UNITED STATES REPORTS VOLUME 437 CASES ADJUDGED IN THE SUPREME COURT AT OCTOBER TERM, 1977 June 14 Through June 23, 1978 HENRY PUTZEL, jr. REPORTER OF DECISIONS UNITED STATES GOVERNMENT PRINTING OFFICE WASHINGTON : 1980 For sale by the Superintendent of Documents, U.S. Government Printing Office Washington, D.C. 20402 Erratum 425 U. S. 149, n. 5, line 12: “demonstate” should be “demonstrate”. li JUSTICES OF THE SUPREME COURT DURING THE TIME OF THESE REPORTS WARREN E. BURGER, Chief Justice. WILLIAM J. BRENNAN, Jr., Associate Justice. POTTER STEWART, Associate Justice. BYRON R. WHITE, Associate Justice. THURGOOD MARSHALL, Associate Justice. HARRY A. BLACKMUN, Associate Justice. LEWIS F. POWELL, Jr., Associate Justice. WILLIAM H. REHNQUIST, Associate Justice. JOHN PAUL STEVENS, Associate Justice. retired STANLEY REED, Associate Justice. WILLIAM 0. DOUGLAS, Associate Justice. OFFICERS OF THE COURT GRIFFIN B. BELL, Attorney General. WADE H. McCREE, Jr., Solicitor General. MICHAEL RODAK, Jr., Clerk. HENRY PUTZEL, jr., Reporter of Decisions. ALFRED WONG, Marshal. ROGER F. JACOBS, Librarian. in 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, Warren E. Burger, 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, Warren E. Burger, Chief Justice. For the Fifth Circuit, Lewis F. Powell, Jr., Associate Justice. For the Sixth Circuit, Potter Stewart, Associate Justice. For the Seventh Circuit, John Paul Stevens, Associate Justice. For the Eighth Circuit, Harry A. Blackmun, Associate Justice. For the Ninth Circuit, William H. Rehnquist, Associate Justice. For the Tenth Circuit, Byron R. White, Associate Justice. December 19, 1975. (For next previous allotment, see 404 U. S., p. v.) IV TABLE OF CASES REPORTED Note: All undesignated references herein to the United States Code are to the 1970 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 Abrams v. United States.......................................... 903 Acting Commissioner of Patents and Trademarks v. Flook......... 584 Alabama; Williams v............................................ 908 Alexander v. U. S. Dept, of Housing and Urban Development...... 903 All Island Delivery Service, Inc. v. United States............... 911 American Broadcasting Cos. v. Writers Guild of America......... 411 ARCO Pipe Line Co. v. United States.............................. 902 Arizona; Mincey v................................................ 385 Ashland Oil, Inc. v. Governor of Maryland........................ 117 Assistant Circuit Attorney of St. Louis; Taylor v................ 908 Associated Milk Producers, Inc. v. United States................. 905 Association. For labor union, see name of trade. Association of Bar of City of New York; Kaufman v................ 904 Association of Motion Picture & TV Producers v. Writers Guild... 411 Bair; Moorman Manufacturing Co. v............................. 267 Bamond v. New York............................................... 912 Banks v. U. S. Dept, of Housing and Urban Development.......... 911 Beal v. Franklin................................................. 902 Bensor v. United States.......................................... 906 Beth Israel Hospital v. Labor Board............................. 483 Black; Bonnell v................................................. 907 Bonnell v. Black................................................. 907 BP Pipelines, Inc. v. United States.............................. 902 Bradford v. United States........................................ 908 Bretz v. Crist................................................... 911 Bretz; Crist v.................................................... 28 Brewster v. New York............................................. 907 Briggs; Goodwin v................................................ 904 Burks v. United States............................................. 1 Butler; Fennell v................................................ 906 VI TABLE OF CASES REPORTED Page Buttorff v. United States....................................... 906 Cadillac Overall Supply Co. v. United States.................... 903 California; Gray v.............................................. 908 California; Mertz v............................................. 908 California; Perlman v.......................................... 904 California; Sullivan v.......................................... 908 California v. Texas............................................. 601 Calvert Fire Insurance Co.; Will v.............................. 655 Cameron v. United States........................................ 906 Cane v. Kentucky................................................ 906 Carter; MAPCO, Inc. v........................................... 904 Carter v. Roberts............................................... 902 Chambers v. Oklahoma............................................ 908 Cheshire v. United States....................................... 907 Chicago Health Clubs, Inc. v. Labor Board.................... 904 City. See name of city. Civil Service Comm’n; Epstein v............................... 911 Cole; Harris v.................................................. 903 Coleman v. Virginia............................................. 911 Commerce Bank of Kansas City; Conrad v.......................... 912 Commissioner; Furrer v......................................... 903 Commissioner; Lewin v........................................... 904 Commissioner; Nasim v........................................... 907 Commissioner, Dept, of Labor of Alaska; Hicklin v.............. 518 Commissioner of Internal Revenue. See Commissioner. Commonwealth. See name of Commonwealth. Conrad v. Commerce Bank of Kansas City.......................... 912 Conrad v. First State Bank & Trust Co........................... 912 Continental Oil Co. v. Governor of Maryland..................... 117 Coopers & Lybrand v. Livesay................................... 463 Corrections Commissioner. See name of commissioner. County. See name of county. Crist v. Bretz................................................... 28 Crist; Bretz v.................................................. 911 Davis; Los Angeles County v.................................. 903 Davis v. United States.......................................... 906 Delaware; Hamilton v............................................ 907 Denney v. Texas................................................. 911 Department of Housing and Urban Development; Alexander v.... 903 Department of Housing and Urban Development; Banks v........... 911 Department of Institutions & Agencies; Essex Welfare Bd. v..... 910 Diaz v. United States........................................... 906 Director of Children & Family Servs. of Ill.; Yesterday’s Children v.. 904 TABLE OF CASES REPORTED VII Page Director of penal or correctional institution. See name of director. Director of Revenue of Iowa; Moorman Manufacturing Co. v....... 267 District Judge. See U. S. District Judge. Douglas Oil Co. of California v. Petrol Stops Northwest......... 902 Drummond v. Fulton County Dept, of Family Services.............. 910 Dunn; Perrin v.................................................. 910 East Baton Rouge Parish v. Pierson.............................. 905 Eastex, Inc. v. Labor Board.................................... 556 Eaton v. United States.......................................... 901 Epstein v. Civil Service Comm’n................................. 911 Essex County Welfare Bd. v. Dept, of Institutions & Agencies... 910 Exxon Corp. v. Governor of Maryland............................ 117 Exxon Pipeline Co. v. United States............................. 902 Federal Home Loan Bank Bd.; Mechanic’s Building & Loan Co. v. 904 Fennell v. Butler............................................... 906 Ferrara v. United States........................................ 907 Ferri v. United States.......................................... 907 Field v. United States.......................................... 901 Finney; Hutto v................................................. 678 First Federal Savings & Loan Assn. v. Tax Comm’n of Mass....... 255 First State Bank & Trust Co.; Conrad v.......................... 912 Flook; Parker v................................................. 584 Florida; Rogers v............................................... 906 Franklin; Beal v............................................. 902 Friedman; Neville v............................................. 903 Fulton County Dept, of Family Services; Drummond v.......... 910 Fulton County Dept, of Family Services; Hill v................ 910 Furrer v. Commissioner.......................................... 903 Gardner v. Westinghouse Broadcasting Co........................ 478 General Motors Assembly Division; Johnson v.................. 905 Godbout v. Norton............................................. 901 Golz v. Illinois................................................ 905 Goodwin v. Briggs............................................... 904 Governor of Maryland; Ashland Oil, Inc. v....................... 117 Governor of Maryland; Continental Oil Co. v.................... 117 Governor of Maryland; Exxon Corp, v............................. 117 Governor of Maryland; Gulf Oil Corp, v.......................... 117 Governor of Maryland; Shell Oil Co. v..................... i ./J 117 Gray v. California.............................................. 908 Greene v. Massey................................................. 19 Greyhound Corp. v. Mt. Hood Stages, Inc...................... 322 Greyhound Corp. v. Pacific Trailways.......................... 322 Griffin v. San Bernardino Police Dept........................... 906 VIII TABLE OF CASES REPORTED Page Griffin v. United States......................................... 906 Gulf Oil Corp. v. Governor of Maryland........................... 117 Gust v. U. S. Customs Service.................................... 904 Guzman v. United States.......................................... 906 Hall v. Morgan................................................... 910 Hamilton v. Delaware............................................. 907 Hannah v. New Jersey............................................. 906 Harbin v. Interlake S. S. Co..................................... 905 Harbin; Interlake S. S. Co. v................................... 905 Harris v. Cole................................................... 903 Herman; Volkswagenwerk AG v...................................... 905 Hicklin v. Orbeck................................................ 518 Hill v. Fulton County Dept, of Family Services................... 910 Hill; Tennessee Valley Authority v............................... 153 Hutto v. Finney.................................................. 678 Illinois; Golz v................................................. 905 Institutionalized Juveniles; Secretary of Welfare of Pa. v...... 902 Interlake S. S. Co. v. Harbin.................................... 905 Interlake S. S. Co.; Harbin v.................................... 905 International. For labor union, see name of trade. Jackson; Morgan v................................................ 907 Jackson v. United States......................................... 907 John v. Mississippi.............................................. 634 John; United States v............................................ 634 Johnson v. General Motors Assembly Division...................... 905 Johnson v. United States..................................... 906,907 Joyce; Joyce Beverages, Inc. v............... J................. 905 Joyce Beverages, Inc. v. Joyce................................... 905 Kaufman v. Association of Bar of City of New York................ 904 Kennedy; Yesterday’s Children v.................................. 904 Kentucky; Cane v................................................. 906 King v. United States............................................ 907 Kirk v. United States............................................ 908 Kroger; Owen Equipment & Erection Co. v..................... 365 Labor Board; Beth Israel Hospital v.............................. 483 Labor Board; Chicago Health Clubs, Inc. v........................ 904 Labor Board; Eastex, Inc. v...................................... 556 Labor Board; Quick Pak, Inc. v................................... 903 Labor Board v. Robbins Tire & Rubber Co.......................... 214 Labor Board v. Writers Guild of America.......................... 411 Labor Union. See name of trade. La je v. R. E. Thomason General Hospital......................... 905 LaSalle National Bank; United States v........................... 298 TABLE OF CASES REPORTED IX Page Layton v. Pogue................................................ 911 Lewin v. Commissioner.......................................... 904 Lipscomb; Wise v............................................... 535 Livesay; Coopers & Lybrand v................................... 463 Local. For labor union, see name of trade. Los Angeles County v. Davis.................................... 903 Louisiana; Williams v.......................................... 907 Louisiana Public Serv. Comm’n; South Central Bell Telephone v.. 911 Louisville Fire Fighters Pension Fund; O’Connor v.............. 905 MAPCO, Inc. v. Carter.......................................... 904 Marquez v. United States..................................... 908 Maryland; Watson v............................................. 908 Massachusetts v. White......................................... 902 Massey; Greene v............................................... 19 Mayor of Dallas v. Lipscomb.................................... 535 Meader v. Montana............................................ 908 Mechanic’s Building & Loan Co. v. Federal Home Loan Bank Bd.. 904 Mertz v. California........................................ 908 Mincey v. Arizona.............................................. 385 Mississippi; John v............................................ 634 Mobil Alaska Pipeline Co. v. United States..................... 902 Montana; Meader v.............................................. 908 Moorman Manufacturing Co. v. Bair............................ 267 Morgan; Hall v............................................... 910 Morgan v. Jackson............................,................. 907 Mt. Hood Stages, Inc.; Greyhound Corp, v..................... 322 Nasim v. Commissioner.......................................... 907 National Labor Relations Board. See Labor Board. Neville v. Friedman............................................ 903 New Jersey; Hannah v.............................. ........... 906 New Jersey; Philadelphia v.................................... 617 New York; Bamond v......................................,...... 912 New York; Brewster v....................................... 907 Niagara Mohawk Power Corp. v. Public Service Comm’n of N. Y.. 901 Norton; Godbout v............................................ 901 O’Connor v. Louisville Fire Fighters Pension Fund.............. 905 Oklahoma; Chambers v.......................................... 908 Oppenheimer Fund, Inc. v. Sanders............................ 340 Orbeck; Hicklin v.............................................. 518 Owen Equipment & Erection Co. v. Kroger........................ 365 Pacific Trailways; Greyhound Corp, v........................ 322 Palmer v. United States......................,................. 907 Parish of East Baton Rouge v. Pierson.......................... 905 X TABLE OF CASES REPORTED Page Parker v. Flook................................................... 584 Pavone v. United States........................................... 904 Perlman v. California............................................. 904 Perrin v. Dunn.................................................... 910 Petrol Stops Northwest; Douglas Oil Co. of California v........... 902 Philadelphia v. New Jersey........................................ 617 Pierson; Parish of East Baton Rouge v......................... 905 Poehling; Taylor v................................................ 908 Pogue; Layton v................................................... 911 PPX Enterprises, Inc. v. Scepter Records, Inc..................... 905 President of United States; MAPCO, Inc. v......................... 904 Public Service Comm’n of N. Y.; Niagara Mohawk Power Corp. v. 901 Quick Pak, Inc. v. Labor Board.................................... 903 Rachal v. Superintendent of M. C. I. at Walpole................... 908 Radisich v. Radisich.............................................. 908 R. E. Thomason General Hospital; La je v.......................... 905 Robbins Tire & Rubber Co.; Labor Board v.......................... 214 Roberts; Carter v................................................. 902 Rogers v. Florida................................................. 906 Sanabria v. United States.......................................... 54 San Bernardino Police Dept.; Griffin v............................ 906 Sanders; Oppenheimer Fund, Inc. v................................. 340 Scepter Records, Inc.; PPX Enterprises, Inc. v.................... 905 Scott; United States v............................................. 82 Secretary of Housing and Urban Development v. Cole................ 903 Secretary of Welfare of Pa. v. Franklin........................... 902 Secretary of Welfare of Pa. v. Institutionalized Juveniles........ 902 Seidel v. Texas................................................... 903 Sheet Metal Workers v. Siebler Heating & Air Conditioning, Inc... 911 Shell Oil Co. v. Governor of Maryland............................. 117 Sherwin v. United States.......................................... 909 Siebler Heating & Air Conditioning, Inc.; Sheet Metal Workers v.. 911 Skidmore v. United States......................................... 908 Smith v. United States............................................ 906 South Central Bell Telephone v. Louisiana Public Serv. Comm’n... 911 Sullivan v. California............................................ 908 Superintendent of M. C. I. at Walpole; Rachal v................... 908 Superintendent of penal or correctional institution. See name or state title of superintendent. Tax Comm’n of Massachusetts; First Fed. Savings & Loan Assn. v.. 255 Taylor v. Poehling................................................ 908 Tennessee Valley Authority v. Hill............................... 153 Texas; California v............................................... 601 TABLE OF CASES REPORTED XI Page Texas; Denney v.................................................... 911 Texas; Seidel v......*............................................... 903 Thomas v. United States.............................................. 904 Thomason General Hospital; Laje v.................................... 905 Union. For labor union, see name of trade. United States; Abrams v............................................ 903 United States; All Island Delivery Service, Inc. v................. 911 United States; ARCO Pipe Line Co. v................................ 902 United States; Associated Milk Producers, Inc. v................... 905 United States; Bensor v............................................ 906 United States; BP Pipelines, Inc. v................................ 902 United States; Bradford v.......................................... 908 United States; Burks v...........................U........... 1 United States; Buttorff v.......................................... 906 United States; Cadillac Overall Supply Co. v....................... 903 United States; Cameron v........................................... 906 United States; Cheshire v.......................................... 907 United States; Davis v............................................. 906 United States; Diaz v.............................................. 906 United States; Eaton v........................................... 901 United States; Exxon Pipeline Co. v................................ 902 United States; Ferrara v........................................... 907 United States; Ferri v............................................ 907 United States; Field v............................................. 901 United States; Griffin v........................................... 906 United States; Guzman v............................................ 906 United States; Jackson v........................................... 907 United States v. John................................................ 634 United States; Johnson v....................................... 906,907 United States; King v.............................................. 907 United States; Kirk v.............................................. 908 United States v. LaSalle National Bank............................... 298 United States; Marquez v.......................................... 908 United States; Mobil Alaska Pipeline Co. v......................... 902 United States; Palmer v............................................ 907 United States; Pavone v............................................ 904 United States; Sanabria v........................................... 54 United States v. Scott................................................ 