; and No. 86-559. Dean Witter Reynolds Inc. v. Blumenthal. C. A. 3d Cir. Certiorari granted, judgments vacated, and cases remanded for further consideration in light of Shearson/American Express Inc. v. McMahon, ante, p. 220. Reported below: 797 F. 2d 1197. No. 86-591. Merrill Lynch, Pierce, Fenner & Smith, Inc., et al. v. Badart et al. C. A. 9th Cir. Certiorari granted, judgment vacated, and case remanded for further consideration in light of Shearson/American Express Inc. v. McMahon, ante, p. 220. Reported below: 797 F. 2d 775. No. 86-668. Merrill Lynch, Pierce, Fenner & Smith, Inc. v. Delman. C. A. 9th Cir. Certiorari granted, judgment vacated, and case remanded for further consideration in light of Shearson/American Express Inc. v. McMahon, ante, p. 220. Reported below: 796 F. 2d 478. No. 86-897. Bowen, Secretary of Health and Human Services v. Wilson et al. C. A. 3d Cir. Certiorari granted, judgment vacated, and case remanded for further consideration in light of Bowen v. Yuckert, ante, p. 137. Reported below: 796 F. 2d 36. No. 86-1160. Shearson Lehman Brothers, Inc., et al. v. Mayaja, Inc., et al. C. A. 5th Cir. Certiorari granted, judgment vacated, and case remanded for further consideration in light of Shearson/American Express Inc. n. McMahon, ante, p. 220. Reported below: 803 F. 2d 157. No. 86-1218. E. F. Hutton & Co., Inc., et al. v. Wolfe et al. C. A. 11th Cir. Certiorari granted, judgment vacated, 924 OCTOBER TERM, 1986 June 15, 1987 482 U. S. and case remanded for further consideration in light of Shear-son/American Express Inc. v. McMahon, ante, p. 220. Reported below: 800 F. 2d 1032. No. 86-1376. A. G. Becker Inc. v. First Biscayne Corp, et al. C. A. 11th Cir. Certiorari granted, judgment vacated, and case remanded for further consideration in light of Shear-son/American Express Inc. v. McMahon, ante, p. 220. Reported below: 808 F. 2d 59. No. 86-5842. Wilkerson v. Florida. Sup. Ct. Fla. Motion of petitioner for leave to proceed in forma pauperis and certiorari granted. Judgment vacated and case remanded for further consideration in light of Miller v. Florida, ante, p. 423. Reported below: 494 So. 2d 210. No. 86-6360. Patterson v. Florida. Sup. Ct. Fla. Motion of petitioner for leave to proceed in forma pauperis and certiorari granted. Judgment vacated and case remanded for further consideration in light of Miller n. Florida, ante, p. 423. Reported below: 499 So. 2d 831. No. 86-6422. Abbott v. Florida. Dist. Ct. App. Fla., 1st Dist. Motion of petitioner for leave to proceed in forma pauperis and certiorari granted. Judgment vacated and case remanded for further consideration in light of Miller v. Florida, ante, p. 423. Reported below: 499 So. 2d 65. Miscellaneous Orders No. A-835 (86-1947). Quality Aluminum Products, Inc. v. National Labor Relations Board. C. A. 6th Cir. Application for stay, addressed to The Chief Justice and referred to the Court, denied. No. A-915. Wingo v. Butler, Warden. Application for stay of execution of sentence of death, presented to Justice White, and by him referred to the Court, denied. Justice Brennan and Justice Marshall, dissenting. Adhering to our views that the death penalty is in all circumstances cruel and unusual punishment prohibited by the Eighth and Fourteenth Amendments, Gregg n. Georgia, 428 U. S. 153, 227, 231 (1976), we would grant the application for stay of execu ORDERS 925 482 U. S. June 15, 1987 tion in order to give the applicant time to file a petition for writ of certiorari and would grant the petition and vacate the death sentence in this case. No. A-916. Wingo v. Butler, Warden. Application for stay of execution of sentence of death, presented to Justice White, and by him referred to the Court, denied. Justice Brennan and Justice Marshall, dissenting. Adhering to our views that the death penalty is in all circumstances cruel and unusual punishment prohibited by the Eighth and Fourteenth Amendments, Gregg v. Georgia, 428 U. S. 153, 227, 231 (1976), we would grant the application for stay of execution in order to give the applicant time to file a petition for writ of certiorari and would grant the petition and vacate the death sentence in this case. No. 85-1765. Bankers Life & Casualty Co. v. Crenshaw. Sup. Ct. Miss. [Probable jurisdiction noted, 480 U. S. 915.] Motion of Alliance of American Insurers et al. for leave to file a brief as amici curiae granted. Justice Stevens took no part in the consideration or decision of this motion. No. 86-492. Boyle, Personal Representative of the Heirs and Estate of Boyle v. United Technologies Corp. C. A. 4th Cir. [Certiorari granted, 479 U. S. 1029.] Motion of petitioner for additional time for oral argument denied. Motion of the Solicitor General for leave to participate in oral argument as amicus curiae and for divided argument granted. Justice Powell took no part in the consideration or decision of these motions. No. 86-595. United States v. Fausto. C. A. Fed. Cir. [Certiorari granted, 479 U. S. 1029.] Motion of the Solicitor General to dispense with printing the joint appendix granted. No. 86-714. Westfall et al. v. Erwin et ux. C. A. 11th Cir. [Certiorari granted, 480 U. S. 905.] Motion of the Solicitor General to dispense with printing the joint appendix granted. No. 86-1521. United States v. Wells Fargo Bank et al. D. C. C. D. Cal. [Probable jurisdiction noted sub nom. United States v. Crocker National Bank, 481 U. S. 1047.] Motion of the Solicitor General to dispense with printing the joint appendix granted. 926 OCTOBER TERM, 1986 June 15, 1987 482 U. S. No. 86-728. Honig, California Superintendent of Public Instruction v. Doe et al. C. A. 9th Cir. [Certiorari granted, 479 U. S. 1084.] Motion of respondents to seal the record granted. Motion of respondents to seal the joint appendix granted. No. 86-836. Hazelwood School District et al. v. Kuhl-meier et al. C. A. 8th Cir. [Certiorari granted, 479 U. S. 1053.] Motion of Planned Parenthood Federation of America, Inc., et al. for leave to file a brief as amici curiae granted. No. 86-6895. Brown v. Schweitzer et al. C. A. 7th Cir. Motion of petitioner for leave to proceed in forma pauperis denied. Petitioner is allowed until July 6, 1987, within which to pay the docketing fee required by Rule 45(a) and to submit a petition in compliance with Rule 33 of the Rules of this Court. Justice Brennan, Justice Marshall, and Justice Stevens, dissenting. For the reasons expressed in Brown v. Herald Co., 464 U. S. 928 (1983), we would deny the petition for writ of certiorari without reaching the merits of the motion to proceed in forma pauperis. No. 86-6780. In re Cooper. Petition for writ of mandamus denied. Certiorari Granted No. 86-978. Gardebring, Commissioner of the Minnesota Department of Human Services v. Jenkins. C. A. 8th Cir. Certiorari granted. Reported below: 801 F. 2d 288. No. 86-1696. Equal Employment Opportunity Commission v. Commercial Office Products Co. C. A. 10th Cir. Certiorari granted. Reported below: 803 F. 2d 581. No. 86-1357. United States Department of Justice et al. v. Julian et al. C. A. 9th Cir. Motion of respondent Kenneth Julian for leave to proceed in forma pauperis granted. Certiorari granted. Reported below: 806 F. 2d 1411. No. 86-5309. Ross v. Oklahoma. Ct. Crim. App. Okla. Motion of petitioner for leave to proceed in forma pauperis granted. Certiorari granted limited to Question I presented by the petition. Reported below: 717 P. 2d 117. ORDERS 927 482 U. S. June 15, 1987 Certiorari Denied. (See also No. 86-1711, supra.) No. 85-983. United Transportation Union v. Interstate Commerce Commission et al.; and No. 85-997. Brotherhood of Locomotive Engineers v. Interstate Commerce Commission et al. C. A. D. C. Cir. Certiorari denied. Reported below: 245 U. S. App. D. C. 311, 761 F. 2d 714. No. 85-1054. United Transportation Union et al. v. Missouri Pacific Railroad Co. et al. C. A. 8th Cir. Certiorari before judgment denied. No. 85-1957. Bruno et al. v. Western Pacific Railroad Co. ET AL. Sup. Ct. Del. Certiorari denied. Reported below: 508 A. 2d 72. No. 86-990. RCA Corp. v. Malia et ux. C. A. 3d Cir. Certiorari denied. Reported below: 794 F. 2d 909. No. 86-1340. Russell et al. v. Shearson Lehman Brothers, Inc., fka Shearson/Lehman American Express, Inc., et AL. C. A. 5th Cir. Certiorari denied. Reported below: 806 F. 2d 259. No. 86-1354. Ma y aja, Inc., et al. v. Shearson Lehman Brothers, Inc., et al. C. A. 5th Cir. Certiorari denied. Reported below: 803 F. 2d 157. No. 86-1447. Torres v. Webb, Secretary of the Navy. C. A. 9th Cir. Certiorari denied. Reported below: 805 F. 2d 1039. No. 86-1529. Orlowski v. United States. C. A. 8th Cir. Certiorari denied. Reported below: 808 F. 2d 1283. No. 86-1576. Wolfe v. United States. C. A. 9th Cir. Certiorari denied. Reported below: 798 F. 2d 1241 and 806 F. 2d 1410. No. 86-1578. Burke, by and Through his Next Friend, Draves v. General Motors Corp, et al. C. A. 6th Cir. Certiorari denied. Reported below: 811 F. 2d 604. No. 86-1617. Bowen, Secretary of Health and Human Services v. Polaski et al. C. A. 8th Cir. Certiorari denied. Reported below: 804 F. 2d 456. 928 OCTOBER TERM, 1986 June 15, 1987 482 U. S. No. 86-1620. Smoky Greenhaw Cotton Co., Inc., et al. v. Merrill Lynch, Pierce, Fenner & Smith, Inc., et al. C. A. 5th Cir. Certiorari denied. Reported below: 805 F. 2d 1221. No. 86-1629. 511 Detroit Street, Inc., et al. v. Kelley, Attorney General of Michigan, et al. C. A. 6th Cir. Certiorari denied. Reported below: 807 F. 2d 1293. No. 86-1693. Y et al. u Kuriansky, Deputy Attorney General of New York for Medicaid Fraud Control. Ct. App. N. Y. Certiorari denied. Reported below: 69 N. Y. 2d 232, 505 N. E. 2d 925. No. 86-1694. Savely et al. v. City of Houston, Texas, et al. Ct. App. Tex., 1st Dist. Certiorari denied. Reported below: 708 S. W. 2d 879. No. 86-1695. Reiboldt v. New Jersey. Super. Ct. N. J., App. Div. Certiorari denied. No. 86-1697. Fortin v. Michigan. Ct. App. Mich. Certiorari denied. No. 86-1702. United Dairymen of Arizona v. LaSalvia et ux. C. A. 9th Cir. Certiorari denied. Reported below: 804 F. 2d 1113. No. 86-1703. Lancaster v. Buerkle Buick Honda Co. C. A. 8th Cir. Certiorari denied. Reported below: 809 F. 2d 539. No. 86-1706. Kelly, Superintendent, Attica Correctional Facility, et al. v. Johnstone. C. A. 2d Cir. Certiorari denied. Reported below: 808 F. 2d 214 and 812 F. 2d 821. No. 86-1708. McDermott v. Tate, Superintendent, Chillicothe Correctional Institute. C. A. 6th Cir. Certiorari denied. Reported below: 811 F. 2d 606. No. 86-1723. Walsh v. Union Pacific Railroad Co. et al. C. A. 8th Cir. Certiorari denied. Reported below: 803 F. 2d 412. No. 86-1730. Setera v. Texas A & M University et al. C. A. 1st Cir. Certiorari denied. Reported below: 808 F. 2d 1513. ORDERS 929 482 U. S. June 15, 1987 No. 86-1732. City of Arlington, Texas v. Byrd et al. Ct. App. Tex., 2d Dist. Certiorari denied. Reported below: 713 S. W. 2d 224. No. 86-1734. Arkansas v. Foster. Sup. Ct. Ark. Certiorari denied. Reported below: 290 Ark. 495, 720 S. W. 2d 712. No. 86-1740. Wanless v. Rothballer et al. Sup. Ct. Ill. Certiorari denied. Reported below: 115 Ill. 2d 158, 503 N. E. 2d 316. No. 86-1742. Marmott et al. v. Maryland Lumber Co. et AL. C. A. 4th Cir. Certiorari denied. Reported below: 807 F. 2d 1180. No. 86-1748. Strelkoff v. Pennsylvania. Super. Ct. Pa. Certiorari denied. Reported below: 359 Pa. Super. 630, 515 A. 2d 619. No. 86-1790. Fleming v. Commissioner of Internal Revenue. C. A. 10th Cir. Certiorari denied. No. 86-1797. Glantz et al. v. United States. C. A. 1st Cir. Certiorari denied. Reported below: 810 F. 2d 316. No. 86-1802. American Savings & Loan Assn. v. Morgan Guaranty Trust Company of New York et al. C. A. 9th Cir. Certiorari denied. Reported below: 804 F. 2d 1487. No. 86-1818. Gaggi v. United States. C. A. 2d Cir. Certiorari denied. Reported below: 811 F. 2d 47. No. 86-1826. Goulding v. Feinglass et al. C. A. 7th Cir. Certiorari denied. Reported below: 811 F. 2d 1099. No. 86-6443. Gabel v. Lynaugh, Director, Texas Department of Corrections. C. A. 5th Cir. Certiorari denied. Reported below: 803 F. 2d 814. No. 86-6490. Mars v. Florida. Sup. Ct. Fla. Certiorari denied. Reported below: 498 So. 2d 402. No. 86-6520. Tyler v. Kentucky. Sup. Ct. Ky. Certiorari denied. No. 86-6528. Dixon v. Pennsylvania. Sup. Ct. Pa. Certiorari denied. No. 86-6547. Armstrong v. Kansas. Sup. Ct. Kan. Certiorari denied. Reported below: 240 Kan. 446, 731 P. 2d 249. 