82 United States; Sherwin v........................................... 909 United States; Skidmore v........................................ 908 United States; Smith v............................................. 906 United States; Thomas v............................................ 904 United States; Wedel v............................................. 901 XII TABLE OF CASES REPORTED Page United States; White v........................................... 906 United States; Zenith Radio Corp, v.............................. 443 United States; Zepeda-Santana v.................................. 907 U. S. Customs Service; Gust v................................... 904 U. S. Dept, of Housing and Urban Development; Alexander v.... 903 U. S. Dept, of Housing and Urban Development; Banks v. 911 U. S. District Judge v. Calvert Fire Insurance Co............... 655 U. S. District Judge; Carter v................................... 902 U. S. District Judge; Volkswagenwerk AG v........................ 905 Virginia; Coleman v.............................................. 911 Volkswagenwerk AG v. Herman...................................... 905 Warden. See name of warden. Watson v. Maryland............................................... 908 Wedel v. United States........................................... 901 Westinghouse Broadcasting Co.; Gardner v......................... 478 White; Massachusetts v........................................... 902 White v. United States........................................... 906 Will v. Calvert Fire Insurance Co................................ 655 Williams v. Alabama............................................. 908 Williams v. Louisiana............................................ 907 Wise v. Lipscomb................................................. 535 Writers Guild of America; American Broadcasting Cos. v........ 411 Writers Guild of America; Assn, of Motion Picture Producers v... 411 Writers Guild of America; Labor Board v.......................... 411 Yesterday’s Children v. Kennedy.................................. 904 Zenith Radio Corp. v. United States.............................. 443 Zepeda-Santana v. United States.................................. 907 TABLE OF CASES CITED Page Abel v. United States, 362 U. S. 217 312 Abney v. United States, 431 U. S. 651 468 Abood v. Detroit Bd. of Ed., 431 U. S. 209 566 Abrahamson Chrysler-Plymouth v. NLRB, 561 F. 2d 63 219 Adams Mfg. Co. v. Storen, 304 U. S. 307 289 Adelman v. Nordberg Mfg. Co., 6 F. R. D. 383 358 Aetna Cas. Co. v. Quarles, 92 F. 2d 321 671,672 Aetna State Bank v. Altheimer, 430 F. 2d 750 659 Aldinger v. Howard, 427 U. S. 1 370,372, 373, 375, 376, 379-382 Allegheny County v. Frank Mashuda Co., 360 U. S. 185 673 Allied Stores of Ohio v. Bowers, 358 U. S. 522 291 Altex Ready Mixed Concrete Corp. v. NLRB, 542 F. 2d 295 566 Alyeska Pipeline Service Co. v. Wilderness Society, 421 U. S. 240 689, 697,707,715 American Automobile Ins. Co. v. Freundt, 103 F. 2d 613 671, 672 American Fire & Cas. Co. v. Finn, 341 U. S. 6 373,377 American Mut. Reinsurance Co. v. Calvert Fire Ins. Co., 52 Ill. App. 3d 922 660 American Pipe & Constr. Co. v. Utah, 414 U. S. 538 338,339 American Ship Building Co. v. NLRB, 380 U. S. 300 431, 433,442 American Tobacco Co. v. United States, 328 U. S. 781 16, 74 Page Antoine v. Washington, 420 U. S. 194 638 Apodaca v. Oregon, 406 U. S. 404 37,52 Arcambel v. Wiseman, 3 Dall. 306 707 Arizona v. Washington, 434 U. S. 497 32,35, 93, 100, 105, 108, 109 Arlington Coalition on Transp. v. Volpe, 458 F. 2d 1323 206 Armstrong Co. v. Nu-Enamel Corp., 305 U. S. 315 204 Asbell v. Kansas, 209 U. S. 251 628,631 Ashe v. Swenson, 397 U. S. 436 73 Aspin v. Department of Defense, 160 U. S. App. D. C. 231 228 Au & Son v. NLRB, 538 F. 2d 80 219,240,250 Austin v. New Hampshire, 420 U. S. 656 523,524 Baker v. Baker, Eccles & Co., 242 U. S. 394 607 Baldwin v. G. A. F. Seelig, Inc., 294 U. S. 511 152,288,296,534 622, 624, 627, 628, 631 Baldwin v. Montana Fish & Game Comm’n, 436 U. S. 371 523,526, 529,532 Balistrieri v. Holtzman, 52 F. R. D. 23 353 Ball v. United States, 140 U. S. 118 88 Ballew v. Georgia, 435 U. S. 223 37,909 Baltimore Contractors v. Bod-inger, 348 U. S. 176 476,480 Bankers Life & Cas. Co. v. Holland, 346 U. S. 379 661,662 Barceloneta Shoe Corp. v. Compton, 271 F. Supp. 591 226, 233 XIII XIV TABLE OF CASES CITED Page Bargain Car Wash v. Standard Oil Co., 466 F. 2d 1163 129 Bass, Ratcliff & Gretton v. State Tax Comm’n, 266 U. S. 271 273,274,287,291 Bates v. Clark, 95 U. S. 204 649 Baylor Univ. Medical Center v. NLRB, 188 U. S. App. D. C. 109 489 Beacon Theatres v. Westover, 359 U. S. 500 666,677 Beecher v. Alabama, 389 U. S. 35 398 Bell v. United States, 349 U. S. 81 70 Benton v. Maryland, 395 U. S. 784 22,24,25,32 Berger v. Superior Court, 110 Ariz. 281 289,403 Berland v. Mack, 48 F. R. D. 121 354,363 Best v. Yerkes, 247 Iowa 800 368 Best & Co. v. Maxwell, 311 U. S. 454 136,147,288 Bethlehem Shipbuilding Corp. v. NLRB, 114 F. 2d 930 566 Bibb v. Navajo Freight Lines, 359 U. S. 520 294,295 Blackburn v. Alabama, 361 U. S. 199 398,401,408,409 Blake v. McClung, 172 U. S. 239 524 Blockburger v. United States, 284 U. S. 299 70 Blue Chip Stamps v. Manor Drug Stores, 421 U. S. 723 202, 330 Boren v. Tucker, 239 F. 2d 767 306,319 Boston Stock Exch. v. State Tax Comm’n, 429 U. S. 318 282, 283 288 Boulden v. Holman, 394 U. S. 478 401 Bourget v. Government Employees Ins. Co., 313 F. Supp. 367 352 Bowman v. Chicago & N. R. Co., 125 U. S. 465 622, 628, 631 Bradley v. Milliken, 540 F. 2d 229 692 Bradley v. Richmond School Bd., 416 U. S. 696 695 Page Braverman v. United States, 317 U. S. 49 70,73,74 Breard v. Alexandria, 341 U. S. 622 127 Breed v. Jones, 421 U. S. 519 15 Brillhart v. Excess Ins. Co., 316 U. S. 491 662, 664,667,670-672 Bristol-Myers Co. v. FTC, 138 U. S. App. D. C. 22 246 Brooks v. Dewar, 313 U. S. 354 210 Brown v. Dellinger, 355 S. W. 2d 742 580 Brown v. Mississippi, 297 U. S. 278 401 Brown v. Ohio, 432 U. S. 161 69, 70,72,74 Brunson v. Board of Trustees, 311 F. 2d 107 479 Bryan v. United States, 338 U. S. 552 5-7,9,10,12,13,24 Buck v. Kuykendall, 267 U. S. 307 624 Burbank v. Lockheed Air Terminal, 411 U. S. 624 621 Burks v. United States, 437 U. S. 1 24-27,32,91,97,107,113 Burnet v. Coronado Oil & Gas Co., 285 U. S. 393 101 Burnett v. New York Central R. Co., 380 U. S. 424 338,377 Bums v. Richardson, 384 U. S. 73 539-541,547-549,553 Burton v. United States, 202 U. S. 344 70 Burton Constr. & Shipbldg. Co. v. Broussard, 154 Tex. 50 580 Capital Cities Communications v. NLRB, 409 F. Supp. 971 253 Carmen’s Petition, In re, 165 F. Supp. 942 651 Carroll v. United States, 354 U. S. 394 79 Carsey v. United States, 129 U. S. App. D. C. 205 52,105 Cary v. Curtis, 3 How. 236 372 Catlin v. United States, 324 U. S. 229 467 C. D. R. Enterprises v. Board of Ed., 412 F. Supp. 1164 531 TABLE OF CASES CITED xv Page Center for National Policy Review v. Weinberger, 163 U. S. App. D. C. 368 228,229,248 Central Hardware Co. v. NLRB, 407 U. S. 539 491, 515, 563,572,582 Central Tablet Mfg. Co. v. United States, 417 U. S. 673 65 Chapman v. California, 386 U. S. 18 398 Chapman v. Meier, 420 U. S. 1 539-541,554 Charlotte-Mecklenburg Hospital Auth. v. Perry, 571 F. 2d 195 220 Chatfield, In re, 545 F. 2d 152 590,594,600 Chicago v. Vaccarro, 408 Ill. 587 264 Chicago Typographical Union v. NLRB, 176 U. S. App. D. C. 240 422,426,429 Chimel v. California, 395 U. S. 752 391,393,394 Chitto v. United States, 133 Ct. Cl. 643 638,642 Choctaw Nation v. Oklahoma, 397 U. S. 620 638 Choctaw Nation v. United States, 119 U. S. 1 638,643 Choctaw Nation v. United States, 81 Ct. Cl. 1 638 Chrapliwy v. Uniroyal, Inc., 17 Fed. Rules Serv. 2d 719 357 Christian v. United States, 432 U. S. 910 910 Christiansburg Garment Co. v. EEOC, 434 U. S. 412 689 Church of Holy Trinity v. United States, 143 U. S. 457 187, 204 City. See name of city. City Bank Farmers Trust Co. v. Schnader, 293 U. S. 112 602 Clark v. Watchie, 513 F. 2d 994 675 Clewis v. Texas, 386 U. S. 707 401 Climax Molybdenum Co. v. NLRB, 539 F. 2d 63 220,251 Coal Co. v. Blatchford, 11 Wall. 172 373 Cobbledick v. United States, 309 U. S. 323 467 Page Cochran v. United States, 157 U. S. 286 66 Cochrane v. Deener, 94 U. S. 780 588 Cohen v. Beneficial Loan Corp., 337 U. S. 541 347,468,471 Cohens v. Virginia, 6 Wheat. 264 462,606,615 Coleman v. Doe, 12 Miss. 40 642 Collett, Ex parte, 337 U. S. 55 184 Colorado River Conservation Dist. v. United States, 424 U. S. 800 657,663, 664, 667-670, 672-674 Commissioner of Corps. & Taxation v. Flaherty, 306 Mass. 461 261 Committee for Nuclear Responsibility v. Seaborg, 149 U. S. App. D. C. 380 190 Commonwealth. See also name of Commonwealth. Commonwealth v. Boyle, 346 Mass. 1 59 Commonwealth v. Cook, 6 Serg. & R. 577 34,45 Commonwealth v. Fontain, 127 Mass. 452 264 Commonwealth v. Gould, 78 Mass. 171 14 Commonwealth v. Hart, 149 Mass. 7 45 Commonwealth v. Wade, 34 Mass. 395 45 Complete Auto Transit v. Brady, 430 U. S. 274 292 Connecticut v. Massachusetts, 282 U. S. 660 613 Connor v. Finch, 431 U. S. 407 539-542,551,554 Connor v. Johnson, 402 U. S. 690 540,542,554 Connor v. Waller, 421 U. S. 656 542 Connor v. Williams, 404 U. S. 549 540,553,554 Conway and Lynch v. Queen, 7 Ir. 149 43 Cooley v. Board of Wardens, 12 How. 299 128 Coolidge v. New Hampshire, 403 U. S. 443 390,393,395 XVI TABLE OF CASES CITED Page Corson v. Maryland, 120 U. S. 502 289 Cotier v. Inter-County Orthopaedic Assn., 526 F. 2d 537 674, 676 Couch v. United States, 409 U. S. 322 309 County. See name of county. Crane v. New York, 239 U. S. 195 531 Crawford v. United States, 126 U. S. App. D. C. 156 17 Crescent Oil Co. v. Mississippi, 257 U. S. 129 150 Crist v. Bretz, 437 U. S. 28 27,92 Crooks v. Harrelson, 282 U. S. 55 187,204 Crow Dog, Ex parte, 109 U. S. 556 649,651 Curley v. United States, 81 U. S. App. D. C. 389 16 Daily Mirror, Inc. v. New York News, 533 F. 2d 53 67 Dairyland Ins. Co. v. Mumert, 212 N. W. 2d 436 368 Danley v. United States, 424 U. S. 929 910 Darnell & Son v. Memphis, 208 U. S. 113 288 Davis v. North Carolina, 384 U. S. 737 398,409 Davis v. State, 236 Md. 389 392 Davis v. United States, 160 U. S. 469 3,98 Dayton Bd. of Educ. v. Brink-man, 433 U. S. 406 704,710,713 Dean Milk Co. v. Madison, 340 U. S. 349 126,135-137,146,149 De Canas v. Bica, 424 U. S. 351 132 De Castelet, In re, 562 F. 2d 1236 600 Deepsouth Packing Co. v. Lai-tram Corp., 406 U. S. 518 596 Denver & R. G. W. R. Co. v. United States, 387 U. S. 485 336 Department of Air Force v. Rose, 425 U. S. 352 220 Dery v. Wyer, 265 F. 2d 804 375 Deutsch, In re, 553 F. 2d 689 600 Di Santo v. Pennsylvania, 273 U. S. 34 289 Page Ditlow v. Brinegar, 161 U. S. App. D. C. 154 228,229,248 Domaco Venture Capital Fund v. Teltronics Services, 551 F. 2d 508 472 Donaldson v. United States, 400 U. S. 517 299,304,306-308,312, 313, 315, 316, 319-321 Donnelly v. United States, 228 U. S. 243 649 Douglas v. New Haven R. Co., 279 U. S. 377 524 Douglas v. Seacoast Products, 431 U. S. 265 627 Downs v. United States, 187 U. S. 496 449,459,461,462 Downs v. United States, 113 F. 144 460 Downum v. United States, 372 U. S. 734 35,48,92,93 Dreyer v. Illinois, 187 U. S. 71 45 Dulin v. Lillard, 91 Va. 718 45,46 Duncan v. Louisiana, 391 U. S. 145 53 Dungan v. Morgan Drive- Away, 570 F. 867 334,336 D’Youville Manor, Inc. v. NLRB, 526 F. 2d 3 237 East Carroll Parish School Bd. v. Marshall, 424 U. S. 636 539, 540, 545, 546, 548-554 Eastex, Inc. v. NLRB, 437 U. S. 556 515,516 Ebeling v. Morgan, 237 U. S. 625 70 Eccles v. People’s Bank, 333 U. S. 426 213 Econo-Car Inti v. Antilles Car Rentals, 61 F. R. D. 8 353 Edelman v. Jordan, 415 U. S. 651 608, 615,616, 690, 692, 695, 698, 700-706, 708, 709, 715, 717 Edwards v. California, 314 U. S. 160 526,627 Eibel Process Co. v. Minnesota Paper Co., 261 U. S. 45 590, 600 Eisen v. Carlisle & Jacquelin, 417 U. S. 156 347,349, 350, 355-363, 468, 471 Eisen v. Carlisle & Jacquelin, 370 F. 2d 119 465 TABLE OF CASES CITED XVII Page Eisen v. Carlisle & Jacquelin, 479 F. 2d 1005; 54 F. R. D. 565 344 Electrical Workers v. NLRB, 366 U. S. 667 575 Electrical Workers v. NLRB, 159 U. S. App. D. C. 242 423 Electrical Workers v. NLRB, 159 U. S. App. D. C. 272 424, 434-436 Electromec Design & Develop- ment Co. v. NLRB, 409 F. 2d 631 237 Employees v. Missouri Public Health Dept., 411 U. S. 279 695, 697, 698, 701, 705, 706, 709,710 Enelow v. New York Life Ins. Co., 293 U. S. 379 481 Enterprise Industries v. Texas Co., 136 F. Supp. 420 129 Environmental Defense Fund v. Froehlke, 473 F. 2d 346 190 EPA v. Mink, 410 U. S. 73 220, 226,235,242,245 Ernst & Ernst v. Hochfelder, 425 U. S. 185 330 Estate. See name of estate. Estelle v. Gamble, 429 U. S. 97 685 Ettelson v. Metropolitan Life Ins. Co., 317 U. S. 188 481 Evco v. Jones, 409 U. S. 91 289 Ex parte. See name of party. Fahey, Ex parte, 332 U. S. 258 667 Fairmont Creamery Co. v. Minnesota, 275 U. S. 70 696, 697,707 Fawvor v. Texaco, Inc., 546 F. 2d 636 367 FAA Administrator v. Robertson, 422 U. S. 255 248 FTC v. Sun Oil Co., 371 U. S. 505 129 Felicia v. United States, 495 F. 2d 353 636 Ferguson v. Skrupa, 372 U. S. 726 124 Finch v. United States, 433 U. S. 676 102 First Fed. Savings v. Connelly, 142 Conn. 483 260 First Nat. Bank v. Dunham, 471 F. 2d 712 689 Page First Nat. Bank v. Maine, 284 U. S. 312 602, 607 Fisons, Ltd. v. United States, 458 F. 2d 1241 475 Fitzpatrick v. Bitzer, 427 U. S. 445 693,694,698,699,701- 703, 706, 708, 710, 717 Fleming v. McCurtain, 215 U. S. 56 638 Fleming v. Mohawk Wrecking Co., 331 U. S. Ill 210 Florida Power & Light v. Electrical Workers, 417 U. S. 790 421,422,424- 430, 436, 437, 439-442 Fong Foo v. United States, 369 U. S. 141 11,64,69, 75, 78, 90, 104, 107, 108 Ford Motor Co. v. Beauchamp, 308 U. S. 331 273,292 Ford Motor Co. v. Department of Treasury, 323 U. S. 459 717 Forman v. United States, 361 U. S. 416 8-10,14,17,24 Fortson v. Dorsey, 379 U. S. 433 541,555 Fort Wayne Corrugated Paper Co. v. NLRB, 111 F. 2d 869 564, 568 Foster v. Boise-Cascade, Inc., 20 Fed. Rules Serv. 2d 466 357 Foster Packing Co. v. Haydel, 278 U. S. 1 533,627 Francis v. Resweber, 329 U. S. 459 12 Frankel v. SEC, 460 F. 2d 813 233 Freeman, In re, 573 F. 2d 1237 600 Freeman v. Howe, 24 How. 450 375 Funk Bros. Seed Co. v. Kalo Co., 333 U. S. 127 591,599 Gaffney v. Cummings, 412 U. S. 735 539 Gates v. Dalton, 67 F. R. D. 621 355 Gates v. Henderson, 568 F. 2d 830 405 Gavieres v. United States, 220 U. S. 338 70 General Atomic Co. v. Felter, 434 U. S. 12 666 XVIII TABLE OF CASES CITED Page General Electric Co. v. Marvel Rare Metals Co., 287 U. S. 430 480,481 General Motors Corp. v. District of Columbia, 380 U. S. 553 274,275,287,292,294-296 General Motors Corp. v. Washington, 377 U. S. 436 281, 282,289 Georgia v. Pennsylvania R. Co., 324 U. S. 439 189 Gillespie v. United States Steel Corp., 379 U. S. 148 477 Glasser v. United States, 315 U. S. 60 17 Glen Falls Indemnity Co. v. United States ex rel. Westinghouse Elec. Supply Co., 229 F. 2d 370 375 Gompers v. Bucks Stove Co., 221 U. S. 418 691 Gore v. United States, 357 U. S. 386 70 Gori v. United States, 367 U.S. 364 34 Gosa v. Securities Inv. Co., 449 F. 2d 1330 472 Gottschalk v. Benson, 409 U. S. 63 585,587-591,593,599, 600 Granader v. Public Bank, 417 F. 2d 75 675 Graves v. Bames, 343 F. Supp. 704 555 Great A&P Tea Co. v. Cottrell, 424 U. S. 366 624 Greathouse v. Dem, 289 U. S. 352 213 Great Northern Life Ins. Co. v. Read, 322 U. S. 47 717 Green v. United States, 355 U. S. 184 11, 13, 15, 17, 25, 35, 43, 46, 47, 51, 64, 69, 87, 91, 95, 99, 101, 105 Green v. Wolf Corp., 406 F. 2d 291 472 Greene v. Massey, 437 U. S. 19 32 Greene County Planning Bd. v. FPC, 455 F. 2d 412 206 Greenwald v. Wisconsin, 390 U. S. 519 401,409,410 Gregg v. Georgia, 428 U. S. 153 911 Griffin v. McMann, 310 F. Supp. 72 696 Page G. S. Nicholas & Co. v. United States, 249 U. S. 34 459 Guy v. Baltimore, 100 U. S. 434 288 G&W Electric Specialty Co. v. NLRB, 360 F. 2d 873 567, 568 Gwin, White & Prince v. Hen- neford, 305 U. S. 434 289 Hackett v. General Host Corp., 455 F. 2d 618 465 Hagner v. United States, 285 U. S. 427 66 Hale v. Bimco Trading, 306 U. S. 375 288 Hall v. DeCuir, 95 U. S. 485 294 Halliburton Oil Well Co. v. Reily, 373 U. S. 64 136,288 Hampton v. United States, 425 U. S. 484 114 Hanf v. United States, 235 F. 2d 710 72 Hans Rees’ Sons v. North Carolina ex rel. Maxwell, 283 U. S. 123 274,287 Harris v. New York, 401 U. S. 222 397,398,410 Hartmann v. Scott, 488 F. 2d 1215 465 Harvey’s Wagon Wheel v. NLRB, 550 F. 2d 1139 219,254 Haynes v. Washington, 373 U. S. 503 398,401,408 Healy v. Ratta, 292 U. S. 263 377 Hecht Co. v. Bowles, 321 U. S. 321 193,194,211-213 Heim v. McCall, 239 U. S. 175 531 Hickman v. Taylor, 329 U. S. 495 254,351 H. L. Peterson Co. v. Applewhite, 383 F. 2d 430 375 Hodgson v. Adams Drug Co., 15 Fed. Rules Serv. 2d 828 358 Hoffman v. State, 20 Md. 425 45 Hood & Sons v. Du Mond, 336 U. S. 525 142,288,623,624 Hooley v. Red Carpet Corp., 549 F. 2d 643 473 Hopt v. Utah, 104 U. S. 631; 110 U. S. 574; 114 U. S. 488; 120 U. S. 430 14 TABLE OF CASES CITED XIX Page H. P. Hood & Sons v. Du Mond, 336 U. S. 525 142, 288,623, 624 Hudgens v. NLRB, 424 U. S. 507 504,505, 515, 563, 572, 573, 582 Hughes v. Alexandria Scrap Corp., 426 U. S. 794 127, 290, 295, 623, 627, 628 Hump Hairpin Co. v. Emmerson, 258 U. S. 290 291 Hunt v. Washington Apple Adv. Comm’n, 432 U. S. 333 126,128,136, 137, 146-148, 289, 624 Huron Cement Co. v. Detroit, 362 U. S. 440 130 lannelli v. United States, 420 U. S. 770 70 Illinois v. Somerville, 410 U. S. 458 33,36,38,48,89 I. M. Darnell & Son v. Memphis, 208 U. S. 113 288 Indianapolis v. Chase Nat. Bank, 314 U. S. 63 373,377 In re. See name of party. Insurance Co. v. Comstock, 16 Wall. 258 662 International Harvester Co. v. Evatt, 329 U. S. 416 286 International Harvester Co. v. Ruckelshaus, 155 U. S. App. D. C. 411 502 International Shoe Co. v. Shar-tel, 279 U. S. 429 291 Ireland’s Case, 7 How. St. Tr. 79 42 Jackson v. Bishop, 404 F. 2d 571 685 Jackson v. Bishop, 268 F. Supp. 804 681,682 Jackson v. Denno, 378 U. S. 368 397,398 J. D. Adams Mfg. v. Stören, 304 U. S. 307 289 Jeffers v. United States, 432 U. S. 137 63,75,76,78 Jelfo v. Hickok Mfg. Co., 531 F. 2d 680 474 Jenkins v. Blue Cross Hosp. Ins., 538 F. 2d 164 479 Johnson v. Haydel, 278 U. S. 16 627 Page Johnson v. United States, 333 U. S. 10 392,394,395 Jones v. Diamond, 519 F. 2d 1090 479 Jones v. Nelson, 484 F. 2d 1165 67 Jones v. Rath Packing Co., 430 U. S. 519 621 Joseph v. Chrysler Corp., 513 F. 2d 626 367 Kaiser Engineers v. NLRB, 538 F. 2d 1379 566,567 Katz v. United States, 389 U. S. 347 390,391 Keeble v. United States, 412 U. S. 205 636 Keerl v. Montana, 213 U. S. 135 45 Kellogg Co. v. NLRB, 457 F. 2d 519 567 Kenrose Mfg. Co. v. Fred Whitaker Co., 512 F. 2d 890 367,376 Kepner v. United States, 195 U.S. 100 11,34,35,45,89-91,104 Ker v. California, 374 U. S. 23 37 Kerotest Mfg. Co. v. C-O-Two Fire Equip. Co., 342 U. S. 180 663 Kerr v. U. S. District Court, 426 U. S. 394 661 King v. Kansas City Southern Industries, 479 F. 2d 1259 465, 469 King v. Kinloch, Fost. 16 43 King v. Perkins, Holt. 403 43 King v. United States, 125 U. S. App. D. C. 318 17 Kline v. Burke Constr. Co., 260 U. S. 226 372 Klinefelter v. Superior Court, 108 Ariz. 494 29 Korn v. Franchard Corp., 443 F. 2d 1301 471,472 Kutler v. United States, 423 U. S. 959 910 Labor Board. See NLRB. Labor Union. See name of trade. La Buy v. Howes Leather Co., 352 U. S. 249 665,676 XX TABLE OF CASES CITED Page Page Ladner v. United States, 358 U.S. 169 70 Landis v. North American Co., 299 U. S. 248 663,665 LASA Per L’Industria Del Marmo v. Alexander, 414 F. 2d 143 375 La Tourette v. McMaster, 248 U. S. 465 524 Laurens Fed. Savings v. South Carolina Tax