930 OCTOBER TERM, 1986 June 15, 1987 482 U. S. No. 86-6605. Thomas v. United States. C. A. 11th Cir. Certiorari denied. Reported below: 805 F. 2d 1464. No. 86-6645. Sawyer v. Illinois. Sup. Ct. Ill. Certiorari denied. Reported below: 115 Ill. 2d 184, 503 N. E. 2d 331. No. 86-6724. Hamilton v. Reid, Superintendent, State Correctional Institution at Greensburg. C. A. 3d Cir. Certiorari denied. Reported below: 813 F. 2d 397. No. 86-6751. Boag v. Bramlett, Warden, et al. C. A. 9th Cir. Certiorari denied. Reported below: 806 F. 2d 263. No. 86-6762. Parks v. Brady. Sup. Ct. N. H. Certiorari denied. No. 86-6769. LaMothe v. Michigan. Sup. Ct. Mich. Certiorari denied. Reported below: 428 Mich. 864. No. 86-6771. Ruether v. Lohmar, Judge, et al. C. A. 8th Cir. Certiorari denied. No. 86-6773. Webb v. Oklahoma. Ct. Crim. App. Okla. Certiorari denied. Reported below: 732 P. 2d 478. No. 86-6775. Garris v. Lyles, Warden. C. A. 4th Cir. Certiorari denied. Reported below: 812 F. 2d 1400. No. 86-6777. Ewers v. City of Garland et al. C. A. 5th Cir. Certiorari denied. Reported below: 811 F. 2d 601. No. 86-6778. Appleby v. Maryland. Ct. Sp. App. Md. Certiorari denied. Reported below: 69 Md. App. 784. No. 86-6786. Martin v. Little, Brown & Co., Inc., et al. C. A. 3d Cir. Certiorari denied. No. 86-6799. Cason v. Cook et al. C. A. 8th Cir. Certiorari denied. Reported below: 810 F. 2d 188. No. 86-6849. Thomas v. United States. C. A. 5th Cir. Certiorari denied. Reported below: 810 F. 2d 478. No. 86-6862. Kimble v. United States. C. A. 11th Cir. Certiorari denied. Reported below: 814 F. 2d 661. No. 86-6875. Dempsey v. Wisconsin. Ct. App. Wis. Certiorari denied. Reported below: 136 Wis. 2d 557, 402 N. W. 2d 390. ORDERS 931 482 U. S. June 15, 1987 No. 86-6894. Pruett v. United States. C. A. 9th Cir. Certiorari denied. Reported below: 813 F. 2d 408. No. 86-6900. Tripp v. Illinois. App. Ct. Ill., 1st Dist. Certiorari denied. No. 86-578. Phillips et al. v. Merrill Lynch, Pierce, Fenner & Smith, Inc., et al. C. A. 8th Cir. Certiorari denied. Justice Blackmun took no part in the consideration or decision of this petition. Reported below: 795 F. 2d 1393. No. 86-1636. Cadwalader, Wickersham & Taft v. United States District Court for the Central District of California (Gottlieb, Real Party in Interest). C. A. 9th Cir. Certiorari denied. Justice White would grant certiorari. No. 86-6175. Wilson v. Denton et al. C. A. 6th Cir. Certiorari denied. Justice White and Justice Blackmun would grant certiorari. Reported below: 811 F. 2d 608. No. 86-6685. Glenn v. Ohio. Sup. Ct. Ohio; No. 86-6727. Correll v. Virginia. Sup. Ct. Va.; No. 86-6732. Pruett v. Virginia. Sup. Ct. Va.; No. 86-6768. Lucas v. Aiken, Warden, et al. Ct. Common Pleas of York County, S. C.; No. 86-6827. Hicks v. Georgia. Sup. Ct. Ga.; and No. 86-6903. Grubbs v. Missouri. Sup. Ct. Mo. Certiorari denied. Reported below: No. 86-6685, 28 Ohio St. 3d 451, 504 N. E. 2d 701; No. 86-6727, 232 Va. 454, 352 S. E. 2d 352; No. 86-6732, 232 Va. 266, 351 S. E. 2d 1; No. 86-6827, 256 Ga. 715, 352 S. E. 2d 762; No. 86-6903, 724 S. W. 2d 494. Justice Brennan and Justice Marshall, dissenting. Adhering to our views that the death penalty is in all circumstances cruel and unusual punishment prohibited by the Eighth and Fourteenth Amendments, Gregg v. Georgia, 428 U. S. 153, 227, 231 (1976), we would grant certiorari and vacate the death sentences in these cases. Rehearing Denied No. 86-6459. Belk v. Chrans, Warden, et al., 481 U. S. 1021; and No. 86-6475. LaVergne v. Holy Name Hospital, 481 U. S. 1022. Petitions for rehearing denied. 932 OCTOBER TERM, 1986 482 U. S. June 15, 16, 18, 1987 No. 86-6483. McHarris v. Spears, Warden, et al., 481 U. S. 1031; and No. 86-6564. May v. Pro-Guard, Inc., 481 U. S. 1032. Petitions for rehearing denied. June 16, 1987 Miscellaneous Order No. A-814 (86-6867). Lowenfield v. Phelps, Secretary, Louisiana Department of Corrections, et al. C. A. 5th Cir. Application for stay of execution of sentence of death, presented to Justice White, and by him referred to the Court, granted pending the disposition by this Court of the petition for writ of certiorari. Should the petition for writ of certiorari be denied, this stay terminates automatically. In the event the petition for writ of certiorari is granted, this stay shall continue pending the sending down of the judgment of this Court. June 18, 1987 Miscellaneous Order No. A-923. Agan v. Florida. Application for stay of execution of sentence of death, presented to Justice Powell, and by him referred to the Court, denied. Justice Brennan and Justice Marshall, dissenting. Adhering to our views that the death penalty is in all circumstances cruel and unusual punishment prohibited by the Eighth and Fourteenth Amendments, Gregg v. Georgia, 428 U. S. 153, 227, 231 (1976), we would grant the application for stay of execution in order to give the applicant time to file a petition for writ of certiorari and would grant the petition and vacate the death sentence in this case. INDEX ABSTENTION. Ordinance criminalizing interruption of police—Constitutional challenge.— Supreme Court would not abstain from deciding merits of appellee’s First Amendment facial challenge to constitutionality of municipal ordinance that made it unlawful to interrupt any police officer in execution of his or her duties and under which appellee was charged and acquitted. Houston v. Hill, p. 451. ADMINISTRATIVE SEARCHES AND SEIZURES. See Constitutional Law, IX. ADMISSION TO FEDERAL DISTRICT COURT BAR. See Attorneys. AIRPORT BANS OF “FIRST AMENDMENT ACTIVITIES.” See Constitutional Law, V, 1. APPEALS. See Standing to Sue. ARBITRATION. See Constitutional Law, X; Standing to Sue. ARBITRATION ACT. See Constitutional Law, X; Federal Arbitration Act. ASYLUM STATES. See Extradition Act. ATTORNEYS. See also Civil Rights Attorney’s Fees Awards Act of 1976. Admission to bar—Federal District Court—Rules.— Federal District Court in Louisiana was not empowered to adopt a Rule requiring members of Louisiana Bar who applied for admission to court’s bar to live or maintain an office in Louisiana; Rule’s residence requirement is unnecessary and arbitrarily discriminates against out-of-state members of Louisiana Bar, such as petitioner, who neither lived in nor had an office in Louisiana, and in-state office requirement is similarly unnecessary and irrational. Frazier v. Hebbe, p. 641. ATTORNEY’S FEES. See Civil Rights Attorney’s Fees Awards Act of 1976. AUTOMOBILE JUNKYARDS. See Constitutional Law, IX. CALIFORNIA. See Constitutional Law, X; Extradition Act; Standing to Sue. CAPITAL MURDER. See Constitutional Law, II. 933 934 INDEX CHILD CUSTODY DISPUTES. See Extradition Act. CHILD WITNESSES. See Constitutional Law, I, 1; VIL CIVIL RIGHTS ACT OF 1866. See also Civil Rights Act of 1964. Racial discrimination suit against employer and unions—Limitations period—Retroactivity.