Comm’n, 365 U. S. 517 258,263 Lecor, Inc. v. U. S. District Court, 502 F. 2d 104 676 Lee v. United States, 432 U. S. 23 63,68,71,75, 80, 86, 89, 92, 94, 97, 102, 103, 105, 107-110 Leh v. General Petroleum Corp., 382 U. S. 54 331,336 Lehman Bros. v. Schein, 416 U. S. 386 27 Le Roy v. Tatham, 14 How. 156 598 Letter Carriers v. Austin, 418 U. S. 264 576 Lincoln Fed. Labor Union v. Northwestern Iron & Metal, 335 U. S. 525 124 Lippa’s, Inc. v. Lenox, Inc., 305 F. Supp. 182 332 Lisenba v. California, 314 U. S. 219 408 Lockerty v. Phillips, 319 U. S. 182 372 Locomotive Engineers v. Missouri, K. & T. R. Co., 363 U. S. 528 213 Logan v. United States, 144 U. S. 263 45 Louisiana ex rel. Francis v. Res- weber, 329 U. S. 459 12 Lovato v. New Mexico, 242 U. S. 199 45,48 Ludwig v. Massachusetts, 427 U. S. 618 52 Lukenas v. Brvce’s Mountain Resort, 538 F. 2d 594 475 Luria Steel Corp. v. Ogden Corp., 484 F. 2d 1016 331,332 Lutheran Hospital v. NLRB, 564 F. 2d 208 489 Lynumm v. Illinois, 372 U. S. 528 398 Lyons v. Westinghouse Electric Co., 222 F. 2d 184 675 Machinists v. Street, 367 U. S. 740 566 Mackay Radio Co. v. Radio Corp., 306 U. S. 86 591,599 Mahala v. State, 18 Tenn. 532 45 Mahan v. Howell, 410 U. S. 315 540 Malinski v. New York, 324 U. S. 401 408 Malloy v. Hogan, 378 U. S. 1 37 Manchester Fed. Savings v. State Tax Comm’n, 105 N. H. 17 260 Mancusi v. DeForte, 392 U. S. 364 395 Marbury v. Madison, 1 Cranch 137 194 Maryland Cas. Co. v. Consumers Finance Service, 101 F. 2d 514 671,672 Massachusetts v. Missouri, 308 U. S. 1 606,609,610, 612,614 Masters, Mates & Pilots v. NLRB, 539 F. 2d 554 438 Maxey v. State, 251 Ind. 645 392 Maxwell, Ex parte, 11 Nev. 428 45 McClanahan v. Arizona Tax Comm’n, 411 U. S. 164 638 McClard v. United States, 386 F. 2d 495 16 McClellan v. Carland, 217 U. S. 268 662,663,670,677 McCready v. Virginia, 94 U. S. 391 528 McDonald v. State, 79 Wis. 651 45,46 McDonald v. United States, 335 U. S. 451 392,394 McGough v. First Arlington Nat. Bank, 519 F. 2d 552 675 McLaren v. Fleischer, 256 U. S. 477 458 Meachum v. Fano, 427 U. S. 215 686 Meador, In re, 16 F. Cas. 1294 311 Meat Drivers v. United States, 371 U. S. 94 623 TABLE OF CASES CITED XXI Page Memphis Steam Laundry Cleaner v. Stone, 342 U. S. 389 288 Mercantile Bank v. New York, 121 U. S. 138 258 Mercantile Nat. Bank v. Lang-deau, 371 U. S. 555 469 Meredith v. Winter Haven, 320 U. S. 228 670 Merrill Lynch, Pierce, Fenner, & Smith v. Ware, 414 U. S. 117 132 Michigan v. Tyler, 436 U. S. 499 391-393 Michigan Nat. Bank v. Michigan, 365 U. S. 467 259,264 Milberg v. Western Pacific R. Co., 443 F. 2d 1301 473 Miller v. Allstate Ins. Co., 66 Wash. 2d 871 264 Milliken v. Bradley, 418 U. S. 717 712 Milliken v. Bradley, 433 U. S. 267 687, 690-692, 704, 710-714 Mine Workers v. Gibbs, 383 U. S. 715 369-371, 373, 378, 379, 381-383 Minnesota v. Barber, 136 U. S. 313 136 Minnesota Mining & Mfg. Co. v. New Jersey Co., 381 U. S. 311 332,333 Miranda v. Arizona, 384 U. S. 436 396,397,410 Mississippi Power Co. v. Peabody Coal Co., 69 F. R. D. 558 353 Missouri v. Illinois, 200 U. S. 496 614 Missouri v. Iowa, 7 How. 660 695 Mitchell v. State, 42 Ohio St. 383 45,46 Mizell v. Attorney General, 442 F. Supp. 868 29 Monell v. New York City Dept, of Social Services, 436 U. S. 658 48,372,373, 701-703, 705, 708, 709 Monroe v. Pape, 365 U. S. 167 372, 701,709 Moore v. New York Cotton Exch., 270 U. S. 593 375,376 Page Morgan v. State, 13 Ind. 215 45,46 Morgan v. Virginia, 328 U. S. 373 294 Morgantown v. Royal Ins. Co., 337 U. S. 254 481 Morton v. Mancari, 417 U. S. 535 189,190,192 Mourning v. Family Publications Service, 411 U. S. 356 455 Mullaney v. Anderson, 342 U. S. 415 526,527 Murdock v. Memphis, 20 Wall. 590 522 Nadeau v. Helgemoe, 561 F. 2d 411 686 Nash v. Florida Industrial Comm’n, 389 U. S. 235 251 National Bellas Hess, Inc. v. Department of Revenue, 386 U. S. 753 273 NLRB v. Alamo Express Co., 430 F. 2d 1032 565 NLRB v. Allis-Chalmers Mfg. Co., 388 U. S. 175 440 NLRB v. Babcock & Wilcox Co., 351 U. S. 105 492, 505, 515, 516, 563, 570, 571, 581, 582 NLRB v. Baptist Hospital, 576 F. 2d 107 489,514, 517 NLRB v. Bretz Fuel Co., 210 F. 2d 392 567 NLRB v. Clement Bros., 407 F. 2d 1027 226,233 NLRB v. Erie Resistor Corp., 373 U. S. 221 433,501 NLRB v. Exchange Parts Co., 375 U. S. 405 251 NLRB v. Fansteel Metallurgical Corp., 306 U. S. 240 580, 581,583 NLRB v. Gissel Packing Co., 395 U. S. 575 251 NLRB v. Hardeman Garment Corp., 557 F. 2d 559 219,240 NLRB v. Insurance Agents, 361 U. S. 477 431 NLRB v. Interboro Contractors, 432 F. 2d 854 237 NLRB v. Iron Workers, 434 U. S. 335 430,431,500,508 NLRB v. J. G. Boswell Co., 136 F. 2d 585 565 XXII TABLE OF CASES CITED Page NLRB v. Leslie Metal Arts Co., 509 F. 2d 811 567 NLRB v. Lizdale Knitting Mills, 523 F. 2d 978 239 NLRB v. Magnavox Co., 415 U. S. 322 516,575 NLRB v. National Survey Service, 361 F. 2d 199 226,239 NLRB v. News Syndicate Co., 365 U. S. 695 431 NLRB v. Peter Cailler Kohler Swiss Chocolates Co., 130 F. 2d 503 566 NLRB v. Rex Disposables, 494 F. 2d 588 237 NLRB v. Safway Steel Scaffolds Co., 383 F. 2d 273 237 NLRB v. Savair Mfg. Co., 414 U. S. 270 251 NLRB v. Scrivener, 405 U. S. 117 239,251 NLRB v. Sears, Roebuck & Co., 421 U. S. 132 220,242,247 NLRB v. Truck Drivers, 353 U. S. 87 431,501 NLRB v. Valley Mold Co., 530 F. 2d 693 237 NLRB v. Vapor Blast Mfg. Co., 287 F. 2d 402 237,239 NLRB v. Washington Aluminum Co., 370 U. S. 9 567 NLRB v. Weingarten, Inc., 420 U. S. 251 500,508,566 National League of Cities v. Usery, 426 U. S. 833 699 National Leather Co. v. Massachusetts, 277 U. S. 413 291 National Railroad Passenger Corp. v. National Assn, of Railroad Passengers, 414 U. S. 453 188 National Wildlife Federation v. Coleman, 529 F. 2d 359 205 Neilson v. Hartford, Web. Pat. Cases 295 592 New England Medical Center Hospital v. NLRB, 548 F. 2d 377 219,241,253 New York v. Latrobe, 279 U. S. 421 291 New York v. New Jersey, 256 U. S. 296 614,615 Page Nicholas & Co. v. United States, 249 U. S. 34 459 Nielsen, In re, 131 U. S. 176 70 Nippert v. Richmond, 327 U.S. 416 288 Nissan Motor Corp., In re, 552 F. 2d 1088 348,355-357,359,360 Nolan v. State, 55 Ga. 521 45,46 Norfolk & W. R. Co. v. North Carolina ex rel. Maxwell, 297 U. S. 682 274 Norfolk & W. R. Co. v. State Tax Comm’n, 390 U. S. 317 273, 274 North Carolina v. Pearce, 395 U. S. 711 13 North Dakota v. Minnesota, 263 U. S. 583 695 Northern Pacific R. Co. v. United States, 356 U. S. 1 133 Northwest Airlines v. Minnesota, 322 U. S. 292 286 Northwestern States Portland Cement Co. v. Minnesota, 358 U. S. 450 283,289 Norwegian Nitrogen Products Co. v. United States, 288 U. S. 294 450 Novak v. Beto, 453 F. 2d 661 686 O’Berry v. Wainwright, 546 F. 2d 1204 405 O’Brian v. Commonwealth, 72 Ky. 333 46 Ohralik v. Ohio State Bar Assn., 436 U. S. 447 583 Oliphant v. Suquamish Indian Tribe, 435 U. S. 191 651 Oregon v. Hass, 420 U. S. 714 397, 398,410 O’Reily v. Morse, 15 How. 62 592,595,598 Ott v. Speedwriting Pub. Co., 518 F. 2d 1143 465,473 Palmore v. United States, 411 U. S. 389 372 Paramount Pictures v. Langer, 23 F. Supp. 890 124 Parden v. Terminal R. Co., 377 U. S. 184 701 Parker v. W. W. Moore & Sons, 528 F. 2d 764 367 Parkinson v. April Industries, 520 F. 2d 650 476 TABLE OF CASES CITED XXIII Page Parson v. Anderson, 354 F. Supp. 1060 392 Pasadena Bd. of Ed. v. Spangler, 427 U. S. 424 713,714 Patrick v. State, 227 A. 2d 486 392 Patterson v. New York, 432 U. S. 197 113 Paul v. Virginia, 8 Wall. 168 524 Penn Mutual Life Ins. Co. v. Austin, 168 U. S. 685 213 Pennsylvania v. Mimms, 434 U. S. 106 406,407 Pennsylvania v. West Virginia, 262 U. S. 553 532,533,627 People v. Brooks, 7 Ill. App. 3d 767 392 People v. Gardner, 62 Mich. 307 45,46 People v. Goodwin, 18 Johns. 187 41,45 People v. Hill, 12 Cal. 3d 731 392 People v. Mitchell, 39 N. Y. 2d 173 392 People v. Olcott, 2 Johns. Cas. 301 41 People v. Webb, 38 Cal. 467 45,46 Peterson Co. v. Applewhite, 383 F. 2d 430 375 Petillo v. New Jersey, 562 F. 2d 903 405 Petty v. Tennessee-Missouri Bridge Comm’n, 539 IT. S. 275 701 Pfizer Inc. v. Government of India, 434 U. S. 308 703 Phelps v. Oaks, 117 U. S. 236 375 Phelps Dodge Corp. v. NLRB, 313 U. S. 177 501,564,568 Pike v. Bruce Church, Inc., 397 U. S. 137 289,295, 624 Pipefitters v. United States, 407 U. S. 385 566 Pizitz, Inc. v. Volpe, 467 F. 2d 208 206 Pointer v. Texas, 380 U. S. 400 37 Popkin v. Wheelabrator-Frye, Inc., 20 Fed. Rules Serv. 2d 125 355,363 Posadas v. National City Bank, 296 U. S. 497 189 Poss v. NLRB, 565 F. 2d 654 253 Page Powe v. United States, 11 F. 2d 598 74 Powell v. United States, 135 U. S. App. D. C. 254 17 Power Reactor Co. v. Electri- cians, 367 U. S. 396 450 Price v. Georgia, 398 U. S. 323 13, 15 Price v. Lucky Stores, 501 F. 2d 1177 479 Queen v. Charlesworth, 1 B. & S. 460 42,43 Queen v. Drury, 3 Cox Crim. Cas. 544 14 Queen v. Winsor, 10 Cox C. C. 276 43 Rader v. Balfour, 440 F. 2d 4ßQ 339 Radio Officers v. NLRB, 347’ U. S. 17 433 Ragland v. Mueller, 460 F. 2d 1196 206 Railroad Co. v. Husen, 95 U. S. 465 631 Railway Clerks v. Allen, 373 U. S. 113 566 Ray v. Atlantic Richfield Co., 435 U. S. 151 621 Raymond Motor Transp. v. Rice, 434 U. S. 429 283, 289,294,623,624 Real Silk Hosiery Mills v. Portland, 268 U. S. 325 288 Reck v. Pate, 367 U. S. 433 401 Regional Rail Reorganization Act Cases, 419 U. S. 102 193,209 Reid v. Colorado, 187 U. S. 137 628 Reisman v. Caplin, 375 U. S. 440 304,306 Renegotiation Bd. v. Bannercraft Clothing Co., 415 U. S. 1 220,242 Republic Aviation v. NLRB, 324 U. S. 793 433, 491, 493, 495, 500, 501, 504, 508-515, 563, 568, 570-576, 581, 582 Reynolds v. Sims, 377 U. S. 533 541 Rice v. Olson, 324 U. S. 786 651 Rice v. Santa Fe Elevator Corp., 331 U. S. 218 621 XXIV TABLE OF CASES CITED Page Richman, In re, 563 F. 2d 1026 595,600 Risdon Locomotive Works v. Medart, 158 U. S. 68 592 Robbins v. Shelby County Taxing Dist., 120 U. S. 489 288 Robinson v. Georgia, 435 U. S. 991 910 Rocca v. Thompson, 223 U. S. 317 331 Roche v. Evaporated Milk Assn., 319 U. S. 21 661, 662,667,676 Roger J. Au & Son v. NLRB, 538 F. 2d 80 219,240,250 Roman v. Sincock, 377 U. S. 695 541 Root v. Gauper, 438 F. 2d 361 392 Rubber-Tip Pencil Co. v. Howard, 20 Wall. 498 598 Russell v. United States, 369 U. S. 749 66 Saalfrank v. O’Daniel, 533 F. 2d 325 367 Ste. Marie v. Eastern R. Assn., 72 F. R. D. 443 355 St. John’s Hospital v. NLRB, 557 F. 2d 1368 489,513 Sample v. Eyman, 469 F. 2d 819 389,403 Samuel v. University of Pittsburg, 506 F. 2d 355 475 Sanabria v. United States, 437 U. S. 54 32,91, 96, 104, 106, 108-110 Santa Fe Industries v. Green, 430 U. S. 462 330 Sapir v. United States, 348 U. S. 373 7-11,24 Schacht v. United States, 398 U. S. 58 62 Schmerber v. California, 384 U. S. 757 394 Scott v. Fancher, 369 F. 2d 842 375 Seagram & Sons, Inc. v. Hostetter, 384 U. S. 35 130,131 Sears, Roebuck & Co. v. Carpenters, 436 U. S. 180 582, 583 SEC v. Sloan, 436 U. S. 103 192, 706 Page Serfass v. United States, 420 U.S. 377 11,33,35-37,48-50,66, 77,86,95,109,112 Sewell v. Georgia, 435 U. S. 982 909,910 Seymour v. Superintendent, 368 U. S. 351 651 Shayne v. Madison Square Garden, 491 F. 2d 397 472 Shelly & Anderson Furniture Mfg. Co. v. NLRB, 497 F. 2d 1200 567 Short v. United States, 91 F. 2d 614 74 Sibbach v. Wilson & Co., 312 U. S. 1 370 Sierra Club v. Froehlke, 534 F. 2d 1289 208 Signal Delivery Service v. Highway Truck Drivers, 68 F. R. D. 318 690 Simmons v. United States, 142 U. S. 148 34,45 Sims v. Georgia, 385 U. S. 538 397 Skehan v. Board of Trustees, 538 F. 2d 53 696 Sligh v. Kirkwood, 237 U. S. 52 631 Smith v. Merchants & Farmers Bank, 574 F. 2d 982 479 Smith Petroleum Service v. Monsanto Chemical Co., 420 F. 2d 1103 375 Snyder v. Harris, 394 U. S. 332 370,374,377 Sorrells v. United States, 287 U. S. 435 204 South Carolina Hwy. Dept. v. Barnwell Bros., 303 U. S. 177 623,628 South Dakota v. Opperman, 428 U. S. 364 390 Southern Pacific v. Arizona ex rel. Sullivan, 325 U. S. 761 293, 628 Southern R. Co. v. Mayfield, 340 U. S. 1 524 Sprague v. Ticonic Nat. Bank, 307 U. S. 161 697 Springfield Institution for Savings v. Worcester Fed. S. & L. Assn., 329 Mass. 184 261 Standard Oil Co. v. FTC, 340 U. S. 231 132 TABLE OF CASES CITED XXV Page Standard Pressed Steel Co. v. Washington Revenue Dept., 419 U. S. 560 280 Standard Sanitary Mfg. Co. v. United States, 226 U. S. 20 307 Stanwood w Green, 22 F. Cas. 1077 311 State. See also name of State. State v. Cunningham, 166 Mont. 530 29 State v. Davis, 80 N. C. 384 46 State v. Duke, 110 Ariz. 320 389, 403 State v. Garrigues, 2 N. C. 188 45 State v. Gosser, 50 N. J. 438 392 State v. Hardin, 90 Nev. 10 392 State v. Minnesota Fed. Sav- ings, 218 Minn. 229 260 State v. M’Kee, 1 Bailey 651 45, 46 State v. Mojarro Padilla, 107 Ariz. 134 29 State v. Paterno, 43 La. Ann. 514 45,46 State v. Pires, 55 Wis. 2d 597 392 State v. Redman, 17 Iowa 329 46 State v. Roe, 12 Vt. 93 45 State v. Sample, 107 Ariz. 407 389, 391,403 State v. Sommers, 60 Minn. 90 45, 46 State v. Texaco, Inc., 14 Wis. 2d 625 130 State v. Van Ness, 82 N. J. L. 181 45 State v. Ward, 48 Ark. 36 45,46 State ex rel. Berger v. Superior Court, 110 Ariz. 281 389,403 State Farm Fire & Cas. Co. v. Tashire, 386 U. S. 523 373, 384,608 Stewart-Warner Corp. v. Westinghouse Electric Corp., 325 F 2d 822 481 Stone v. Powell, 428 U. S. 465 98, 402,404,405 Straub v. Vaisman & Co., 540 F. 2d 591 689 Strawbridge v. Curtiss, 3 Cranch 267 373,380 Stroble v. California, 343 U. S. 181 398,407 Page Sugarman v. Dougall, 413 U. S. 634 531 Sumpter v. DeGroote, 552 F. 2d 1206 11 Susman v. Lincoln American Corp., 561 F. 2d 86 475 Swann v. Charlotte-Mecklen- burg Bd. of Ed., 402 U. S. 1 687,713 Switzerland Cheese Assn. v. E. Home’s Market, 385 U. S. 23 481,482 Talley v. Stephens, 247 F. Supp. 683 681,682 Teal v. Georgia, 435 U. S. 989 910 Teamsters v. NLRB, 365 U. S. 667 431,433 Teat v. State, 53 Miss. 439 45 Telegraphers v. Railway Ex- press Agency, 321 U. S. 342 338 Temple-Eastex, Inc. v. NLRB, 410 F. Supp. 183 253 Terry v. Ohio, 392 U. S. 1 390, 393,406 Texas v. Florida, 306 U. S. 398 601,602,604- 607, 609, 610, 613, 615 Texas v. New Jersey, 379 U. S. 674 613 Texas v. New York, 300 U. S. 642 609 Textile Workers v. Darlington Co., 380 U. S. 263 251 Thermtron Products v. Her- mansdorfer, 423 U. S. 336 662, 670 Thomas v. Consumers Power Co., 58 Mich. App. 486 264 Thompson v. United States, 155U. S. 271 34,45 Thomson v. Gaskill, 315 U. S. 442 377 Tilghman v. Proctor, 102 U. S. 707 589-592,599,600 Title Guarantee Co. v. NLRB, 534 F. 2d 484 219, 220,233,241,253 Toledo Scale Co. v. Computing Scale Co., 261 U. S. 399 689 Toomer v. Witsell, 334 U. S. 385 524-527,624,627 Townsend v. Sain, 372 U. S. 293 398,401,407 XXVI TABLE OF CASES CITED Page Page Tradesmens Nat. Bank v. Oklahoma Tax Comm’n, 309 U. S. 560 264 Train v. Natural Resources Def. Council, 421 U. S. 60 450 Trans Alaska Pipeline Rate Cases, 436 U. S. 631 521 Travis v. Yale & Towne Mfg. Co., 252 U. S. 60 524 Treinies v. Sunshine Mining Co., 308 U. S. 66 614 Trono v. United States, 199 U. S. 521 12,15 Trupiano v. United States, 334 U. S. 699 390 Tubby v. State, 327 So. 2d 272 637 Turner’s Case, 89 Eng. Rep. 158 33 Udall v. Tailman, 380 U. S. 1 450, 458 Underwood Typewriter Co. v. Chamberlain, 254 U. S. 113 273, 274,286,287,291 Unemployment Compensation Comm’n v. Aragon, 329 U. S. 143 450 Union Refrigerator Transit Co. v. Kentucky, 199 U. S. 194 602 United Airlines v. McDonald, 432 U. S. 385 469,481 United States, Ex parte, 287 U. S. 241 667 United States, In re, 286 F. 2d 556 64 United States v. Alberti, 568 F. 2d 617 66,69 United States v. American Trucking Assns., 310 U. S. 534 204,207 United States v. Auto Workers, 352 U. S. 567 566 United States v. Ball, 163 U. S. 662 13,14,63, 64,69, 88-90, 96, 99, 104, 109 United States v. Barber, 219 U. S. 72 66 United States v. Barone, 330 F. 2d 543 392 United States v. Bass, 490 F. 2d 846 . 4 United States v. Becker, 461 F. 2d 230 71 United States v. Bigelow, 14 D. C. 393 44,45,47 United States v. Billingsley, 469 F. 2d 1208 305,306,311 United States v. Bobo, 477 F. 2d 974 74 United States v. Borden Co., 308 U. S. 188 190 United States v. Bowdach, 561 F. 2d 1160 901 United States v. Brick, 502 F. 2d 219 71 United States v. Brignoni- Ponce, 422 U. S. 873 406 United States v. Ceccolini, 435 U. S. 268 100 United States v. Celestine, 215 U. S. 278 649,654 United States v. Ceraso, 467 F. 2d 653 71 United States v. Chadwick, 433 U. S. 1 393 United States v. Choctaw Nation, 179 U. S. 494 638 United States v. Cohen, 197 F. 2d 26 74 United States v. CIO, 335 U. S. 106 566 United States v. Coolidge, 25 F. Cas. 622 44 United States v. DiMuro, 540 F. 2d 503 71,74 United States v. Dinitz, 424 U. S. 600 34.62,63,94,109 United States v. Duell, 172 U. S. 576 662 United States v. Edwards, 415 U. S. 800 391 United States v. 416.81 Acres of Land, 514 F. 2d 627 352 United States v. Frankfort Distilleries, 324 U. S. 293 136 United States v. Gibert, 25 F. Cas. 1287 44 United States v. Goldman, 277 U. S. 229 66 United States v. Graham, 534 F. 2d 1357 71 United States v. Haskell, 26 F. Cas. 207 41,44 United States v. Hawes, 529 F. 2d 472 70 TABLE OF CASES CITED XXVII Page United States v. Hodge & Zweig, 548 F. 2d 1347 305, 306,311,314 United States v. Howard, 360 F. 2d 373 353 United States v. Hunter, 478 F. 2d 1019 71 United States v. Jackson, 560 F. 2d 112 72 United States v. James, 408 F. Supp. 527 392 United States v. Jeffers, 342 U. S. 48 391 United States v. Jenkins, 420 U. S. 358 63,84, 86, 87, 90, 91, 94, 95, 97, 100-103, 107-110 United States v. Jones, 533 F. 2d 1387 70 United States v. Jom, 400 U. S. 470 11,36,66,93,96,99,105 United States v. Kagama, 118 U. S. 375 651 United States v. Keen, 26 F. Cas. 686 91,92 United States v. Kopp, 429 U. S. 121 100 United States v. Kordel, 397 U. S. 1 307 United States v. Lafko, 520 F. 2d 622 306,313,314 United States v. Langston, 118 U. S. 389 209 United States v. Leon, 534 F. 2d 667 71 United States v. Lovasco, 431 U. S. 783 111,115 United States v. MacDonald, 435 U. S. 850 111,468 United States v. Marion, 404 U. S. 307 111 United States v. Marrifield, 515 F. 2d 877 71 United States v. Martin Linen Supply Co., 430 U. S. 564 10, 35, 36, 63, 64, 69, 71, 79, 90, 91, 95-97, 102, 104, 106, 111, 112 United States v. Maurice, 26 F. Cas. 1211 , 703 United States v. Mayes, 512 F. 2d 637 74 Page United States v. McCarthy, 514 F. 2d 368 305,314 United States v. McGowan, 302 U. S. 535 648 United States v. Mine Workers, 330 U. S. 258 690,691 United States v. Morgan Guaranty Trust Co., 572 F. 2d 36 305,312,316,320,321 United States v. Morris, 26 F. Cas. 1323 45 United States v. Morrison, 429 U. S. 1 61,100 United States v. Morrison, 531 F. 2d 1089 62 United States v. Musquiz, 445 F. 2d 963 24 United States v. National State Bank, 454 F. 2d 1249 319 United States v. Nelson, 419 F. 2d 1237 16 United States v. New York Tel. Co., 434 U. S. 159 547 United States v. O’Connor, 118 F. Supp. 248 306,312 United States v. Orito, 413 U. S. 139 910 United States v. Pelican, 232 U. S. 442 649 United States v. Perez, 9 Wheat. 579 33,44,93 United States v. Powell, 379 U. S. 48 313,314,320,321 United States v. Robinson, 414 U. S. 218 391 United States v. Rose, 429 U. S. 5 100 United States v. Russell, 411 U. S. 423 97 United States v. Ryan, 284 U. S. 167 204 United States v. Sabella, 272 F. 2d 206 72 United States v. Sanges, 144 U. S. 310 85 United States v. Schaefer, 510 F. 2d 1307 71 United States v. Scott, 437 U. S. 82 32,51 United States v. Shoemaker, 27 F. Cas. 1067 45 United States v. Sisson, 399 U. S. 267 48,64,85,96,112 XXVIII TABLE OF CASES CITED Page United States v. Smaldone, 485 F. 2d 1333 71 United States v. State Tax Comm’n, 481 F. 2d 963 257,259 United States v. State Tax Comm’n, 505 F. 2d 633 637 United States v. Tanner, 471 F. 2d 128 74 United States v. Tateo, 377 U. S. 463 14,15,26,34 United States v. Theodore, 479 F. 2d 749 314 United States v. Troupe, 438 F. 2d 117 305 United States v. U. S. District Court, 407 U. S. 297 395 United States v. Universal C. I. T. Credit Corp., 344 U. S. 218 65,69 United States v. Wall Corp., 154 U. S. App. D. C. 309 305, 314 United States v. Watson, 28 F. Cas. 499 45 United States v. Weingarden, 473 F. 2d 454 305,312,314,319 United States v. Weiss, 293 F. 992 74 United States v. Weller, 401 U. S. 254 85 United States v. Wheeler, 435 U. S. 313 32,651 United States v. Wilson, 420 U. S. 332 14,15,34,41,63, 78, 85-87, 96, 104-106 United States v. Wolfenbarger, 426 F. 2d 992 16 United States v. Wright, 53 F. 2d 300 653 United States v. Young, 503 F. 2d 1072 74 United States v. Zack, 521 F. 2d 1366 305,314,319 United States ex rei. Greathouse v. Dern, 289 U. S. 352 213 United States ex rei. Griffin v. McMann, 310 F. Supp. 72 696 United States ex rei. Parson v. Anderson, 354 F. Supp. 1060 392 United States ex rei. Petillo v. New Jersey, 562 F. 2d 903 405 Page United States Fidelity & Guaranty Co. v. United States ex rel. Struthers Wells Co., 209 U. S. 306 206 United States Glue Co. v. Oak Creek, 247 U. S. 321 291 United States Steel Corp. v. Multistate Tax Comm’n, 434 U. S. 452 282 Universal Camera Corp. v. NLRB, 340 U. S. 474 501,507 Utah v. United States, 304 F. 2d 23 696 Vale v. Louisiana, 399 U. S. 30 390,391 Vaux’s Case, 4 Co. Rep. 44a 88 Virginia, Ex parte, 100 U. S. 339 702 Wabash, St. L. & P. R. Co. v. Illinois, 118 U. S. 557 128 Wade v. Hunter, 336 U. S. 684 15,34,36,38,47,51,63,101 Ward v. Maryland, 12 Wall. 418 524-526 Warden v. Hayden, 387 U. S. 294 392,394 Wayne v. United States, 115 U. S. App. D. C. 234 392 Weems v. United States, 217 U. S. 349 685 Weiner v. Shearson, Hammill & Co., 521 F. 2d 817 659 Weisberg v. Department of Justice, 106 U. S. App. D. C. 71 228,246 Wellman Industries v. NLRB, 490 F. 2d 427 226, 233 Welton v. Missouri, 91 U. S. 275 288,624 West v. Kansas Nat. Gas Co., 221 U. S. 229 532,533,627 Western Live Stock v. Bureau of Revenue, 303 U. S. 250 281, 289 Western Union Telegraph Co. v. Pennsylvania, 368 U. S. 71 612 West Point Wholesale Grocery Co. v. Opelika, 354 U. S. 390 288 Whitcomb v. Chavis, 403 U. S. 124 541,554 White v. Dunbar, 119 U. S. 47 590 TABLE OF CASES CITED XXIX Page White v. Regester, 412 U. S. 755 541,550,555 Whitebread’s Case, 7 How. St. Tr. 311 43 White Motor Co. v. United States, 372 U. S. 253 139 Will v. United States, 389 U. S. 90 79,661,666,676 Williams v. Florida, 399 U. S. 78 37 Williams v. Lee, 358 U. S. 217 651 Williams v. Mumford, 167 U. S. App. D. C. 125 469,479 Williams v. Wallace Silver- smiths, Inc., 566 F. 2d 364 479 Winsor v. Queen, 1 Q. B. 390 36 Winton v. Amos, 255 U. S. 373 638 Winton’s Estate v. Amos, 51 Ct. Cl. 284 644 Wisconsin River Valley Dist. Council v. NLRB, 532 F. 2d 47 422,426,435 Page Wong Sun v. United States, 371 U. S. 471 395 Wood v. Strickland, 420 U. S. 308 699,710 Wood v. United States, 16 Pet. 342 190 Worcester v. Georgia, 6 Pet. 515 640 Worcester County Trust Co. v. Riley, 302 U. S. 292 607, 608,612,615,616 Yaffe v. Powers, 454 F. 2d 1362 479 Yakus v. United States, 321 U. S. 414 213 Yates v. United States, 354 U. S. 298 8-10 Young, Ex parte, 209 U. S. 123 608,690,717 Zahn v. International Paper Co., 414 U. S. 291 372,373 Zuber v. Allen, 396 U. S. 168 456 CASES ADJUDGED IN THE SUPBEME COUBT OF THE UNITED STATES AT OCTOBER TERM, 1977 BURKS v. UNITED STATES CERTIORARI TO THE UNITED STATES COURT OF APPEALS FOR THE SIXTH CIRCUIT No. 76-6528. Argued November 28, 1977—Decided June 14, 1978 Petitioner, in support of his insanity defense to a bank robbery charge, offered expert testimony, and the Government offered expert and lay testimony in rebuttal. Before the case was submitted to the jury, the District Court denied a motion for acquittal. The jury found petitioner guilty as charged, and thereafter his motion for a new trial on the ground that the evidence was insufficient to support the verdict was denied. The Court of Appeals, holding that the Government had failed to rebut petitioner’s proof as to insanity, reversed and remanded to the District Court to determine whether a directed verdict of acquittal should be entered or a new trial ordered, citing, inter alia, as authority for such a remand 28 U. S. C. § 2106, which authorizes federal appellate courts to remand a cause and “direct the entry of such appropriate judgment, decree, or order, or require such further proceedings to be had as may be just under the circumstances.” Held: The Double Jeopardy Clause of the Fifth Amendment precludes a second trial once the reviewing court has found the evidence insufficient to sustain the jury’s verdict of guilty, and the only “just” remedy available for that court under 28 U. 8. C. § 2106 is the entry of a judgment of acquittal. Pp. 5-18. (a) For the purposes of determining whether the Double Jeopardy Clause precludes a second trial after the reversal of a conviction, a reversal based on insufficiency of evidence is to be distinguished from a reversal for trial error. In holding the evidence insufficient to sustain 1 2 OCTOBER TERM, 1977 Opinion of the Court 437 U. S. guilt, an appellate court determines that the prosecution has failed to prove guilt beyond a reasonable doubt. Given the requirements for entry of a judgment of acquittal, to permit a second trial would negate the purpose of the Double Jeopardy Clause to forbid a second trial in which the prosecution would be afforded another opportunity to supply evidence that it failed to muster in the first trial. Pp. 15-17. (b) It makes no difference that a defendant has sought a new trial as one of his remedies, or even as the sole remedy, and he does not waive his right to a judgment of acquittal by moving for a new trial. Bryan v. United States, 338 U. S. 552; Sapir n. United States, 348 U. S. 373; Yates v. United States, 354 U. S. 298; and Forman v. United States, 361 U. S. 416, are overruled to the extent that they suggest such a waiver. Pp. 17-18. 547 F. 2d 968, reversed and remanded. Burger, C. J., delivered the opinion of the Court, in which all other Members joined except Blackmun, J., who took no part in the consideration or decision of the case. Bart C. Durham III argued the cause and filed briefs for petitioner. Frank H. Easterbrook argued the cause for the United States pro hac vice. With him on the brief were Acting Solicitor General Friedman, Assistant Attorney General Civiletti, and Michael W. Farrell. Mr. Chief Justice Burger delivered the opinion of the Court. We granted certiorari to resolve the question of whether an accused may be subjected to a second trial when conviction in a prior trial was reversed by an appellate court solely for lack of sufficient evidence to sustain the jury’s verdict. I Petitioner Burks was tried in the United States District Court for the crime of robbing a federally insured bank by use of a dangerous weapon, a violation of 18 U. S. C. § 2113 (d) (1976 ed.). Burks’ principal defense was insanity. To prove this BURKS v. UNITED STATES 3 1 Opinion of the Court claim petitioner produced three expert witnesses who testified, albeit with differing diagnoses of his mental condition, that he suffered from a mental illness at the time of the robbery, which rendered him substantially incapable of conforming his conduct to the requirements of the law. In rebuttal the Government offered the testimony of two experts, one of whom testified that although petitioner possessed a character disorder, he was not mentally ill. The other prosecution witness acknowledged a character disorder in petitioner, but gave a rather ambiguous answer to the question of whether Burks had been capable of conforming his conduct to the law. Lay witnesses also testified for the Government, expressing their opinion that petitioner appeared to be capable of normal functioning and was sane at the time of the alleged offense. Before the case was submitted to the jury, the court denied a motion for a judgment of acquittal. The jury found Burks guilty as charged. Thereafter, he filed a timely motion for a new trial, maintaining, among other things, that “[t]he evidence was insufficient to support the verdict.” The motion was denied by the District Court, which concluded that petitioner’s challenge to the sufficiency of the evidence was “utterly without merit.” 1 On appeal petitioner narrowed the issues by admitting the affirmative factual elements of the charge against him, leaving only his claim concerning criminal responsibility to be resolved. With respect to this point, the Court of Appeals agreed with petitioner’s claim that the evidence was insufficient to support the verdict and reversed his conviction. 547 F. 2d 968 (CA6 1976). The court began by noting that “the government has the burden of proving sanity [beyond a reasonable doubt] once a prima facie defense of insanity has been raised.” 2 Id., 1 Petitioner did not file a post-trial motion for judgment of acquittal, which he was entitled to do under Fed. Rule Crim. Proc. 29 (c). 2 Although the Court of Appeals did not cite Davis v. United States, 160 U. S. 469 (1895), that decision would require this allocation of burdens. 4 OCTOBER TERM, 1977 Opinion of the Court 437 U. S. at 969. Petitioner had met his obligation, the court indicated, by presenting “the specific testimony of three experts with unchallenged credentials.” Id., at 970. But the reviewing court went on to hold that the United States had not fulfilled its burden since the prosecution’s evidence with respect to Burks’ mental condition, even when viewed in the light most favorable to the Government, did not “effectively rebu[t]” petitioner’s proof with respect to insanity and criminal responsibility. Ibid. In particular, the witnesses presented by the prosecution failed to “express definite opinions on the precise questions which this Court has identified as critical in cases involving the issue of sanity.” Ibid. At this point, the Court of Appeals, rather than terminating the case against petitioner, remanded to the District Court “for a determination of whether a directed verdict of acquittal should be entered or a new trial ordered.” Ibid. Indicating that the District Court should choose the appropriate course “from a balancing of the equities,” ibid., the court explicitly adopted the procedures utilized by the Fifth Circuit in United States v. Bass, 490 F. 2d 846, 852-853 (1974), “as a guide” to be used on remand: “[W]e reverse and remand the case to the district court where the defendant will be entitled to a directed verdict of acquittal unless the government presents sufficient additional evidence to carry its burden on the issue of defendant’s sanity. As we noted earlier, the question of sufficiency of the evidence to make an issue for the jury on the defense of insanity is a question of law to be decided by the trial judge. ... If the district court, sitting without the presence of the jury, is satisfied by the government’s presentation, it may order a new trial. . . . Even if the government presents additional evidence, the district judge may refuse to order a new trial if he finds from the record that the prosecution had the opportunity fully to develop its case or in fact did so at the first trial.” BURKS v. UNITED STATES 5 1 Opinion of the Court The Court of Appeals assumed it had the power to order this “balancing” remedy by virtue of the fact that Burks had explicitly requested a new trial. As authority for this holding the court cited, inter alia, 28 U. S. C. § 2106,3 and Bryan v. United States, 338 U. S. 552 (1950). 547 F. 2d, at 970. II The United States has not cross-petitioned for certiorari on the question of whether the Court of Appeals was correct in holding that the Government had failed to meet its burden of proof with respect to the claim of insanity. Accordingly, that issue is not open for review here. Given this posture, we are squarely presented with the question of whether a defendant may be tried a second time when a reviewing court has determined that in a prior trial the evidence was insufficient to sustain the verdict of the jury.4 Petitioner’s argument is straightforward. He contends that the Court of Appeals’ holding was nothing more or less than a decision that the District Court had erred by not granting his motion for a judgment of acquittal. By implication, he argues, the appellate reversal was the operative equivalent of a district court’s judgment of acquittal, entered either before or after verdict. Petitioner points out, however, that had the District Court found the evidence at the first trial inadequate, as the Court of Appeals said it should have done, a second trial would violate the Double Jeopardy Clause of the 3 Title 28 U. S. C. § 2106 provides: “The Supreme Court or any other court of appellate jurisdiction may affirm, modify, vacate, set aside or reverse any judgment, decree, or order of a court lawfully brought before it for review, and may remand the cause and direct the entry of such appropriate judgment, decree, or order, or require such further proceedings to be had as may be just under the circumstances.” 4 There is no claim in this case that the trial court committed error by excluding prosecution evidence which, if received, would have rebutted any claim of evidentiary insufficiency. 6 OCTOBER TERM, 1977 Opinion of the Court 437 U.S. Fifth Amendment. Therefore, he maintains, it makes no difference that the determination of evidentiary insufficiency was made by a reviewing court since the double jeopardy considerations are the same, regardless of which court decides that a judgment of acquittal is in order. The position advanced by petitioner has not been embraced by our prior holdings. Indeed, as the Court of Appeals here recognized, Bryan v. United States, supra, would appear to be contrary. In Bryan the defendant was convicted in the District Court for evasion of federal income tax laws. Bryan had moved for a judgment of acquittal both at the close of the Government’s case and when all of the evidence had been presented. After the verdict was returned he renewed these motions, but asked—in the alternative—for a new trial. These motions were all denied. The Court of Appeals reversed the conviction on the specific ground that the evidence was insufficient to sustain the verdict and remanded the case for a new trial. Certiorari was then granted to determine whether the Court of Appeals had properly ordered a new trial, or whether it should have entered a judgment of acquittal. In affirming the Court of Appeals, this Court decided, first, that the Court of Appeals had statutory authority, under 28 U. S. C. § 2106, to direct a new trial. But Bryan had also maintained that notwithstanding § 2106 a retrial was prohibited by the Double Jeopardy Clause, a contention which was dismissed in one paragraph: “Petitioner’s contention that to require him to stand trial again would be to place him twice in jeopardy is not persuasive. He sought and obtained the reversal of his conviction, assigning a number of alleged errors on appeal, including denial of his motion for judgment of acquittal, i . . [W]here the accused successfully seeks review of a conviction, there is no double jeopardy upon a new trial.’ Francis v. Res weber, 329 U. S. 459, 462. See Trono v. United States, 199 U. S. 521, 533^534.” 338 U. S., at 560. BURKS v. UNITED STATES 7 1 Opinion of the Court Five years after Bryan was decided, a similar claim of double jeopardy was presented to the Court in Sapir v. United States, 348 U. S. 373 (1955). Sapir had been convicted of conspiracy by a jury in the District Court. After the trial court denied a motion for acquittal, he obtained a reversal in the Court of Appeals, which held that the motion should have been granted since the evidence was insufficient to sustain a conviction. In a brief per curiam opinion, this Court, without explanation, reversed the Court of Appeals’ decision to remand the petitioner’s case for a new trial. Concurring in the Sapir judgment, which directed the dismissal of the indictment, Mr. Justice Douglas indicated his basis for reversal: “The correct rule was stated in Kepner v. United States, 195 U. S. 100, at 130, Tt is, then, the settled law of this court that former jeopardy includes one who has been acquitted by a verdict duly rendered . . . .’ If the jury had acquitted, there plainly would be double jeopardy to give the Government another go at this citizen. If, as in the Kepner case, the trial judge had rendered a verdict of acquittal, the guarantee against double jeopardy would prevent a new trial of the old offense. I see no difference when the appellate court orders a judgment of acquittal for lack of evidence.” Id., at 374. Up to this point, Mr. Justice Douglas’ explication is, of course, precisely that urged on us by petitioner, and presumably would have been applicable to Bryan as well. But the concurrence in Sapir then undertook to distinguish Bryan: “If petitioner [Sapir] had asked for a new trial, different considerations would come into play, for then the defendant opens the whole record for such disposition as might be just. See Bryan v. United States, 338 U. S. 552.” 348 U. S., at 374. (Emphasis added.) 8 OCTOBER TERM, 1977 Opinion of the Court 437 U. S. Shortly after Sapir, in Yates v. United States, 354 U. S. 298 (1957), the Court adopted much the same reasoning as that employed by the Sapir concurrence. In Yates, this Court— without citing Sapir—ordered acquittals for some defendants in the case, but new trials for others, when one of the main contentions of the petitioners concerned the insufficiency of the evidence. As an explanation for the differing remedies, the Court stated: “We think we may do this by drawing on our power under 28 U. S. C. § 2106, because under that statute we would no doubt be justified in refusing to order acquittal even where the evidence might be deemed palpably insufficient, particularly since petitioners have asked in the alternative for a new trial as well as for acquittal. See Bryan v. United States, 338 U. S. 552.” 354 U. S., at 328. The Yates decision thus paralleled Sapir’s concurrence in the sense that both would allow a new trial to correct evidentiary insufficiency if the defendant had requested such relief—even as an alternative to a motion for acquittal. But the language in Yates was also susceptible of a broader reading, namely, that appellate courts have full authority to order a new trial as a remedy for evidentiary insufficiency, even when the defendant has moved only for a judgment of acquittal. Three years later in Forman v. United States, 361 U. S. 416 (1960), the Court again treated these questions. There a conviction was reversed by the Court of Appeals due to an improper instruction to the jury, i. e., trial error, as opposed to evidentiary insufficiency. Although the petitioner in Forman had moved both for a new trial and judgment of acquittal, he argued that a new trial would not be appropriate relief since he had requested a judgment of acquittal with respect to the specific trial error on which this Court agreed with the Court of Appeals. Without distinguishing between a reversal due to trial error and reversal resulting solely from evidentiary BURKS v. UNITED STATES 9 1 Opinion of the Court insufficiency, this Court held that a new trial did not involve double jeopardy: “It is elementary in our law that a person can be tried a second time for an offense when his prior conviction for that same offense has been set aside by his appeal. United States v. Ball, 163 U. S. 662, 672 (1896). . . . Even though petitioner be right in his claim that he did not request a new trial with respect to the portion of the charge dealing with the statute of limitations, still his plea of double jeopardy must fail. Under 28 U. S. C. § 2106, the Court of Appeals has full power to go beyond the particular relief sought. See Ball, and other cases, supra.” Id., at 425. Until this stage in the Forman opinion the Court seemed to adopt the more expansive implication of Yates, i. e., that an appellate court’s choice of remedies for an unfair conviction— whether reversal be compelled by failure of proof or trial error—would not turn on the relief requested by the defendant. The Forman decision, however, was not entirely free from ambiguity. In the course of meeting the petitioner’s argument that Sapir demanded a judgment of acquittal, the Court noted two differences between those cases. In the first place, “the order to dismiss in Sapir was based on the insufficiency of the evidence, which could be cured only by the introduction of new evidence”; in Forman, however, “ ‘[t]he jury was simply not properly instructed.’ ” 361 U. S., at 426: In addition, “Sapir made no motion for a new trial in the District Court, while here petitioner [Forman] filed such a motion. That was a decisive factor in Sapir’s case.” Ibid. (Emphasis added.) The Court’s holdings in this area, beginning with Bryan, can hardly be characterized as models of consistency and clarity. Bryan seemingly stood for the proposition that an appellate court could order whatever relief was “appropriate” 10 OCTOBER TERM, 1977 Opinion of the Court 437 U. S. or “equitable,” regardless of what considerations prompted reversal. A somewhat different course was taken by the concurrence in Sapir, where it was suggested that a reversal for evidentiary insufficiency would require a judgment of acquittal unless the defendant had requested a new trial. Yates, on the contrary, implied that new trials could be ordered to cure prior inadequacies of proof even when the defendant had not so moved. While not completely resolving these ambiguities, Forman suggested that a reviewing court could go beyond the relief requested by a defendant and order a new trial under some circumstances. In discussing Sapir, however, the Forman Court intimated that a different result might follow if the conviction was reversed for evidentiary insufficiency and the defendant had not requested a new trial. After the Bryan-Forman line of decisions at least one proposition emerged: A defendant who requests a new trial as one avenue of relief may be required to stand trial again, even when his conviction was reversed due to failure of proof at the first trial. Given that petitioner here appealed from a denial of a motion for a new trial—although he had moved for acquittal during trial—our prior cases would seem to indicate that the Court of Appeals had power to remand on the terms it ordered. To reach a different result will require a departure from those holdings. Ill It is unquestionably true that the Court of Appeals’ decision “represente [d] a resolution, correct or not, of some or all of the factual elements of the offense charged.” United States v. Martin Linen Supply Co., 430 U. S. 564, 571 (1977). By deciding that the Government had failed to come forward with sufficient proof of petitioner’s capacity to be responsible for criminal acts, that court was clearly saying that Burks’ criminal culpability had not been established. If the District Court had so held in the first instance, as the reviewing court said it should have done, a judgment of acquittal would have BURKS v. UNITED STATES 11 1 Opinion of the Court been entered5 and, of course, petitioner could not be retried for the same offense. See Fong Foo v. United States, 369 U. S. 141 (1962); Kepner v. United States, 195 U. S. 100 (1904). Consequently, as Mr. Justice Douglas correctly perceived in Sapir, it should make no difference that the reviewing court, rather than the trial court, determined the evidence to be insufficient, see 348 U. S., at 374. The appellate decision unmistakably meant that the District Court had erred in failing to grant a judgment of acquittal. To hold otherwise would create a purely arbitrary distinction between those in petitioner’s position and others who would enjoy the benefit of a correct decision by the District Court. See Sumpter v. DeGroote, 552 F. 2d 1206, 1211-1212 (CA7 1977). The Double Jeopardy Clause forbids a second trial for the purpose of affording the prosecution another opportunity to supply evidence which it failed to muster in the first proceeding.6 This is central to the objective of the prohibition against successive trials. The Clause does not allow “the State . . . to make repeated attempts to convict an individual for an alleged offense,” since “[t]he constitutional prohibition against ‘double jeopardy’ was designed to protect an individual from being subjected to the hazards of trial and possible conviction more than once for an alleged * offense.” Green v. United States, 355 U. S. 184, 187 (1957); see Serfass v. United States, 420 U. S. 377, 387-388 (1975); United States v. Jorn, 400 U. S. 470, 479 (1971). 5 When a district court determines, at the close of either side’s case, that the evidence is insufficient, it “shall order the entry of [a] judgment of acquittal . . . .” Fed. Rule Crim. Proc. 29; see C. Wright, Federal Practice and Procedure § 462, p. 245 (1969). 6 We recognize that under the terms of the remand in this case the District Court might very well conclude, after “a balancing of the equities,” that a second trial should not be held. Nonetheless, where the Double Jeopardy Clause is applicable, its sweep is absolute. There are no “equities” to be balanced, for the Clause has declared a constitutional policy, based on grounds which are not open to judicial examination. 12 OCTOBER TERM, 1977 Opinion of the Court 437 U. S. Nonetheless, as the discussion in Part II, supra, indicates, our past holdings do not appear consistent with what we believe the Double Jeopardy Clause commands. A close reexamination of those precedents, however, persuades us that they have not properly construed the Clause, and accordingly should no longer be followed. Reconsideration must begin with Bryan v. United States. The brief and somewhat cursory examination of the double jeopardy issue there was limited to stating that “ ‘where the accused successfully seeks review of a conviction, there is no double jeopardy upon a new trial,’ ” 338 U. S., at 560, citing Louisiana ex rel. Francis v. Resweber, 329 U. S. 459,462 (1947), and Trono v. United States, 199 U. S. 521, 533-534 (1905). These two cited authorities, which represent the totality of the Court’s analysis, add little, if anything, toward resolving the double jeopardy problem presented by Bryan. Resweber involved facts completely unrelated to evidentiary insufficiency. There, in what were admittedly “unusual circumstances,” 329 U. S., at 461, the Court decided that a State would be allowed another chance to carry out the execution of one properly convicted and under sentence of death after an initial attempted electrocution failed due to some mechanical difficulty. In passing, the opinion stated: “But where the accused successfully seeks review of a conviction, there is no double jeopardy upon a new trial. United States v. Ball, 163 U. S. 662, 672.” Id., at 462. Trono made a similar comment, citing Ball for the proposition that “if the judgment of conviction be reversed on [the defendant’s] own appeal, he cannot avail himself of the once-in-jeopardy provision as a bar to a new trial of the offense for which he was convicted.” 199 U. S., at 533-5345 7 Trono arose from a murder prosecution in the Philippines. After a nonjury trial the defendants were acquitted of the crime of murder, but were convicted of the lesser included offense of assault. They appealed to the Supreme Court of the Philippine Islands, which reversed the judgment BURKS v. UNITED STATES 13 1 Opinion of the Court The common ancestor of these statements in Resweber and Trono, then, is United States v. Ball, which provides a logical starting point for unraveling the conceptual confusion arising from Bryan and the cases which have followed in its wake. This is especially true since Ball appears to represent the first instance in which this Court considered in any detail the double jeopardy implications of an appellate reversal. North Carolina v. Pearce, 395 U. S. 711, 719-720 (1969). Ball came before the Court twice, the first occasion being on writ of error from federal convictions for murder. On this initial review, those defendants who had been found guilty obtained a reversal of their convictions due to a fatally defective indictment. On remand after appeal, the trial court dismissed the flawed indictment and proceeded to retry the defendants on a new indictment. They were again convicted and the defendants came once more to this Court, arguing that their second trial was barred because of former jeopardy. The Court rejected this plea in a brief statement: “[A] defendant, who procures a judgment against him upon an indictment to be set aside, may be tried anew upon the same indictment, or upon another indictment, for the same offence of which he had been convicted. Hopt v. Utah, 104 U. S. 631; 110 U. S. 574; 114 U. S. 488; 120 U. S. 430; Regina v. Drury, 3 Cox Crim. Cas. 544; S. C. 3 Car. & Kirw. 193; Common wealth v. Gould, 12 Gray, 171.” 163 U. 8., at 672. and entered convictions for murder, increasing their sentences as well. This Court affirmed, although “it seems apparent that a majority of the Court was unable to agree on any common ground for the conclusion that an appeal of a lesser offense destroyed a defense of a former jeopardy on a greater offense for which the defendant had already been acquitted.” Green v. United States, 355 U. S. 184, 187 (1957). Green expressly confined the Trono decision to “its peculiar factual setting,” namely, an interpretation of a “statutory provision against double jeopardy pertaining to the Philippine Islands.” 355 U. S., at 187; see Price v. Georgia, 398 U. 8. 323, 327-328, n. 3 (1970). 14 OCTOBER TERM, 1977 Opinion of the Court 437 U. S. The reversal in Ball was therefore based not on insufficiency of evidence but rather on trial error, i. e., failure to dismiss a faulty indictment. Moreover, the cases cited as authority by Ball were ones involving trial errors.8 We have no doubt that Ball was correct in allowing a new trial to rectify trial error: “The principle that [the Double Jeopardy Clause] does not preclude the Government’s retrying a defendant whose conviction is set aside because of an error in the proceedings leading to conviction is a well-established part of our constitutional jurisprudence.” United States v. Tateo, 377 U. S. 463, 465 (1964) (emphasis supplied). See United States v. Wilson, 420 U. S. 332, 341 n. 9 (1975); Forman, 361 U. S., at 425. As we have seen in Part II, supra, the cases which have arisen since Ball generally do not distin 8 Hopt v. Utah, 120 U. S. 430 (1887), was the last of four appeals by a defendant from a murder conviction in the Territory of Utah. On the first three appeals the convictions were reversed and new trials ordered because of trial errors, e. g., improper instruction, 104 U. S. 631 (1882); absence of the accused during a portion of the trial, improper hearsay testimony received, and prejudicial instruction, 110 U. S. 574 (1884); and inadequate record due to failure to record jury instructions, 114 U. S. 488 (1885). No claim of evidentiary insufficiency was sustained by the Court, and indeed no discussion of double jeopardy appears. Commonwealth v. Gould, 78 Mass. 171 (1858), was a state case in which a defendant was ordered tried on a superseding indictment, after the original indictment had been challenged. Finally, in the English case, Queen n. Drury, 3 Cox Crim. Cas. 544, 175 Eng. Rep. 516 (Q. B. 1849), the defendants had been given an improper sentence after being found guilty at a trial to which no other error was assigned. The court allowed a retrial, saying: “A man who has been tried, convicted and attainted on an insufficient indictment, or on a record erroneous in any other part, is in so much jeopardy literally that punishment may be lawfully inflicted on him, unless the attainder be reversed in a Court of Error; and yet when that is done, he may certainly be indicted again for the same offense, and the rule would be held to apply, that he had never been in jeopardy under the former indictment.” Id., at 546, 175 Eng. Rep., at 520. BURKS v. UNITED STATES 15 1 Opinion of the Court guish between reversals due to trial error and those resulting from evidentiary insufficiency. We believe, however, that the failure to make this distinction has contributed substantially to the present state of conceptual confusion existing in this area of the law. eConsequently, it is important to consider carefully the respective roles of these two types of reversals in double jeopardy analysis. Various rationales have been advanced to support the policy of allowing retrial to correct trial error,9 but in our view the most reasonable justification is that advanced by Tateo, supra, at 466: “It would be a high price indeed for society to pay were every accused granted immunity from punishment because of any defect sufficient to constitute reversible error in the proceedings leading to conviction.” See Wilson, supra, at 343-344, n. 11; Wade v. Hunter, 336 U. S. 684, 688-689 (1949). In short, reversal for trial error, as distinguished from evidentiary insufficiency, does not constitute a decision to the effect that the government has failed to prove its case. As such, it implies nothing with respect to the guilt or innocence of the defendant. Rather, it is a determination that a defendant has been convicted through a judicial process which is defective in some fundamental respect, e. g., incorrect receipt or rejection of evidence, incorrect instructions, or prosecutorial misconduct. When this occurs, the accused has a strong interest in obtaining a fair readjudication of his guilt free from error, just as society maintains a valid concern for insuring that the guilty are punished. See Note, Double Jeopardy: A New Trial After 9 It has been suggested, for example, that an appeal from a conviction amounts to a “waiver” of double jeopardy protections, see Trono v. United States, 199 U. S. 521, 533 (1905); but see Green, supra, at 191-198; or that the appeal somehow continues the jeopardy which attached at the first trial, see Price v. Georgia, supra, at 326; but see Breed n. Jones, 421 U. S. 519, 534 (1975). 16 OCTOBER TERM, 1977 Opinion of the Court 437 U. S. Appellate Reversal for Insufficient Evidence, 31 U. Chi. L. Rev. 365, 370 (1964). The same cannot be said when a defendant’s conviction has been overturned due to a failure of proof at trial, in which case the prosecution cannot complain of prejudice, for it has been given one fair opportunity to offer whatever proof it could assemble.10 Moreover, such an appellate reversal means that the government’s case was so lacking that it should not have even been submitted to the jury. Since we necessarily afford absolute finality to a jury’s verdict of acquittal—no matter how erroneous its decision—it is difficult to conceive how society has any greater interest in retrying a defendant when, on review, it is decided as a matter of law that the jury could not properly have returned a verdict of guilty. The importance of a reversal on grounds of evidentiary insufficiency for purposes of inquiry under the Double Jeopardy Clause is underscored by the fact that a federal court’s role in deciding whether a case should be considered by the jury is quite limited. Even the trial court, which has heard the testimony of witnesses firsthand, is not to weigh the evidence or assess the credibility of witnesses when it judges the merits of a motion for acquittal. See United States v. Wolfenbarffer, 426 F. 2d 992, 994 (CA6 1970); United States v. Nelson, 419 F. 2d 1237, 1241 (CA9 1969); McClard v. United States, 386 F. 2d 495, 497 (CA8 1968); Curley v. United States, 81 U. S. App. D. C. 389, 392, 160 F. 2d 229, 232-233, cert, denied, 331 U. S. 837 (1947). The prevailing rule has long been that a district judge is to submit a case to the jury if the evidence and inferences therefrom most favorable to the prosecution would warrant the jury’s finding the defendant guilty beyond a reasonable doubt. See C. Wright, Federal Practice and 10 In holding the evidence insufficient to sustain guilt, an appellate court determines that the prosecution has failed to prove guilt beyond a reasonable doubt. See American Tobacco Co. v. United States, 328 U. S. 781, 787 n. 4 (1946). BURKS v. UNITED STATES 17 1 Opinion of the Court Procedure §467, pp. 259-260 (1969); e. g., Powell v. United States, 135 U. S. App. D. C. 254, 257, 418 F. 2d 470, 473 (1969); Crawjord v. United States, 126 U. S. App. D. C. 156, 158, 375 F. 2d 332, 334 (1967). Obviously a federal appellate court applies no higher a standard; rather, it must sustain the verdict if there is substantial evidence, viewed in the light most favorable to the Government, to uphold the jury’s decision. See Glasser v. United States, 315 U. S. 60, 80 (1942). While this is not the appropriate occasion to re-examine in detail the standards for appellate reversal on grounds of insufficient evidence, it is apparent that such a decision will be confined to cases where the prosecution’s failure is clear.11 Given the requirements for entry of a judgment of acquittal, the purposes of the Clause would be negated were we to afford the government an opportunity for the proverbial “second bite at the apple.” In our view it makes no difference that a defendant has sought a new trial as one of his remedies, or even as the sole remedy. It cannot be meaningfully said that a person “waives” his right to a judgment of acquittal by moving for a new trial. See Green v. United States, 355 U. S., at 191-198. Moreover, as Forman, 361 U. S., at 425, has indicated, an appellate court is authorized by § 2106 to “go beyond the particular relief sought” in order to provide that relief which 11 When the basic issue before the appellate court concerns the sufficiency of the Government’s proof of a defendant’s sanity (as it did here), a reviewing court should be most wary of disturbing the jury verdict: “There may be cases where the facts adduced as to the existence and impact of an accused’s mental condition may be so overwhelming as to require a judge to conclude that no reasonable juror could entertain a reasonable doubt. But in view of the complicated nature of the decision to be made—intertwining moral, legal, and medical judgments—it will require an unusually strong showing to induce us to reverse a conviction because the judge left the critical issue of criminal responsibility with the jury.” King n. United States, 125 U. S. App. D. C. 318, 324, 372 F. 2d 383,389 (1967) (footnote omitted). 18 OCTOBER TERM, 1977 Opinion of the Court 437 U. S. would be “just under the circumstances.” Since we hold today that the Double Jeopardy Clause precludes a second trial once the reviewing court has found the evidence legally insufficient, the only “just” remedy available for that court is the direction of a judgment of acquittal. To the extent that our prior decisions suggest that by moving for a new trial, a defendant waives his right to a judgment of acquittal on the basis of evidentiary insufficiency, those cases are overruled. Accordingly, the judgment of the Court of Appeals is reversed, and the case is remanded for proceedings consistent with this opinion. Reversed and remanded. Mr. Justice Blackmun took no part in the consideration or decision of this case. GREENE v. MASSEY 19 Syllabus GREENE v. MASSEY, CORRECTIONAL SUPERINTENDENT CERTIORARI TO THE UNITED STATES COURT OF APPEALS FOR THE FIFTH CIRCUIT No. 76-6617. Argued November 28, 1977—Decided June 14, 1978 On appeal of the first-degree murder convictions of petitioner and another, the Florida Supreme Court reversed by a per curiam opinion and ordered a new trial. That opinion, which a majority of four justices joined, stated that “the evidence was definitely lacking in establishing beyond a reasonable doubt that the defendants committed murder in the first degree,” and that the “interests of justice require a new trial.” Three justices dissented without opinion. Three of the justices who had joined the per curiam also filed a “special concurrence,” which, though concerned only with trial error, concluded that “[f]or the reasons stated the judgments should be reversed and remanded for a new trial so we have agreed to the Per Curiam order doing so.” Before the second trial defendants unsuccessfully contended in the state courts that the per curiam opinion was tantamount to a finding that the trial court should have directed a verdict of not guilty and that a second trial for first-degree murder would constitute double jeopardy; and the defendants were retried and convicted of first-degree murder. Petitioner and his codefendant, by appeal in the state courts and petitioner by application for habeas corpus in the District Court and Court of Appeals, unavail-ingly pressed their double jeopardy claims. Held: Burks v. United States, ante, p. 1, precludes a second trial once a reviewing court has determined that the evidence introduced at trial is insufficient to sustain the verdict. Standing by itself, the per curiam would therefore clearly compel the conclusion that petitioner’s second trial violated the Double Jeopardy Clause. But the special concurrence leaves open the possibility that three of the justices who joined the per curiam were concerned simply with trial error and joined in the remand solely to give the defendants an error-free trial—even though they were satisfied that the evidence was sufficient to support the verdict. So that the ambiguity can be resolved, the case is remanded to the Court of Appeals for reconsideration in light of the Court’s opinion and Burks, supra. Pp. 24-27. 