—In petitioner employees’ suit against employer and unions, asserting racial discrimination claims under 42 U. S. C. § 1981, inter alia, Court of Appeals was correct in selecting Pennsylvania 2-year limitations period governing personal injury actions as most analogous state statute of limitations to govern § 1981 suits, and in concluding that statute should be applied retroactively to petitioners. Goodman v. Lukens Steel Co., p. 656. CIVIL RIGHTS ACT OF 1871. See Civil Rights Attorney’s Fees Awards Act of 1976. CIVIL RIGHTS ACT OF 1964. Racial discrimination—Violation by unions.—In petitioner employees’ suit against employer and unions, asserting racial discrimination claims under Title VII of Act and 42 U. S. C. § 1981, courts below properly held that unions violated both Title VII and § 1981 in failing to challenge employer’s discriminatory discharges of probationary employees, in failing and refusing to assert instances of racial discrimination as grievances, and in tolerating and tacitly encouraging racial harassment. Goodman v. Lukens Steel Co., p. 656. CIVIL RIGHTS ATTORNEY’S FEES AWARDS ACT OF 1976. Prevailing party.— Where, after he was found guilty of misconduct in connection with prison riot and sentenced to disciplinary confinement solely on basis of undisclosed informant’s statements, respondent brought suit against petitioner prison officials for damages and injunctive relief under 42 U. S. C. § 1983, and ultimately summary judgment for petitioners was entered on basis of qualified immunity and Court of Appeals affirmed, respondent was not a “prevailing party” eligible for attorney’s fees under 42 U. S. C. § 1988 on basis of Court of Appeals’ statement that its holding that respondent’s constitutional rights had been violated was “a form of judicial relief.” Hewitt v. Helms, p. 755. CLOSING OF PLANT AS REQUIRING SEVERANCE PAY. See Preemption of State Law by Federal Law. COLLECTIVE BARGAINING. See Labor. COLLECTIVE-BARGAINING AGREEMENTS. See Removal of Actions. INDEX 935 COMPACTS BETWEEN STATES. Pecos River Compact—Remedy for past breaches. —Supreme Court is not limited to ordering prospective relief but may provide a remedy for past breaches of Pecos River Compact between New Mexico and Texas, since, although such a Compact, when approved by Congress, becomes law of United States, it is still a contract, subject to construction and application in accordance with its terms; case is returned to Special Master for his recommendations as to whether New Mexico should be allowed to elect to pay money damages for past shortages in amount of water it was required to deliver to Texas under Compact, and a decree is entered with respect to New Mexico’s current and future obligation to deliver water pursuant to Compact. Texas v. New Mexico, p. 124. COMPETENCY OF MINOR WITNESSES TO TESTIFY. See Constitutional Law, I, 1; VII. CONSTITUTIONAL LAW. See also Abstention; Civil Rights Attorney’s Fees Awards Act of 1976; Extradition Act; Judicial Review, 2, 3; Judgments; Property Taxes; Standing to Sue. I. Due Process. 1. Minor witnesses—Competency hearing—Exclusion of defendant. — In prosecution of respondent for committing sodomy with two minor girls, his rights under Due Process Clause of Fourteenth Amendment were not violated by excluding him from hearing to determine competency of girls to testify; his due process right to be present at critical stages of a criminal proceeding if his presence would contribute to fairness of procedure was not implicated. Kentucky v. Stincer, p. 730. 2. State parole-release statute. —With respect to respondent prisoners who were denied parole, a Montana statute—which provided that a prisoner eligible for parole “shall” be released when there is a reasonable probability that no detriment will result to him or community and specified that parole shall be ordered for best interests of society and when State Board of Pardons believes that prisoner is able and willing to assume obligations of a law-abiding citizen—clearly created a liberty interest in parole release that is protected by Due Process Clause of Fourteenth Amendment. Board of Pardons v. Allen, p. 369. II. Eighth Amendment. Death sentence—Consideration of victim impact statement. —Introduction of a victim impact statement at sentencing phase of petitioner’s capital murder trial violated Eighth Amendment, and therefore Maryland statute was invalid to extent it required consideration of this information, which is irrelevant to a capital sentencing decision and admission of which creates a 936 INDEX CONSTITUTIONAL LAW-Continued. constitutionally unacceptable risk that jury may impose death penalty in an arbitrary and capricious manner. Booth v. Maryland, p. 496. III. Ex Post Facto Laws. Sentencing guidelines.— Application of Florida’s revised sentencing guidelines law to petitioner, where crimes occurred before law’s effective date, violated Ex Post Facto Clause of Article I of Constitution. Miller v. Florida, p. 423. IV. Freedom of Religion. Teaching of theory of evolution—State prohibition. — Louisiana’s “Creationism Act,” which forbids teaching of theory of evolution in public schools unless accompanied by instruction in theory of “creation science,” is facially invalid as violative of Establishment Clause of First Amendment, because it lacks a clear secular purpose. Edwards v. Aguillard, p. 578. V. Freedom of Speech. 1. Airport ban of “First Amendment activities Overbreadth. — Resolution of Los Angeles Board of Airport Commissioners banning all “First Amendment activities” within central terminal area at Los Angeles International Airport, challenged by a nonprofit religious corporation and a minister for that organization who stopped distributing religious literature in airport’s central terminal area when warned against doing so by an airport officer, is facially unconstitutional under First Amendment overbreadth doctrine regardless of whether forum involved is public or nonpublic. Board of Airport Comm’rs v. Jews for Jesus, Inc., p. 569. 2. Ordinance criminalizing interruption of police—Overbreadth.—A municipal ordinance making it unlawful to interrupt a police officer in performance of his or her duties is unconstitutionally overbroad under First Amendment; ordinance criminalizes a substantial amount of, and is susceptible of regular application to, constitutionally protected speech and accords police unconstitutional enforcement discretion. Houston v. Hill, p. 451. 3. Regulation of inmate-to-inmate correspondence.