546 F. 2d 51, reversed and remanded. Burger, C. J., delivered the opinion of the Court, in which Brennan, Stewart, White, Marshall, Powell, and Stevens, JJ., joined. Powell, 20 OCTOBER TERM, 1977 Opinion of the Court 437 U. S. J., filed a concurring opinion, post, p. 27. Rehnquist, J., filed an opinion concurring in the judgment, post, p. 27. Blackmun, J., took no part in the consideration or decision of the case. John T. Chandler argued the cause for petitioner pro hoc vice. With him on the briefs was Donald C. Peters. Harry M. Hipler, Assistant Attorney General of Florida, argued the cause for respondent pro hoc vice. With him on the brief were Robert L. Shevin, Attorney General, and Basil S. Diamond, Assistant Attorney General. Mr. Chief Justice Burger delivered the opinion of the Court. We granted certiorari to decide whether a State may retry a defendant after his conviction has been reversed by an appellate court on the ground that the evidence introduced at the prior trial was insufficient, as a matter of law, to sustain the jury’s verdict. I On September 7, 1965, petitioner Greene and José Manuel Sosa were indicted by a Florida grand jury for the murder of Nicanor Martinez. The indictment charged that Sosa “did hire, procure, aid, abet and counsel” Greene to murder Martinez and that petitioner had carried out the premeditated plan, shooting the victim to death with a pistol. A state-court jury subsequently found the defendants guilty of first-degree murder, without a recommendation of mercy. Pursuant to Florida law at the time, the trial court sentenced both defendants to death. On appeal to the Florida Supreme Court, the convictions of Greene and Sosa were reversed and new trials ordered. The reviewing court was sharply divided, however, with a majority composed of four justices joining a brief per curiam opinion which disposed of the case in the following terms : “After a careful review of the voluminous evidence here we are of the view that the evidence was definitely lacking GREENE v. MASSEY 21 19 Opinion of the Court in establishing beyond a reasonable doubt that the defendants committed murder in the first degree, and that the interests of justice require a new trial. The judgments are accordingly reversed and remanded for a new trial.” Sosa v. State, 215 So. 2d 736, 737 (1968). (Emphasis added.) Three justices dissented without opinion; we can do no more than speculate that the dissenting justices concluded there was sufficient evidence to support the jury verdict. In addition, a separate “special concurrence” was filed on behalf of three of the four justices who had also joined the per curiam opinion remanding for a new trial. These three concurring justices undertook a detailed examination of various asserted trial errors and found that on at least one claim the trial court had committed reversible error.1 This point concerned the improper admission of certain hearsay evidence which, in the opinion of the concurring justices, had a “potential probative force” that could have been “highly incriminating or critical to the establishment of an ultimate fact in dispute.” Id., at 745. While the concurrence of the three justices makes no mention of evidentiary insufficiency as such, the opinion concludes: “For the reasons stated the judgments should be reversed and remanded for a new trial so we have agreed to the Per Curiam order doing so.” Id., at 746. The “reasons stated” by the concurring justices thus concerned trial error, but paradoxically, the three explicitly joined the court’s per curiam opinion which rested exclusively on the 1 The concurrence also concluded that the trial court had improperly ruled on a question concerning a subpoena duces tecum, the result of which was that the defense may have been deprived of evidence to which it was entitled. It is not clear from the opinion whether the concurring justices would have regarded this error, in and of itself, as requiring reversal. 22 OCTOBER TERM, 1977 Opinion of the Court 437 U. S. ground that the evidence was insufficient to support the verdict. The case was then remanded, and after some intervening procedural maneuvering, the defendants were ordered retried in the Circuit Court of Orange County, Fla. Prior to their second trial, however, the defendants filed a suggestion for a writ of prohibition, claiming that their retrial would violate the Double Jeopardy Clause of the Federal Constitution, as it was applied to the States by Benton v. Maryland, 395 U. S. 784 (1969). They contended that the per curiam opinion of the State Supreme Court was tantamount to a finding that thè trial court should have directed a verdict of not guilty and hence a second trial for first-degree murder would constitute double jeopardy. When the trial court refused to issue the writ, review was sought in the Second District Court of Appeal of Florida. That court likewise declined to issue a writ of prohibition, but expressly stated that it was not rendering “an opinion as to the propriety of a new trial after a reversal for lack of sufficient evidence to establish, as a matter of law, the essential elements of the crime charged.” Sosa v. Maxwell, 234 So. 2d 690, 692 (1970). Rather, the District Court of Appeal was of the view that the Supreme Court’s reversal “appear [ed] to be based on a finding that the evidence, though technically sufficient, [was] so tenuous as to prompt an appellate court to exercise its discretion and, in the interest of justice, grant a new trial.” Id., at 691.2 Considering the case 2 The District Court of Appeal noted that “on many occasions” Florida courts had “held that where the weight of evidence appears ... to be very weak, although apparently legally sufficient if all permissible inferences are made and certain witnesses believed or disbelieved, a new trial may be granted.” 234 So. 2d, at 691. That court construed the language in the per curiam opinion of the State Supreme Court “as indicating that although some evidence on all elements of the crime was present, a grave doubt that affirmance would be in the interests of justice was raised in the minds of those members of the supreme court joining in the per curiam decision.” Id., at 691 n. 1. GREENE v. MASSEY 23 19 Opinion of the Court in this posture, the court indicated that it could find no precedent in Florida law which would bar a retrial on double jeopardy grounds.3 Certiorari was subsequently sought in the Supreme Court of Florida, which denied the petition without comment. 240 So. 2d 640 (1970). Greene and Sosa were then retried. On January 15, 1972, they were convicted of first-degree murder and each received a life sentence,, the second jury having recommended mercy. From this judgment they appealed to the Fourth District Court of Appeal of Florida, raising again their contention that the second trial violated the Double Jeopardy Clause. While conceding “the point to be academically intriguing,” Greene v. State, 302 So. 2d 202, 203 (1974), that court refused to reach the merits of the double jeopardy claim, holding instead that the Court of Appeal’s earlier disposition of the issue was res judicata. Greene and Sosa applied for a writ of certiorari in this Court and certiorari was denied. Greene n. Florida, 421 U. S. 932 (1975). Having exhausted all avenues of direct relief, petitioner Greene4 applied for a writ of habeas corpus in the United States District Court, arguing once more that his second trial was held in violation of the Double Jeopardy Clause. Although the District Court was sympathetic to petitioner’s claim,5 it felt constrained by prior Fifth Circuit precedent to 3 Although the District Court of Appeal thus failed to decide whether the State might retry a defendant after his conviction has been reversed on the ground that the evidence was insufficient to support the verdict, it did opine in dictum that in such circumstances “the trial judge should have directed a verdict of acquittal.” Id., at 692. 4 Sosa was not a party to the federal habeas corpus action; accordingly, our holding here has no effect on his conviction. 5 In its unreported order dismissing the petition, the District Court stated that “if this were a question of first impression in the Fifth Circuit, this Court might be inclined to grant the petition. Regardless of whether an appellate court or a trial jury makes the determination that the evidence is insufficient to sustain a finding of guilt as to a particular charge, 24 OCTOBER TERM, 1977 Opinion of the Court 437 IL S. dismiss the petition. From this ruling petitioner appealed to the Court of Appeals, which affirmed the District Court on the basis of an earlier Fifth Circuit case, United States v. Musquiz, 445 F. 2d 963 (1971). 546 F. 2d 51 (1977). The Musquiz decision had interpreted several of this Court’s cases6 to mean that under 28 U. S. C. § 2106, a court of appeals could order a new trial after a conviction had been reversed due to evidentiary insufficiency “if a motion for a new trial was made in the trial court.” 546 F. 2d, at 56. Noting that Greene had made a motion for a new trial after his first conviction, and that the Florida Supreme Court had “review power at least equal to that possessed by this Court [of Appeals] under § 2106,” ibid., the court held that a new trial had been a constitutionally permissible remedy. We granted certiorari, 432 U. S. 905 (1977), to review the judgment of the United States Court of Appeals. II In Burks v. United States, ante, p. 1, decided today, we have held that the Double Jeopardy Clause precludes a second trial once a reviewing court has determined that the evidence introduced at trial was insufficient to sustain the verdict. Since the constitutional prohibition against double jeopardy is fully applicable to state criminal proceedings, Benton v. Maryland, supra, we are bound to apply the standard announced in Burks to the case now under review. If we were confronted only with the per curiam opinion of the Florida Supreme Court, reversal in this case would follow. and regardless of whether a petitioner moves for a new trial on other grounds in addition to asserting the ground of insufficiency of evidence, it would seem that the double jeopardy clause would preclude giving the prosecution a second chance.” 6These included Forman v. United States, 361 IL S. 416 (I960); Sapir v. United States, 348 IL S. 373 (1955); Bryan v. United States, 338 IL S. 552 (1950). GREENE v. MASSEY 25 19 Opinion of the Court The per curiam disposition, standing by itself, leaves no room for interpretation by us other than that a majority of the State Supreme Court was “of the view that the evidence was definitely lacking in establishing beyond a reasonable doubt that the defendants committed murder in the first degree.,. By using the precise terminology “lacking in establishing beyond a reasonable doubt,” the highest court in Florida seems to have clearly said that there was insufficient evidence to permit the jury to convict petitioner at his first trial.7 The dispositive per curiam opinion makes no reference to the trial errors raised on appeal. Viewed in this manner, the reasoning enunciated in Burks would obviously compel the conclusion that Greene’s second trial violated the Double Jeopardy Clause. But the situation is confused by the fact that three of the four justices who joined in the per curiam disposition expressly qualified their action by “specially concurring” in an opinion which discussed only trial error. One could interpret this action to mean that the three concurring justices were con 7 Arguably, the per curiam opinion might be read as meaning that although there was insufficient evidence to convict the defendants of “murder in the first degree,” there was nonetheless evidence to support a conviction for a lesser included offense, e. g., second-degree murder, see Fla. Stat. §782.04 (1977). At the time of the Florida Supreme Court’s holding in this case, the Double Jeopardy Clause was not applicable to state proceedings, and hence that court conceivably did not see any need to consider whether, under the Federal Constitution, a retrial would be allowed only for some lesser included offense. Cf. Green n. United States, 355 U. S. 184 (1957). Indeed, even if Benton v. Maryland, 395 U. S. 784 (1969), had been decided prior to the State Supreme Court’s action, the Florida court might have reasonably concluded from our decisions that a retrial for first-degree murder was permissible under the Double Jeopardy Clause. See Burks, ante, at 10. Given our decision today to remand this case for reconsideration by the Court of Appeals, we need not reach the question of whether the State could, consistent with the Double Jeopardy Clause, try Greene for a lesser included offense in the event that his first-degree murder conviction is voided. 26 OCTOBER TERM, 1977 Opinion of the Court 437 U. S. cerned simply with trial error and joined in the remand solely to afford Greene and Sosa a fair, error-free trial—even though they were satisfied that the evidence was sufficient to support the verdict. A reversal grounded on such a holding, of course, would not prevent a retrial.8 See Burks, ante, at 15-16; United States v. Tateo, 377 U. S. 463,465 (1964). The problem with this interpretation is that the opinion concludes by expressly stating that the three concurring justices had “agreed to the Per Curiam order . . . .” When the concurrence is considered in light of the language of the per curiam opinion, it could reasonably be said that the concurring justices thought that the legally competent evidence adduced at the first trial was insufficient to prove guilt. That is, they were of the opinion that once the inadmissible hearsay evidence was discounted, there was insufficient evidence to permit the jury to convict? Given the varying interpretations10 that can be placed on the actions of the several Florida appellate courts, we conclude that this case should be remanded to the Court of Appeals for reconsideration in light of this opinion and Burks v. United 8 Even if this view of the concurrence is accepted, it would still mean that only a plurality of the Florida Supreme Court embraced the conclusion that reversal was justified solely on trial-error grounds. We leave resolution of this ambiguity to the Court of Appeals on remand, which will undoubtedly be in a better position to understand how Florida law would construe such a disposition. 9 We express no opinion as to the double jeopardy implications of a retrial following such a holding. 10 We note that the Second District Court of Appeal attached still another interpretation to the Florida Supreme Court’s action, namely, that a new trial was being granted “in the interests of justice,” even though the evidence was technically sufficient to support a verdict of guilty. See supra, at 22 n. 2. We are unaware, however, of the amount of weight that Florida law would afford to a district court of appeal’s interpretation of its Supreme Court’s actions. Nor are we willing to express an opinion as to the double jeopardy implications of a retrial ordered on such grounds. We leave both of these considerations to the Court of Appeals on remand. GREENE v. MASSEY 27 19 Rehnquist, J., concurring in judgment States, ante, p. 1. The Court of Appeals will be free to direct further proceedings in the District Court or to certify unresolved questions of state law to the Florida Supreme Court. See Fla. Stat. § 25.031 (1977), Fla. App. Rule 4.61; Lehman Bros. v. Schein, 416 U. S. 386 (1974). Reversed and remanded. Mr. Justice Blackmun took no part in the consideration or decision of this case. Mr. Justice Powell, concurring. I concur in the opinion of the Court except insofar as it states that the constitutional prohibition against double jeopardy is fully applicable to state criminal proceedings. See Crist v. Bretz, post, p. 40 (Powell, J., dissenting). I believe, however, that under our decision today in Burks v. United States, ante, p. 1, a fundamental component of the prohibition against double jeopardy is the right not to be retried once an appellate court has found the evidence insufficient as a matter of law to support the jury’s guilty verdict. Mr. Justice Rehnquist, concurring in the judgment. For the reasons stated by Mr. Justice Powell in his dissenting opinion in Crist v. Bretz, post, p. 40, I do not agree with the Court’s premise, ante, at 24, that “the constitutional prohibition against double jeopardy is fully applicable to state criminal proceedings.” Even if I did agree with that view, I would want to emphasize more than the Court does in its opinion the varying practices with respect to motions for new trial and other challenges to the sufficiency of the evidence both at the trial level and on appeal in the 50 different States in the Union. Thus, to the extent that Florida practice in this regard differs from practice in the federal system, the impact of the Double Jeopardy Clause may likewise differ with respect to a particular proceeding. I therefore concur only in the Court’s judgment. 28 OCTOBER TERM, 1977 437 U.S. Syllabus CRIST, WARDEN, et al. v. BRETZ et al. APPEAL FROM THE UNITED STATES COURT OF APPEALS FOR THE NINTH CIRCUIT No. 76-1200. Argued November 1, 1977—Reargued March 22, 1978— Decided June 14, 1978 The federal rule that jeopardy attaches in a jury trial when the jury is empaneled and sworn, a rule that reflects and protects the defendant’s interest in retaining a chosen jury, is an integral part of the Fifth Amendment guarantee against double jeopardy made applicable to the States by the Fourteenth Amendment. Hence, a Montana statute providing that jeopardy does not attach until the first witness is sworn cannot constitutionally be applied in a jury trial. Pp. 32-38. 546 F. 2d 1336, affirmed. Stewart, J., delivered the opinion of the Court, in which Brennan, White, Marshall, Blackmun, and Stevens, JJ., joined. Blackmun, J., filed a concurring opinion, post, p. 38. Burger, C. J., filed a dissenting opinion, post, p. 39. Powell, J., filed a dissenting opinion, in which Burger, C. J., and Rehnquist, J., joined, post, p. 40. Robert S. Keller, Special Assistant Attorney General of Montana, reargued the cause for appellants. With him on the briefs was Michael T. Greely, Attorney General. W. William Leaphart, by appointment of the Court, 431 U. S. 963, reargued the cause and filed briefs for appellee Cline. Charles F. Moses reargued the cause and filed briefs for appellee Bretz. Kenneth S. Geller argued the cause on the reargument for the United States as amicus curiae urging reversal. On the brief were Solicitor General McCree, Assistant Attorney General Civiletti, Deputy Solicitor General Easterbrook, and Alan J. Sobol. CRIST v. BRETZ 29 28 Opinion of the Court Mr. Justice Stewart delivered the opinion of the Court. This case involves an aspect of the constitutional guarantee against being twice put in jeopardy. The precise issue is whether the federal rule governing the time when jeopardy attaches in a jury trial is binding on Montana through the Fourteenth Amendment. The federal rule is that jeopardy attaches when the jury is empaneled and sworn; a Montana statute provides that jeopardy does not attach until the first witness is sworn.1 I The appellees, Merrel Cline2 and L. R. Bretz, were brought to trial in a Montana court on charges of grand larceny, obtaining money and property by false pretenses, and several counts of preparing or offering false evidence. A jury was empaneled and sworn following a three-day selection process. Before the first witness was sworn, however, the appellees filed a motion drawing attention to the allegation in the 1 Montana Rev. Codes Ann. §95-1711 (3) (1947) provides in pertinent part: “[A] prosecution based upon the same transaction as a former prosecution is barred by such former prosecution under the following circumstances: ... (d) The former prosecution was improperly terminated. Except as provided in this subsection, there is an improper termination of a prosecution if the termination is for reasons not amounting to an acquittal, and it takes place after the first witness is sworn but before verdict. . . .” See also State n. Cunningham, 166 Mont. 530, 535-536, 535 P. 2d 186, 189. In addition to Montana, Arizona also holds that jeopardy does not attach until “proceedings commence,” although this may be as early as the opening statement. Klinefelter v. Superior Court, 108 Ariz. 494, 495, 502 P. 2d 531, 532; State n. Mojarro Padilla, 107 Ariz. 134, 139-140, 483 P. 2d 549, 553. Until recently, New York had a similar rule. See Mizell n. Attorney General, 442 F. Supp. 868 (EDNY). 