— Missouri prison regulation permitting correspondence between immediate family members who are inmates at different institutions and between inmates “concerning legal matters,” but allowing other inmate correspondence only if each inmate’s classification/treatment team deems it in parties’ best interests, is reasonable and facially valid; it is logically related to prison officials’ legtimate security concerns and does not deprive prisoners of all means of expression but simply bars communication with a limited class of people— other inmates—with whom authorities have particular cause to be concerned. Turner v. Safley, p. 78. INDEX 937 CONSTITUTIONAL LAW-Continued. VI. Free Exercise of Religion. Prison regulations.— New Jersey prison regulations (1) requiring prisoners to work outside buildings where respondent Muslim prisoners were housed and weekly Muslim services were held on Friday afternoons and (2) prohibiting prisoners assigned to outside work from returning to buildings during day are reasonably related to legitimate penological interests and therefore do not offend Free Exercise Clause of First Amendment. O’Lone v. Estate of Shabazz, p. 342. VII. Right to Confront Witnesses. Minor witnesses—Competency hearing—Exclusion of defendant.—In prosecution of respondent for committing sodomy with two minor girls, his rights under Confrontation Clause of Sixth Amendment were not violated by excluding him from hearing to determine girls’ competency to testify; such exclusion did not interfere with respondent’s opportunity for cross-examination because girls were cross-examined in open court and any questions asked during hearing could have been repeated during direct and cross-examination. Kentucky v. Stincer, p. 730. VIII. Right to Marry. Prisoners.— Constitutional right of prisoners to marry is impermissibly burdened by Missouri regulation permitting an inmate to marry only with prison superintendent’s permission, which can be given only when there are “compelling reasons” to do so; regulation is facially invalid as not being reasonably related to legitimate penological objectives. Turner v. Safley, p. 78. IX. Searches and Seizures. Warrantless inspection of “closely regulated” business—Automobile junkyard. —Warrantless search of respondent’s automobile junkyard pursuant to New York statute authorizing such search fell within exception to Fourth Amendment’s warrant requirement for administrative inspections of “closely regulated” businesses; statute does not violate Fourth Amendment on ground that it was designed simply to give police an expedient means of enforcing penal sanctions for possession of stolen property. New York v. Burger, p. 691. X. Supremacy Clause. Federal Arbitration Act—Pre-emption of state law.— Under Supremacy Clause, § 2 of Federal Arbitration Act, which mandates enforcement of arbitration agreements, pre-empted provision of California Labor Code, which permits wage collection actions to be maintained without regard to any private agreement to arbitrate and which appellee claimed authorized his breach-of-contract suit against appellants, his former employer and two 938 INDEX CONSTITUTIONAL LAW-Continued. of its employees, regarding dispute over securities sales commission, despite provision in a form appellee executed in connection with employment application whereby he agreed to arbitrate any dispute with employer. Perry v. Thomas, p. 483. XL Taking of Property. “Temporary” regulatory taking—Inverse condemnation—Compensation.— Appellant church’s claim that California Supreme Court in Agins v. Tiburon, 24 Cal. 3d 266, 598 P. 2d 25, improperly held that Fifth Amendment, as made applicable to States through Fourteenth Amendment, does not require compensation as a remedy for “temporary” regulatory takings, was properly presented to United States Supreme Court; appellant—who claimed that county ordinance, which prohibited construction or reconstruction of any building or structure in interim flood protection area that included land on which stood appellant’s campground buildings that had been destroyed by flood, denied appellant all use of campground—was entitled to recover in inverse condemnation compensation for time before a final determination that ordinance constituted a “taking” of property. First English Evangelical Lutheran Church v. Los Angeles County, p. 304. CORRESPONDENCE BETWEEN PRISONERS. See Constitutional Law, V, 3; VIII; Judicial Review, 3. COSTS IN COURT PROCEEDINGS. See Procedure. COURTS OF APPEALS. See Jurisdiction. “CREATIONISM ACT.” See Constitutional Law, IV; Judgments. CREWING PROCEDURES OF RAILROADS. See Judicial Review, 1. CRIMINAL LAW. See Abstention; Constitutional Law, 1,1; II; III, V, 2; VII; IX; Extradition Act. CROSS-EXAMINATIONS. See Constitutional Law, I, 1; VII. CUSTODY OF CHILDREN. See Extradition Act. DEATH SENTENCES. See Constitutional Law, II. DISABILITY BENEFITS UNDER SOCIAL SECURITY ACT. See Social Security Act. DISCOVERY. See Hague Convention on the Taking of Evidence Abroad in Civil or Commercial Matters. DISTRICT COURTS. See Attorneys; Hague Convention on the Taking of Evidence Abroad in Civil or Commercial Matters; Judgments; Removal of Actions. DUE PROCESS. See Constitutional Law, I; VII. INDEX 939 EIGHTH AMENDMENT. See Constitutional Law, IL ELIGIBILITY FOR PAROLE. See Constitutional Law, I, 2. EMINENT DOMAIN. See Constitutional Law, XI. EMPLOYEE RETIREMENT INCOME SECURITY ACT OF 1974 (ERISA). See Pre-emption of State Law by Federal Law. EMPLOYER AND EMPLOYEES. See Civil Rights Act of 1866; Civil Rights Act of 1964; Constitutional Law, X; Labor; Pre-emption of State Law by Federal Law; Removal of Actions; Standing to Sue. EMPLOYMENT CONTRACTS. See Removal of Actions. EQUAL-FOOTING DOCTRINE. See Federal-State Relations. ESTABLISHMENT OF RELIGION. See Constitutional Law, IV. EXCLUSION OF ACCUSED FROM WITNESS COMPETENCY HEARING. See Constitutional Law, I, 1; VII. EXEMPTIONS FROM STATE TAXATION. See Property Taxes. EXPERT WITNESS FEES. See Procedure. EX POST FACTO LAWS. See Constitutional Law, III. EXTRADITION ACT. Child custody dispute—Extradition of parent.— Extradition Act, which implements Extradition Clause of Article IV of Constitution and requires an asylum State to give up to a demanding State a fugitive against whom a properly certified indictment has been lodged, prohibited California Supreme Court from refusing to permit extradition to Louisiana of father and grandfather who, after having picked up father’s minor children in Louisiana where they were living with their mother to whom custody had originally been awarded in California divorce proceeding, were charged, in Louisiana, based on mother’s affidavit, with violating Louisiana parental kidnaping statute; it was for Louisiana courts to determine whether mother’s affidavit was fraudulent, whether modified California decrees awarding custody to father established him as children’s lawful custodian under full faith and credit provision of federal Parental Kidnaping Prevention Act, and whether father and grandfather were, accordingly, not guilty of violating Louisiana statute. California v. Superior Court of California, p. 400. FACIAL OVERBREADTH. See Constitutional Law, V, 1. FEDERAL ARBITRATION ACT. See also Constitutional Law, X. Securities Exchange Act claims—RICO claims.