2 We were informed during argument that the conviction of Merrel Cline has been reversed, see State v. Cline, 170 Mont. 520, 555 P. 2d 724, and the charges against him dismissed. This appeal, therefore, has become moot as to him. 30 OCTOBER TERM, 1977 Opinion of the Court 437 U. S. false-pretenses charge that the defendants’ illegal conduct began on January 13, 1974.3 Effective January 1, 1974, the particular statute relied on in that count of the information, Mont. Rev. Codes Ann. § 94r-1805 (1947), had been repealed. The prosecutor moved to amend the information, claiming that “1974” was a typographical error, and that the date on which the defendants’ alleged violation of the statute had commenced was actually January 13, 1973, the same date alleged in the grand larceny count. The trial judge denied the prosecutor’s motion to amend the information and dismissed the false-pretenses count. The State promptly but unsuccessfully asked the Montana Supreme Court for a writ of supervisory control ordering the trial judge to allow the amendment. Returning to the trial court, the prosecution then asked the trial judge to dismiss the entire information so that a new one could be filed. That motion was granted, and the jury was dismissed. A new information was then filed, charging the appellees with grand larceny and obtaining money and property by false pretenses. Both charges were based on conduct commencing January 13,1973. Other than the change in dates, the new false-pretenses charge described essentially the same offense charged in the earlier defective count. After a second jury had been selected and sworn, the appellees moved to dismiss the new information, claiming that the Double Jeopardy Clauses of the United States and Montana Constitutions barred a second prosecution. The motion was denied, and the trial began. The appellees were found guilty on the false-pretenses count, and sentenced to terms of imprisonment. The Montana Supreme Court, which had previously denied appellees habeas corpus relief, State ex rel. Bretz v. Sheriff, 167 Mont. 363, 539 P. 2d 1191, affirmed the judgment as to Bretz on the ground that under state law 3 The motion asked that the prosecution’s evidence be limited to the time period alleged in the information. CRIST v. BRETZ 31 28 Opinion of the Court jeopardy had not attached in the first trial. State v. Cline, 170 Mont. 520, 555 P. 2d 724. In the meantime the appellees had brought a habeas corpus proceeding in a Federal District Court, again alleging that their convictions had been unconstitutionally obtained because the second trial violated the Fifth and Fourteenth Amendment guarantee against double jeopardy. The federal court denied the petition, holding that the Montana statute providing that jeopardy does not attach until the first witness is sworn does not violate the United States Constitution. The court held in the alternative that even if jeopardy had attached, a second prosecution was justified, as manifest necessity supported the first dismissal. Cunningham v. District Court, 406 F. Supp. 430 (Mont.).4 The Court of Appeals for the Ninth Circuit reversed. 546 F. 2d 1336. It held that the federal rule governing the time when jeopardy attaches is an integral part of the constitutional guarantee, and thus is binding upon the States under the Fourteenth Amendment. The appellate court further held that there had been no manifest necessity for the Montana trial judge’s dismissal of the defective count, and, accordingly, that a second prosecution was not constitutionally permissible.5 Appellants appealed pursuant to 28 U. S. C. § 1254 (2), seeking review only of the holding of the Court of Appeals that Montana is constitutionally required to recognize that, for purposes of the constitutional guarantee against double jeopardy, jeopardy attaches in a criminal trial when the jury is empaneled and sworn. We postponed consideration of probable jurisdiction sub nom. Crist v. Cline, 430 U. S. 982, and the case was argued. Thereafter the case was set for 4 The Cunningham case, involving the same issue, was consolidated with the appellees’ case. 5 In this Court the appellants specifically waived any challenge to the Court of Appeals’ ruling on manifest necessity, and we intimate no view as to its correctness. 32 OCTOBER TERM, 1977 Opinion of the Court 437 U. S. reargument, 434 U. S. 980, and the parties were asked to address the following two questions: “1. Is the rule heretofore applied in the federal courts— that jeopardy attaches in jury trials when the jury is sworn—constitutionally mandated? “2. Should this Court hold that the Constitution does not require jeopardy to attach in any trial—state or federal, jury or non jury—until the first witness is sworn?” II A The unstated premise of the questions posed on reargument is that if the rule “that jeopardy attaches in jury trials when the jury is sworn” is “constitutionally mandated,” then that rule is binding on Montana, since “the double jeopardy prohibition of the Fifth Amendment . . . [applies] to the States through the Fourteenth Amendment,” and “the same constitutional standards” must apply equally in federal and state courts. Benton v. Maryland, 395 U. S. 784, 794-795. The single dispositive question, therefore, is whether the federal rule is an integral part of the constitutional guarantee. The Double Jeopardy Clause of the Fifth Amendment is stated in brief compass: “[N]or shall any person be subject for the same offence to be twice put in jeopardy of life or limb.” But this deceptively plain language has given rise to problems both subtle and complex, problems illustrated by no less than eight cases argued here this very Term.6 This case, however, presents a single straightforward issue concerning the point during a jury trial when a defendant is deemed to have been put in jeopardy, for only if that point has once been 6 In addition to the present case, see Arizona v. Washington, 434 U. S. 497; United States v. Wheeler, 435 U. S. 313; Burks v. United States, ante, p. 1; Greene v. Massey, ante, p. 19; Sanabria v. United States, post, p. 54; Swisher v. Brady, No. 77-653; United States v. Scott, post, p. 82. CRIST v. BRETZ 33 28 Opinion of the Court reached does any subsequent prosecution of the defendant bring the guarantee against double jeopardy even potentially into play. Serfass v. United States, 420 U. S. 377, 388; Illinois v. Somerville, 410 U. S. 458, 467. The Fifth Amendment guarantee against double jeopardy derived from English common law, which followed then, as it does now,7 the relatively simple rule that a defendant has been put in jeopardy only when there has been a conviction or an acquittal—after a complete trial.8 A primary purpose served by such a rule is akin to that served by the doctrines of res judicata and collateral estoppel—to preserve the finality of judgments.9 And it is clear that in the early years of our national history the constitutional guarantee against double jeopardy was considered to be equally limited in scope. As Mr. Justice Story explained: “[The Double Jeopardy Clause] does not mean, that [a person] shall not be tried for the offence a second time, if the jury shall have been discharged without giving any verdict; . . . for, in such a case, his life or limb cannot judicially be said to have been put in jeopardy.” 3 J. Story, Commentaries on the Constitution § 1781, pp. 659-660 (1833). But this constitutional understanding was not destined to endure. Beginning with this Court’s decision in United 711 Halsbury’s Laws of England If 242 (4th ed. 1976). 8 Established at least by 1676, Turner’s Case, 89 Eng. Rep. 158, the rule was embodied in defensive pleas of former conviction or former acquittal. Although the pleas did not mention jeopardy, Blackstone commented that they were based on the “universal maxim . . . that no man is to be brought into jeopardy of his life, more than once, for the same offence.” 4 W. Blackstone, Commentaries *335. See generally J. Sigler, Double Jeopardy 1-37 (1969). 9 See Mayers & Yarbrough, Bis Vexari: New Trials and Successive Prosecutions, 74 Harv. L. Rev. 1 (1960). See also M. Friedland, Double Jeopardy 6 (1969); ALI, Administration of the Criminal Law: Double Jeopardy 7 (1935). 34 OCTOBER TERM, 1977 Opinion of the Court 437 U. S. States v. Perez, 9 Wheat. 579, it became firmly established by the end of the 19th century that a defendant could be put in jeopardy even in a prosecution that did not culminate in a conviction or an acquittal, and this concept has been long established as an integral part of double jeopardy jurisprudence.10 Thus in Wade v. Hunter, 336 U. S. 684, 688, the Court was able accurately to say: “Past cases have decided that a defendant, put to trial before a jury, may be subjected to the kind of ‘jeopardy’ that bars a second trial for the same 10 In perhaps the first expression of this concept, a state court in 1822 concluded that jeopardy may attach prior to a verdict, because “[t]here is a wide different between a verdict given and the jeopardy of a verdict.” Commonwealth v. Cook, 6 Serg. & R. 577, 596 (Pa.). In the Perez case, the trial judge had discharged a deadlocked jury, and the defendant argued in this Court that the discharge was a bar to a second trial. The case has long been understood as standing for the proposition that jeopardy attached during the first trial, but that despite the former jeopardy a second trial was not barred by the Double Jeopardy Clause because there was a “manifest necessity” for the discharge of the first jury. See, e. g., United States v. Tateo, 377 U. S. 463,467; Wade v. Hunter, 336 U. S. 684, 689-690. In fact, a close reading of the short opinion in that case could support the view that the Court was not purporting to decide a constitutional question, but simply settling a problem arising in the administration of federal criminal justice. But to cast suoh a new light on Perez at this late date would be of academic interest only. In two cases decided in the wake of Perez the Court simply followed its precedential authority: Simmons v. United States, 142 IT. S. 148; Thompson v. United States, 155 U. S. 271. But it had become clear at least by the time of Kepner v. United States, 195 U. S. 100, decided in 1904, that jeopardy does attach even in a trial that does not culminate in a jury verdict: “[A] person has been in jeopardy when he is regularly charged with a crime before a tribunal properly organized and competent to try him .... Undoubtedly in those jurisdictions where a trial of one accused of crime can only be to a jury, and a verdict of acquittal or conviction must be by a jury, no legal jeopardy can attach until a jury has been called and charged with the deliverance of the accused.” Id., at 128. See also United States v. Dinitz, 424 U. S. 600; United States v. Wilson, 420 U. S. 332, 343-344; Gori v. United States, 367 U. S. 364. CRIST v. BRETZ 35 28 Opinion of the Court offense even though his trial is discontinued without a verdict.” See also, e. g., Arizona v. Washington, 434 U. S. 497. The basic reason for holding that a defendant is put in jeopardy even though the criminal proceeding against him terminates before verdict was perhaps best stated in Green n. United States, 355 U. S. 184, 187-188: “The underlying idea, one that is deeply ingrained in at least the Anglo-American system of jurisprudence, is that the State with all its resources and power should not be allowed to make repeated attempts to convict an individual for an alleged offense, thereby subjecting him to embarrassment, expense and ordeal and compelling him to live in a continuing state of anxiety and insecurity, as well as enhancing the possibility that even though innocent he may be found guilty.” Although it has thus long been established that jeopardy may attach in a criminal trial that ends inconclusively, the precise point at which jeopardy does attach in a jury trial might have been open to argument before this Court’s decision in Downum v. United States, 372 U. S. 734.11 There the Court held that the Double Jeopardy Clause prevented a second prosecution of a defendant whose first trial had ended just after the jury had been sworn and before any testimony had been taken. The Court thus necessarily pinpointed the stage in a jury trial when jeopardy attaches, and the Downum case has since been understood as explicit authority for the proposition that jeopardy attaches when the jury is empaneled and sworn. See United States v. Martin Linen Supply Co., 430 U. S. 564, 569; Serfass v. United States, 420 U. S., at 388. The reason for holding that jeopardy attaches when the jury is empaneled and sworn lies in the need to protect the interest of an accused in retaining a chosen jury. That 11 But see Kepner n. United States, supra, at 128; n. 10, supra. 36 OCTOBER TERM, 1977 Opinion of the Court 437 U. S. interest was described in Wade v. Hunter, supra, as a defendant’s “valued right to have his trial completed by a particular tribunal.” 336 U. S., at 689. It is an interest with roots deep in the historic development of trial by jury in the Anglo-American system of criminal justice.12 Throughout that history there ran a strong tradition that once banded together a jury should not be discharged until it had completed its solemn task of announcing a verdict.13 Regardless of its historic origin, however, the defendant’s “valued right to have his trial completed by a particular tribunal” is now within the protection of the constitutional guarantee against double jeopardy, since it is that “right” that lies at the foundation of the federal rule that jeopardy attaches when the jury is empaneled and sworn. United States v. Martin Linen Supply Co., supra; Serfass v. United States, supra, at 388; Illinois v. Somerville, 410 U. S., at 467; United States v. Jorn, 400 U. S. 470, 478-480, 484-485 (plurality opinion). 12 Trial juries were at first merely a substitute for other inscrutable methods of decisionmaking, such as trial by battle, compurgation, and ordeal. See 1 W. Holdsworth, A History of English Law 317 (7th ed. 1956). See also T. Plucknett, A Concise History of the Common Law 125 (5th ed. 1956). They soon evolved, however, into a more rational instrument of decisionmaking—serving as a representative group of peers to sit in judgment on a defendant’s guilt. 13 Illustrative of this tradition was the practice of keeping the jury together unfed and without drink until it delivered its unanimous verdict. See Y. B. Trin. 14 Hen. VII, pl. 4. See Plucknett, supra, at 119. As Lord Coke put the matter: “A jury sworn and charged in case of life or member, cannot be discharged by the court or any other, but they ought to give a verdict.” 1 E. Coke, Institutes 227 (b) (6th ed. 1861). And an English court said as late as 1866: “ [The rule] seems to command the confinement of the jury till death if they do not agree, and to avoid any such consequence an exception was introduced in practice which Blackstone has described by the words 'except in case of evident necessity.’ ” Winsor v. The Queen, [1866] 1 Q. B. 390, 394. CRIST v. BRETZ 37 28 Opinion of the Court B It follows that Montana’s view as to when jeopardy attaches is impermissible under the Fourteenth Amendment unless it can be said that the federal rule is not “at the core” of the Double Jeopardy Clause. See Pointer v. Texas, 380 U. S. 400, 406; Malloy v. Hogan, 378 U. S. 1, 11; Ker v. California, 374 IT. S. 23, 33. In asking us to hold that it is not, appellants argue that the federal standard is no more than an arbitrarily chosen rule of convenience,14 similar in its lack of constitutional status to the federal requirement of a unanimous verdict by 12 jurors, which has been held not to bind the States. Apodaca v. Oregon, 406 U. S. 404; Williams v. Florida, 399 U. S. 78. But see Ballew v. Georgia, 435 IT. S. 223. If the rule that jeopardy attaches when the jury is sworn were simply an arbitrary exercise of linedrawing, this argument might well be persuasive, and it might reasonably be concluded that jeopardy does not constitutionally attach until the first witness is sworn, to provide consistency in jury and non jury trials.15 Indeed, it might then be concluded that the point of the attachment of jeopardy- could be moved a few steps forward or backward without constitutional significance.16 But the federal rule as to when jeopardy attaches in a jury 14 The United States as amicus curiae makes a similar argument. 15 In nonjury trials jeopardy does not attach until the first witness is sworn. Serfass v. United States, 420 U. S. 377, 388. 16 The United States alternatively proposes a due process sliding “interest balancing test” under which the further the trial has proceeded the more the justification required for a midtrial termination. Montana alternatively proposes that jeopardy should not be held to attach until a prima facie case has been made, on the premise that only then will a defendant truly be in jeopardy. The legal literature provides at least one other approach: jeopardy should attach “as soon as the process of selecting the jury begins.” See Schulhofer, Jeopardy and Mistrials, 125 U. Pa. L. Rev. 449, 512-514 (1977). 38 OCTOBER TERM, 1977 Blackmun, J., concurring 437 U. S. trial is not only a settled part of federal constitutional law. It is a rule that both reflects and protects the defendant’s interest in retaining a chosen jury. We cannot hold that this rule, so grounded, is only at the periphery of double jeopardy concerns. Those concerns—the finality of judgments, the minimization of harassing exposure to the harrowing experience of a criminal trial, and the valued right to continue with the chosen jury—have combined to produce the federal law that in a jury trial jeopardy attaches when the jury is empaneled and sworn. We agree with the Court of Appeals that the time when jeopardy attaches in a jury trial “serves as the lynchpin for all double jeopardy jurisprudence.” 