— Claims of respondent customers of petitioner brokerage firm under agreements providing for arbitration of any controversy relating to their accounts that petitioner had violated antifraud provisions of § 10(b) of Securities Exchange Act of 1934 940 INDEX FEDERAL ARBITRATION ACT-Continued. and Racketeer Influenced and Corrupt Organizations Act were arbitrable under Arbitration Act. Shearson/American Express Inc. v. McMahon, p. 220. FEDERAL DISTRICT COURTS. See Attorneys; Hague Convention on the Taking of Evidence Abroad in Civil or Commercial Matters; Judgments; Removal of Actions. FEDERAL INCOME TAXES. See Internal Revenue Code. FEDERAL PARENTAL KIDNAPING ACT. See Extradition Act. FEDERAL RULES OF CIVIL PROCEDURE. See Hague Convention on the Taking of Evidence Abroad in Civil or Commercial Matters. FEDERAL-STATE RELATIONS. See also Constitutional Law, X; Pre-emption of State Law by Federal Law; Removal of Actions; Standing to Sue. Title to lakebed—Equal-footing doctrine.— Title to Utah Lake’s bed passed to Utah under equal-footing doctrine upon Utah’s admission to Union; even assuming that a federal reservation of lakebed could defeat Utah’s claim to title under that doctrine, such defeat was not accomplished on facts. Utah Division of State Lands v. United States, p. 193. FEDERAL TORT CLAIMS ACT. See Jurisdiction. FIFTH AMENDMENT. See Constitutional Law, XI. FIRST AMENDMENT. See Abstention; Constitutional Law, IV; V; VI; Judicial Review, 2. FLORIDA. See Constitutional Law, III. FOREIGN LITIGANTS AS SUBJECT TO DISCOVERY. See Hague Convention on the Taking of Evidence Abroad in Civil or Commercial Matters. FOURTEENTH AMENDMENT. See Constitutional Law, I; VII; XI. FOURTH AMENDMENT. See Constitutional Law, IX. FREEDOM OF RELIGION. See Constitutional Law, IV; VI; Judicial Review, 2. FREEDOM OF SPEECH. See Abstention; Constitutional Law, V. “GINNIE MAES.” See Property Taxes. GOVERNMENT NATIONAL MORTGAGE ASSOCIATION. See Property Taxes. GUIDELINES FOR SENTENCING. See Constitutional Law, III. INDEX 941 HAGUE CONVENTION ON THE TAKING OF EVIDENCE ABROAD IN CIVIL OR COMMERCIAL MATTERS. Discovery from foreign litigant.—In suit for injuries resulting from crash of aircraft built and sold by corporation owned by France, Federal District Court was not required to employ procedures set forth in Convention for obtaining documents and information from defendants over whom court has personal jurisdiction; Convention does not provide exclusive or mandatory procedures but was intended to establish optional procedures for obtaining evidence abroad, and international comity does not require in all instances that American litigants first resort to Convention procedures before initiating discovery under Federal Rules of Civil Procedure. Société Nationale Industrielle Aérospatiale v. United States District Court, p. 522. ILLINOIS. See Property Taxes. INCOME TAXES. See Internal Revenue Code. INMATE MARRIAGES. See Constitutional Law, V, 3; VIII; Judicial Review, 3. INMATE-TO-INMATE CORRESPONDENCE. See Constitutional Law, V, 3; VIII; Judicial Review, 3. INTERGOVERNMENTAL TAX IMMUNITY. See Property Taxes. INTERNAL REVENUE CODE. Income tax—Underpayment—Negligence penalty.—Section 6653(a)(1) of Code providing that if “any part of any underpayment” is due to negligence a penalty of “5 percent of the underpayment” shall be added to tax clearly establishes that penalty is imposed on entire amount of underpayment, not just on part of underpayment attributable to negligence; thus, where respondent asphalt manufacturer was found to have negligently deducted on its tax return expense of transporting two trucks from their place of purchase to respondent’s place of business because they had detoured to pick up equipment bought by respondent’s shareholders in their individual capacities, resulting in a tax deficiency, negligence penalty on full amount of respondent’s deficiency was proper. Commissioner v. Asphalt Products Co., p. 117. INTERNATIONAL COMITY. See Hague Convention on the Taking of Evidence Abroad in Civil or Commercial Matters. INTERNMENT CAMPS FOR JAPANESE-AMERICANS. See Jurisdiction. INTERRUPTION OF POLICE IN PERFORMANCE OF DUTIES. See Abstention; Constitutional Law, V, 2. INTERSTATE COMMERCE COMMISSION. See Judicial Review, 1. 942 INDEX INTERSTATE COMPACTS. See Compacts Between States. INVERSE CONDEMNATION. See Constitutional Law, XI. JAPANESE-AMERICANS. See Jurisdiction. JUDGMENTS. Summary judgments—No genuine issue of material fact.—In action by appellee parent, teachers, and religious leaders challenging constitutionality of Louisiana’s “Creationism Act,” which forbids teaching of theory of evolution in public schools unless accompanied by instruction in theory of “creation science,” Federal District Court did not err in granting summary judgment upon finding that appellant Louisiana officials had failed to raise a genuine issue of material fact. Edwards v. Aguillard, p. 578. JUDICIAL REVIEW. 1. ICC orders—Railroad trackage rights—Crewing procedures.— Although respondent unions’ petitions for Court of Appeals review of Interstate Commerce Commission’s order were timely filed, they should have been dismissed since order itself—whereby ICC refused to reconsider earlier order refusing to clarify, with respect to railroads’ crewing procedures, its prior order granting petitioner railroad and another railroad right to use tracks of third railroad—was unreviewable; similarly, although unions’ petition for Court of Appeals review of order refusing to clarify its prior trackage rights approval order was timely filed, it too should have been dismissed since refusal order itself was not appealable. ICC v. Locomotive Engineers, p. 270. 2. Review of prison regulations—Standard of scrutiny.—In respondent Muslim prisoners’ action contending that New Jersey prison regulations (1) requiring prisoners to work outside buildings where respondents were housed and weekly Muslim services were held on Friday afternoons and (2) prohibiting prisoners assigned to outside work from returning to buildings during day violated respondents’ rights under Free Exercise Clause of First Amendment, Court of Appeals, in appeal from District Court’s holding that no constitutional violation occurred, erred in placing burden on prison officials to disprove availability of alternative methods of accommodating prisoners’ religious rights; that approach fails to reflect respect and deference Constitution allows for prison administrators’ judgment. O’Lone v. Estate of Shabazz, p. 342. 3. Review of prison rules—Standard of scrutiny.—In reviewing constitutionality of Missouri prison regulations relating to inmate marriages and inmate-to-inmate correspondence, it was not appropriate to apply a strict standard of scrutiny, but rather a lesser standard whereby inquiry is made whether regulations were “reasonably related” to legitimate penological interests should be applied. Turner v. Safley, p. 78. JUNKYARDS. See Constitutional Law, IX. INDEX 943 JURISDICTION. See also Hague Convention on the Taking of Evidence Abroad in Civil or Commercial Matters; Removal of Actions. Little Tucker Act and FTC A claims—Federal Circuit.— United States Court of Appeals for Federal Circuit rather than appropriate regional Court of Appeals had jurisdiction over an appeal from District Court’s decision in respondent Japanese-American organization’s and individuals’ suit raising both a nontax claim under Little Tucker Act and a claim under Federal Tort Claims Act and seeking damages and declaratory relief for injuries suffered when Government during World War II placed Japanese-Americans in internment camps. United States v. Hohri, p. 64. JUST COMPENSATION. See Constitutional Law, XI. KIDNAPING. See Extradition Act. LABOR. “Successor” employer—Duty to bargain with union. — A “successor” employer’s duty to bargain with union representing predecessor employer’s employees is not limited to situation where union was only recently certified before transition in employers, but where, as here, union was certified for more than a year before change in employers, it was entitled to presumption of majority support; evidence supported National Labor Relations Board’s conclusion that petitioner, who acquired textile plant after it had gone out of business, was “successor” employer based upon determination of “substantial continuity” between petitioner and predecessor plant owner despite a 7-month hiatus between predecessor’s going out of business and petitioner’s start-up; NLRB’s “substantial and representative complement” rule—which fixes date when a successor’s duty to bargain with predecessor’s employees’ union arises—is reasonable and was properly applied in this case, as was NLRB’s “continuing demand” rule whereby a union’s premature demand for bargaining continues in effect until successor acquires “substantial and representative complement” of employees that triggers it obligation to bargain. Fall River Dyeing & Finishing Corp. v. NLRB, p. 27. LABOR MANAGEMENT RELATIONS ACT, 1947. See Removal of Actions. LABOR UNIONS. See Civil Rights Act of 1866; Civil Rights Act of 1964; Judicial Review, 1; Labor. LAKEBED RIGHTS. See Federal-State Relations. LAND-USE REGULATIONS. See Constitutional Law, XL LAWYERS. See Attorneys; Civil Rights Attorney’s Fees Awards Act of 1976. LIBERTY INTERESTS IN PAROLE. See Constitutional Law, I, 2. 944 INDEX LIMITATION OF ACTIONS. See Civil Rights Act of 1866. LITTLE TUCKER ACT. See Jurisdiction. LOS ANGELES. See Constitutional Law, V, 1. LOSING PARTY’S LIABILITY FOR PREVAILING PARTY’S EXPERT WITNESS FEES. See Procedure. LOUISIANA. See Constitutional Law, IV; Extradition Act; Judgments. MAINE. See Pre-emption of State Law by Federal Law. MARRIAGE BY PRISONERS. See Constitutional Law, V, 3; VIII. Judicial Review, 3. MARYLAND. See Constitutional Law, II. MINOR WITNESSES. See Constitutional Law, I, 1; VII. MONTANA. See Constitutional Law, I, 2. MORTGAGES. See Property Taxes. NATIONAL LABOR RELATIONS ACT. See Labor; Pre-emption of State Law by Federal Law. NATIONAL LABOR RELATIONS BOARD. See Labor. NEGLIGENCE PENALTIES FOR UNDERPAYMENT OF INCOME TAXES. See Internal Revenue Code. NEW JERSEY. See Constitutional Law, VI; Judicial Review, 2. NEW MEXICO. See Compacts Between States. NEW YORK. See Constitutional Law, IX. OVERBREADTH. See Constitutional Law, V, 1. PARENTAL KIDNAPING ACT. See Extradition Act. PARENTS AND CHILDREN. See Extradition Act. PAROLE. See Constitutional Law, I, 2. PECOS RIVER COMPACT. See Compacts Between States. PENALTIES FOR UNDERPAYMENT OF INCOME TAXES. See Internal Revenue Code. PLANT CLOSING AS REQUIRING SEVERANCE PAY. See Preemption of State Law by Federal Law. PRE-EMPTION OF STATE LAW BY FEDERAL LAW. See also Constitutional Law, X; Removal of Actions. State severance pay statute—ERISA—NLRA. —Maine statute requiring employers, in event of a plant closing, to provide one-time severance pay- INDEX 945 PRE-EMPTION OF STATE LAW BY FEDERAL LAW-Continued. ment to employees not covered by an express contract providing for severance pay is not pre-empted by Employee Retirement Income Security Act of 1974, since it does not “relate to any employee benefit plan” under that statute’s pre-emption provision; nor is Maine statute pre-empted by National Labor Relations Act, since it does not intrude on collective bargaining but is a valid and unexceptional exercise of State’s police power compatible with NLRA. Fort Halifax Packing Co. v. Coyne, p. 1. PRESENTENCE REPORTS. See Constitutional Law, II. PREVAILING PARTIES. See Civil Rights Attorney’s Fees Awards Act of 1976; Procedure. PREVAILING PARTY’S ENTITLEMENT TO COST OF EXPERT WITNESS FEES. See Procedure. PRISONER MARRIAGES. See Constitutional Law, V, 3; VIII; Judicial Review, 3. PRISONERS. See Civil Rights Attorney’s Fees Awards Act of 1976; Constitutional Law, I, 2; VI; Judicial Review, 2. PRISONER-TO-PRISONER CORRESPONDENCE. See Constitutional Law, V, 3; VIII; Judicial Review, 3. PRISON RULES AND REGULATIONS. See Constitutional Law, V, 3; VI; VIII; Judicial Review, 2, 3. PRISONS. See Civil Rights Attorney’s Fees Awards Act of 1976; Constitutional Law, VI; Judicial Review, 2. PROCEDURE. See Hague Convention on the Taking of Evidence Abroad in Civil or Commercial Matters; Judgments. Costs—Expert witness fees.— Absent explicit statutory or contractual authorization for taxation of expenses of a litigant’s witness as costs, federal courts are bound by limitations of 28 U. S. C. §§ 1821(b)—which states that a witness “shall be paid” a fee of $30 per day for court attendance— and 1920—which provides that a federal court “may tax” specified items, including witness fees, as costs against losing party; thus, it was not proper for District Court to award, as part of costs of defendant prevailing parties in antitrust action, an amount for expert witness fees in excess of § 1821’s $30-per-day limit, but District Court properly refused to order plaintiff losing party in civil rights action to reimburse defendant for its expert witness fees to extent they exceeded $30-per-day limit. Crawford Fitting Co. v. J. T. Gibbons, Inc., p. 437. PROPERTY TAXES. “Ginnie Maes”—Exemption from state property tax. —Appellant’s “Ginnie Maes”—instruments under which issuing private financial institu- 946 INDEX PROPERTY TAXES—Continued. tion is obliged to make timely payments of principal and interest and Government National Mortgage Association guarantees that such payments will be made as scheduled—were not exempt from Illinois state property taxes either under Revised Statutes § 3701, which exempts from state taxation “all stocks, bonds, Treasury notes, and other obligations of the United States,” or under constitutional principle of intergovernmental tax immunity. Rockford Life Ins. Co. v. Illinois Dept, of Revenue, p. 182. RACIAL DISCRIMINATION. See Civil Rights Act of 1866; Civil Rights Act of 1964. RACKETEER INFLUENCED AND CORRUPT ORGANIZATIONS ACT. See Federal Arbitration Act. RAILROADS. See Judicial Review, 1. RELEASE FROM PRISON. See Constitutional Law, I, 2. REMOVAL OF ACTIONS. Breach of employment contract—State-law complaint.— Where, after being hired in positions covered by employer’s collective-bargaining agreement with a union, eventually assuming positions outside bargaining unit, subsequently being downgraded to unionized positions, allegedly assured to be temporary, and then being notified that they would be laid off when plant was closed, respondents brought action in California state court based solely on state law, alleging that employer had breached their individual employment contracts, action was not removable to Federal District Court; respondents’ state-law contract claims were not “completely preempted” by claims under § 301 of Labor Management Relations Act, 1947, which confers federal jurisdiction as to suits for violations of collectivebargaining agreements and governs claims founded directly on rights created by such agreements and claims substantially dependent on analysis of such agreements but says nothing about content or validity of individual employment contracts such as those respondents alleged were breached. Caterpillar Inc. v. Williams, p. 386. RETROACTIVITY OF STATUTES OF LIMITATIONS. See Civil Rights Act of 1866. RIGHT TO CONFRONT WITNESSES. See Constitutional Law, I, 1; VII. RIGHT TO CROSS-EXAMINE. See Constitutional Law, I, 1; VII. RIPARIAN RIGHTS. See Compacts Between States. RULES GOVERNING ADMISSION TO FEDERAL DISTRICT COURT BAR. See Attorneys. INDEX 947 RULES OF CIVIL PROCEDURE. See Hague Convention on the Taking of Evidence Abroad in Civil or Commercial Matters. SCHOOLS. See Constitutional Law, IV; Judgments. SEARCHES AND SEIZURES. See Constitutional Law, IX. SECURITIES EXCHANGE ACT OF 1934. See Federal Arbitration Act. SENTENCING GUIDELINES. See Constitutional Law, III. SENTENCING PROCEEDINGS. See Constitutional Law, II. SEVERANCE PAY ON PLANT CLOSING. See Pre-emption of State Law by Federal Law. “SEVERITY REGULATION” AS TO DISABILITY UNDER SOCIAL SECURITY ACT. See Social Security Act. SIXTH AMENDMENT. See Constitutional Law, I, 1; VII. SOCIAL SECURITY ACT. Disability benefits—“Severity regulation.”—Regulation providing that if claimaint for disability benefits under Act does not have impairment that significantly limits his ability to do basic work activities, it will be found that he does not have a severe impairment and is therefore not disabled, and that his age, education, and work experience will not be considered, is valid on its face under Act’s language and legislative history; regulation increases efficiency and reliability of disability evaluation process by identifying at an early stage those claimants whose medical impairments are so slight that it is unlikely they would be found to be disabled even if their age, education, and experience were taken into account. Bowen v. Yuck-ert, p. 137. SODOMY. See Constitutional Law, I, 1; VII. STANDARD OF REVIEW IN DETERMINING CONSTITUTIONALITY OF PRISON REGULATIONS. See Judicial Review, 2, 3. STANDING TO SUE. Arbitration agreement—Enforcement—Appeal.— Resolution of whether appellants, appellee’s former employer and its employees, lacked “standing” to enforce agreement to arbitrate any dispute with employer was not prerequisite to their having standing under Article III of Constitution to maintain appeal to Supreme Court from lower court’s refusal to cofhpel arbitration in appellee’s breach-of-contract suit against appellants. Perry v. Thomas, p. 483. STATE PROPERTY TAXES. See Property Taxes. STATUTES OF LIMITATIONS. See Civil Rights Act of 1866. 948 INDEX STOLEN PROPERTY. See Constitutional Law, IX. STRICT SCRUTINY ANALYSIS OF PRISON RULES. See Judicial Review, 3. “SUCCESSOR” EMPLOYER’S DUTY TO BARGAIN WITH UNION. See Labor. SUMMARY JUDGMENTS. See Judgments. SUPREMACY CLAUSE. See Constitutional Law, X. SUPREME COURT. See Abstention; Constitutional Law, XI; Standing to Sue. TAKING OF REAL PROPERTY. See Constitutional Law, XI. TAXATION. See Internal Revenue Code; Property Taxes. TEACHING OF THEORY OF EVOLUTION. See Constitutional Law, IV; Judgments. “TEMPORARY” REGULATORY TAKING OF REAL PROPERTY. See Constitutional Law, XI. TEXAS. See Compacts Between States. THEORY OF “CREATION SCIENCE.” See Constitutional Law, IV; Judgments. THEORY OF EVOLUTION. See Constitutional Law, IV; Judgments. TRACKAGE RIGHTS OF RAILROADS. See Judicial Review, 1. TUCKER ACT. See Jurisdiction. UNDERPAYMENT OF INCOME TAXES. See Internal Revenue Code. UNFAIR LABOR PRACTICES. See Labor. UNIONS. See Civil Rights Act of 1866; Civil Rights Act of 1964; Judicial Review, 1; Labor. UNITED STATES COURTS OF APPEALS. See Jurisdiction. UNITED STATES DISTRICT COURTS. See Attorneys; Hague Convention on the Taking of Evidence Abroad in Civil or Commercial Matters; Judgments; Removal of Actions. UTAH. See Federal-State Relations. VICTIM IMPACT STATEMENTS. See Constitutional Law, II. WARRANTLESS SEARCHES AND SEIZURES. See Constitutional Law, IX. INDEX 949 WATER RIGHTS. See Compacts Between States; Federal-State Relations. WITNESSES. See Constitutional Law, I, 1; VIL WITNESS FEES. See Procedure.