546 F. 2d, at 1343. In Illinois v. Somerville, supra, at 467, a case involving the application of the Double Jeopardy Clause through the Fourteenth Amendment, the Court said that “jeopardy ‘attached’ when the first jury was selected and sworn.” Today we explicitly hold what Somerville assumed: The federal rule that jeopardy attaches when the jury is empaneled and sworn is an integral part of the constitutional guarantee against double jeopardy. The judgment is Affirmed. Mr. Justice Blackmun, concurring. Although I join the Court’s opinion, I write to emphasize the fact that I am not content to rest the result, as the Court seems to be, ante, at 36, solely on the defendant’s “valued right to have his trial completed by a particular tribunal,” a factor mentioned by Mr. Justice Black, speaking for the Court, in Wade v. Hunter, 336 U. S. 684, 689 (1949). That approach would also support a conclusion that jeopardy attaches at the very beginning of the jury selection process. See Schulhofer, Jeopardy and Mistrials, 125 U. Pa. L. Rev. 449, 512-514 (1977). Other interests are involved here as well: repetitive stress CRIST v. BRETZ 39 28 Burger, C. J., dissenting and anxiety upon the defendant; continuing embarrassment for him; and the possibility of prosecutorial overreaching in the opening statement. It is perhaps true that each of these interests could be used, too, to support an argument that jeopardy attaches at some point before the jury is sworn. I would bring all these interests into focus, however, at the point where the jury is sworn because it is then and there that the defendant’s interest in the jury reaches its highest plateau, because the opportunity for prosecutorial overreaching thereafter increases substantially, and because stress and possible embarrassment for the defendant from then on is sustained. Mr. Chief Justice Burger, dissenting. As a “rulemaking” matter, the result reached by the Court is a reasonable one; it is the Court’s decision to constitutionalize the rule that jeopardy attaches at the point when the jury is sworn—so as to bind the States—that I reject. This is but another example of how constitutional guarantees are trivialized by the insistence on mechanical uniformity between state and federal practice. There is, of course, no reason why the state and federal rules must be the same. In the period between the swearing of the jury and the swearing of the first witness, the concerns underlying the constitutional guarantee against double jeopardy are simply not threatened in any meaningful sense even on the least sanguine of assumptions about prosecutorial behavior. We should be cautious about constitutionalizing every procedural device found useful in federal courts, thereby foreclosing the States from experimentation with different approaches which are equally compatible with constitutional principles. All things “good” or “desirable” are not mandated by the Constitution. States should remain free to have procedures attuned to the special problems of the criminal justice system at the state and local levels. Principles of federalism should not so readily be com 40 OCTOBER TERM, 1977 Powell, J., dissenting 437U.S. promised for the sake of a uniformity finding sustenance perhaps in considerations of convenience but certainly not in the Constitution. Countless times in the past 50 years this Court has extolled the virtues of allowing the States to serve as “laboratories” to experiment with procedures which differ from those followed in the federal courts. Yet we continue to press the States into a procrustean federal mold. The Court’s holding will produce no great mischief, but it continues, I repeat, the business of trivializing the Constitution on matters better left to the States. Accordingly, I join Mr. Justice Powell’s dissent. Mr. Justice Powell, with whom The Chief Justice and Mr. Justice Rehnquist join, dissenting. The rule that jeopardy attaches in a jury trial at the moment the jury is sworn is not mandated by the Constitution. It is the product of historical accident, embodied in a Court decision without the slightest consideration of the policies it purports to serve. Because these policies would be served equally well by a rule fixing the attachment of jeopardy at the swearing of the first witness, I would uphold the Montana statute. Even if one assumed that the Fifth Amendment now requires the attachment of jeopardy at the swearing of the jury, I would view that rule as incidental to the purpose of the Double Jeopardy Clause and hence not incorporated through the Due Process Clause of the Fourteenth Amendment and not applicable to the States. I therefore dissent. I As the Court correctly observes, ante, at 33, it is clear that in the early years of our national history the constitutional guarantee against double jeopardy was restricted to cases in which there had been a complete trial—culminating in acquittal or conviction. The limited debate on the Double Jeopardy Clause in the House of Representatives confirms this proposi CRIST v. BRETZ 41 28 Powell, J., dissenting tion. 1 Annals of Cong. 753 (1789). See generally United, States v. Wilson, 420 U. S. 332, 339-342 (1975). This was consonant with the prevailing English practice regarding pleas in bar. The pleas of autrefois acquit and autrefois convict, which implemented the maxim, repeated by Blackstone, that no man should twice be placed in jeopardy for the same offense,1 could be interposed only on the basis of an actual verdict of acquittal or conviction.2 It was to these pleas in bar—which embody a res judicata policy, as the Court describes it, ante, at 33—that the Double Jeopardy Clause was directed. See, e. g., United States v. Haskell, 26 F. Cas. 207, 212 (No. 15,321) (CC Pa. 1823) (Washington, J.); People v. Goodwin, 18 Johns. 187, 205 (N. Y. Sup. Ct. 1820); cf. People v. Olcott, 2 Johns. Cas. 301 (N. Y. Sup. Ct. 1801) (Kent, J.). This remains the English rule. See n. 2, supra. But there existed a separate rule of English practice that has become intertwined with the doctrine of pleas in bar in the development of our Double Jeopardy Clause. This was the rule, based upon a dictum of Lord Coke, that once the “[j]uiy is retorned and sworn, their verdict must be heard, and they cannot be discharged . . . .” 3 E. Coke, Institutes 110 (6th ed. 1681); accord, 1 id., at 227 (b). That this rule arose as an aspect of jury practice, rather than as an element of the guarantee against double jeopardy, is supported by several facts. First, it applied in civil cases as well as criminal. Kirk, “Jeopardy” During the Period of the Year Books, 82 IL Pa. L. Rev. 602, 609 (1934). Second, the early cases and treaties laid down no clear standard as to the effect of a failure to follow the rule. See, e. g., C. St. Germain, Doctor and Student 1531, Dialogue 2, ch. 52 (1970). Third, it seems never to have been pleaded successfully in bar of a second 14 W. Blackstone, Commentaries *335. See also 3 E. Coke, Institutes 213-214 (6th ed. 1681). 2 J. Archbold, Pleading, Evidence & Practice in Criminal Cases §§ 435-459 (35th ed. 1962). 42 OCTOBER TERM, 1977 Powell, J., dissenting 437 U. S. prosecution in the period of the Year Books, when the rule is said to have arisen. Kirk, supra, at 611. Fourth, Blackstone dealt with the rule governing the discharge of the jury not in his section on pleas in bar but in his discussion dealing with verdicts. Compare 4 W. Blackstone, Commentaries *335-*338, with id., at *360.3 Hence, it is reasonably clear that the rule forbidding discharge of the jury arose out of the circumstances of medieval England, “when jurors of the counties where the facts occurred were summoned to give testimony at Westminster on a trial based on those facts. It seems not to have been an invariable rule and has never been found to have had any connection, in the cases at English common law, with the problem of two trials for the same offense.” Kirk, supra, at 612 (footnote omitted). Notwithstanding its origin as an aspect of jury practice, the rule against discharge of the jury became a useful defense against Crown oppression in the 17th century. Reaction to the “tyrannical practice,” The Queen v. Charlesworth, 1 B. & S. 460, 500, 121 Eng. Rep. 786, 801 (Q. B. 1861), of discharging juries and permitting reindictment when acquittal appeared likely4 was so strong that the common-law judges 3 Interestingly, Blackstone wrote that the jury could not be discharged, not as soon as it was sworn, but only after evidence had been introduced. 4 W. Blackstone, Commentaries *360. A relatively recent edition of Blackstone, compiled from the earliest editions, indicates that the close of the evidence may have been the point at which the rule against discharge of the jury originally was fixed by that authority. J. Ehrlich, Ehrlich’s Blackstone 941 (1959). 4 2 M. Hale, Pleas of the Crown 294r-295 (W. Stokes & E. Ingersoll ed. 1847). In the infamous Ireland’s Case, 7 How. St. Tr. 79 (1678), five defendants were accused of high treason. The court permitted the jury to deliberate as to three defendants, but instructed the jury that the evidence against Whitebread and Fenwick was not sufficient to convict, even though “so full, as to satisfy a private conscience.” Id., at 121. The court therefore discharged the jury of those two, declaring that it would “be convenient, from what is already proved, to have them stay until more proof may come in.” Ibid. They were reindicted, convicted, and executed, CRIST v. BRETZ 43 28 Powell, J., dissenting declared “that in all capital cases, a juror cannot be withdrawn, though the parties consent to it; that in criminal cases, not capital, a juror may be withdrawn, if both parties consent, but not otherwise . . . .” The King v. Perkins, Holt. 403, 90 Eng. Rep. 1122 (K. B. 1698). Whether or not this strict rule was ever stringently applied, it was modified soon after it was announced. The King n. Kinloch, Post. 16, 168 Eng. Rep. 9 (K. B. 1746). In any event, it seems never to have furnished the basis for a plea of autrefois acquit. Rather, it was viewed as a matter committed to the discretion of the trial judge, from which no writ of error would lie nor any plea in bar of a future prosecution would be allowed. The Queen v. Winsor, 10 Cox C. C. 276, 313-323, 325-326 (Q. B. 1865); The Queen v. Charlesworth, supra, at 507-515, 121 Eng. Rep., at 803-806.5 Thus, while the English judges had adapted Lord Coke’s rule to the protection of interests later recognized in this country as within the sphere of the Double Jeopardy Clause, compare The Queen v. Winsor, supra, at 301-302, with Green v. United States, 355 U. S. 184, 187-188 (1957), they refused to import the rule into the realm of pleas in bar, and it was the latter which informed the framing of the Double Jeopardy Clause. But it was the common-law rule of jury practice—a rule that we well might have come to regard as an aspect of due process if it had not been absorbed in this country by the Whitebread’s Case, 7 How. St. Tr. 311 (1679), despite their pleas of former jeopardy, id., at 315-318. 5 In Conway and Lynch n. The Queen, 1 Ir. 149 (Q. B. 1845), the Irish Court of Queen’s Bench did review on writ of error the prisoners’ convictions after reindictment, holding that where the trial judge failed to state on the record thd condition of necessity which had prompted the discharge of the first jury, there was an abuse of discretion preventing subsequent trial. The English Court of Queen’s Bench, however, rejected this view in Charlesworth and in Winsor. Indeed, that court adopted the view of Justice Crampton, who had dissented in Conway and Lynch. 44 OCTOBER TERM, 1977 Powell, J., dissenting 437 U. S. Double Jeopardy Clause—with which this Court concerned itself in United States v. Perez, 9 Wheat. 579 (1824). Sitting on the Perez Court was Mr. Justice Washington, who one year earlier had written that “the jeopardy spoken of in [the Fifth Amendment] can be interpreted to mean nothing short of the acquittal or conviction of the prisoner, and the judgment of the court thereupon.” United States v. Haskell, 26 F. Cas., at 212. Mr. Justice Story authored the opinion of the Court in Perez. Nine years later he would explain in his treatise on the Constitution that the meaning of the Double Jeopardy Clause is “that a party shall not be tried a second time for the same offence, after he has once been convicted, or acquitted of the offence charged, by the verdict of a jury, and judgment has passed thereon for or against him.” 3 J. Story, Commentaries on the Constitution § 1781, p. 659 (1833).6 It seems most unlikely that either of these Members of the Perez Court thought that the decision was interpreting the Fifth Amendment when it declared that the discharge of a jury, before verdict, on grounds of “manifest necessity” was not a bar to a retrial.7 9 Wheat., at 580. As both Justices Washington and Story believed that the Double Jeopardy Clause embraced only actual acquittal and conviction, they must have viewed Perez as involving the independent rule barring needless dis 6 See also United States v. Coolidge, 25 F. Cas. 622 (No. 14,858) (CC Mass. 1815) (Story, J.). Despite the view clearly expressed in Mr. Justice Story’s Commentaries, there is some evidence that by the year following its publication he was beginning to consider the rule against discharge of the jury as embodying some double jeopardy concerns. See United States v. Gibert, 25 F. Cas. 1287, 1295-1296 (No. 15,204) (CC Mass. 1834). 7 That Perez was not concerned with pleas in bar—and therefore not with the Double Jeopardy Clause—is supported by its recognition of the doctrine of manifest necessity. No “necessity”—for example, discovery of incontrovertible evidence that a previously acquitted person was guilty— sufficed to overcome a valid plea in bar. Necessity went only to the propriety of discharging the jury. See United States v. Bigelow, 14 D. C. 393, 401-403 (1884). CRIST v. BRETZ 45 28 Powell, J., dissenting charges of the jury.8 The decisions of this Court throughout the 19th and early 20th centuries dealing with discharges of the jury are ambiguous, but can be read merely as reaffirming the principle of Perez that discharges before verdict may be justified by manifest necessity, without adding a Fifth Amendment gloss.9 Throughout the 19th century, however, many state courts began to blend the rule against needless discharges of juries into the guarantee against double jeopardy contained in the Federal and State Constitutions.10 It was recognized that the 8 The Court recognizes that Perez probably cannot be viewed as a double jeopardy case. Ante, at 34 n. 10. 9 Simmons v. United States, 142 U. S. 148 (1891); Logan v. United States, 144 U. S. 263 (1892); Thompson v. United States, 155 U. S. 271 (1894); Dreyer v. Illinois, 187 U. S. 71 (1902); Lovato v. New Mexico, 242 U. S. 199 (1916). See also United States v. Morris, 26 F. Cas. 1323 (No. 15,815) (CC Mass. 1851) (Curtis, J.). But see Keerl n. Montana, 213 U. S. 135 (1909); cf. Kepner v. United States, 195 U. S. 100, 128 (1904). See also United States v. Shoemaker, 27 F. Cas. 1067 (No. 16,279) (CC Ill. 1840); United States v. Watson, 28 F. Cas. 499 (No. 16,651) (SDNY 1868). 10 See, e. g., State v. Garrigues, 2 N. C. 188 (1795) (semble); Commonwealth v. Cook, 6 Serg. & R. 577 (Pa. 1822); State v. M’Kee, 1 Bailey 651 (S. C. 1830); Mahala n. State, 18 Tenn. 532 (1837); State v. Roe, 12 Vt. 93 (1840); Morgan v. State, 13 Ind. 215 (1859); People v. Webb, 38 Cal. 467 (1869); Nolan n. State, 55 Ga. 521 (1875); Teat v. State, 53 Miss. 439 (1876); Ex parte Maxwell, 11 Nev. 428, 435 (1876); Mitchell v. State, 42 Ohio St. 383 (1884); State v. Ward, 48 Ark. 36, 2 S. W. 191 (1886); People v. Gardner, 62 Mich. 307, 29 N. W. 19 (1886); Commonwealth n. Hart, 149 Mass. 7, 20 N. E. 310 (1889); State v. Paterno, 43 La. Ann. 514, 9 So. 442 (1891); McDonald n. State, 79 Wis. 651, 48 N. W. 863 (1891); State n. Sommers, 60 Minn. 90, 61 N. W. 907 (1895); Dulin v. Lillard, 91 Va. 718, 20 S. E. 821 (1895). But see, e. g., People v. Goodwin, 18 Johns. 187 (N. Y. Sup. Ct. 1820); Commonwealth v. Wade, 34 Mass. 395 (1835); Hoffman v. State, 20 Md. 425, 433 (1863); United States v. Bigeloxo, 14 D. C. 393 (1884); State v. Van Ness, 82 N. J. L. 181, 83 A. 195 (1912). American treatises also included the rule against discharge of the jury under the heading of Double Jeopardy. See M. Bigelow, Estoppel 36 (2d 46 OCTOBER TERM, 1977 Powell, J., dissenting 437 U. S. discharge rule provided significant protection against being twice vexed: “The right of trial by jury is of but little value to the citizen in a criminal prosecution against him if [the guarantee against double jeopardy] can be violated and the accused left without remedy. If the judge can arbitrarily discharge and impanel juries until one is obtained that will render such a verdict as the state demands, or the attorney for the prosecution desires, and the only protection against such oppression is that a new trial may be ordered in the court trying him, or by the court of last resort, then of what value is this boasted right?” O’ Brian v. Commonwealth, 72 Ky. 333, 339 (1873). Cf. Green v. United States, 355 U. S., at 187-188. Thus, the state courts were putting Lord Coke’s rule to a use similar to that of the 17th-century English judges, but they did so—with no apparent awareness of the novelty of their action—under the rubric of the Double Jeopardy Clause. Given this rather unreflective incorporation of a common-law rule of jury practice into the guarantee against double jeopardy, it is not surprising that the state courts also generally fixed the attachment of jeopardy at the swearing of the jury.11 Because the ed. 1876); 1 J. Bishop, Commentaries on the Criminal Law § 1016 (5th ed. 1872); T. Cooley, Constitutional Limitations 325-327 (2d ed. 1871). See generally ALT, Administration of the Criminal Law, Commentary to §6, pp. 61-72 (1935). The leading English criminal law treatise was to the contrary. See 1 J. Chitty, Criminal Law 451-463, 480 (J. Perkins ed. 1847). 11 See, e. g., State v. M’Kee, supra, at 655; Morgan n. State, supra, at 216; State v. Redman, 17 Iowa 329, 333 (1864); People v. Webb, supra, at 478; Nolan v. State, supra, at 523; State v. Davis, 80 N. C. 384 (1879); Mitchell v. State, supra, at 393; State v. Ward, supra, at 38, 2 S. W. 191; People n. Gardner, supra, at 311, 29 N. W., at 20; State v. Paterno, supra, at 515, 9 So. 442; McDonald n. State, supra, at 653, 48 N. W., at 864; State n. Sommers, supra, at 91, 61 N. W. 907; Dulin v. Lillard, supra, at 722, 20 S. E., at 822; accord, Bishop, supra, n. 10; Cooley, supra, n. 10. CRIST v. BRETZ 47 28 Powell, J., dissenting state courts do not appear to have been aware that they were adapting a separate rule to a different area of individual rights, they perceived no need to examine all the trappings of the rule in light of the new uses to which it was being put.12 It was after more than a century of development in state courts that the “defendant’s valued right to have his trial completed by a particular tribunal” appeared in the decisions of this Court for the first time, also without analysis, as an element of the Double Jeopardy Clause. Wade v. Hunter, 336 U. S. 684, 689 (1949). The policies underlying this “valued right” were not spelled out in JFWe,13 but the rationale expressed in Green v. United States, supra, at 187-188—a case not involving midtrial discharge of the jury—appears to echo the state courts of a century earlier: “. . . [T]he State with all its resources and power should not be allowed to make repeated attempts to convict an individual for an alleged offense, thereby subjecting him to embarrassment, expense and ordeal and compelling him to live in a continuing state of anxiety and insecurity, as well as enhancing the possibility that even though innocent he may be found guilty.” Although neither Wade nor Green confronted the question of when jeopardy attached, the Green Court declared that “ [t]his Court, as well as most others, has taken the position that a defendant is placed in jeopardy once he is put to trial before a jury so that if the jury is discharged without his consent he cannot be tried again.” 355 U. S., at 188. Having accepted almost without articulated thought the doctrine that the Double Jeopardy Clause protects against needless discharge of the jury, this Court proceeded to adopt 12 But see United States v. Bigelow, supra. 13 Similarly, the Court today does not explore the reasons supporting valuation of this particular right, merely announcing that it is “valued.” Ante, at 38. 48 OCTOBER TERM, 1977 Powell, J., dissenting 437U.S. with a similar lack of reason or analysis the implementing rule that jeopardy attaches when the jury is sworn. In Downum v. United States, 372 U. S. 734 (1963), the trial court declared a mistrial after the jury had been sworn but before any witnesses had been called. Finding an absence of “imperious necessity,” id., at 736, the Court held that the Fifth Amendment barred reprosecution. The Downum opinion contains no discussion of the point of jeopardy’s attachment or of the policies underlying the selection of the swearing of the jury as the determinative moment.14 Nevertheless, the swearing of the jury has been accepted since Downum as the constitutional line of demarcation for the attachment of jeopardy, see, e. g., Illinois v. Somerville, 410 U. S. 458, 466 (1973); United States v. Sisson, 399 U. S. 267, 305 (1970), even though no case before this Court has presented a contest over that issue.15 This Court, following the lead of the state courts, simply enlisted the doctrine concerning needless discharge of juries in the service of double jeopardy principles, largely without anal 14 The Government in Downum conceded that jeopardy attaches at the time the jury is sworn. Brief for United States, 0. T. 1962, No. 489, p. 31. In support of this concession, the Government cited Lovato v. New Mexico, 242 U. S. 199 (1916), apparently believing that Lovato had involved discharge of the jury immediately after swearing. In that case, however, the witnesses for both sides had been sworn, so that it actually furnished no support for the concession. Since the parties did not dispute the point of jeopardy’s attachment, the Court did not discuss the matter. Because the rule of attachment was not put in issue and not discussed in Downum, we owe this sub silentio determination less deference than a holding arrived at after full argument and consideration, see Monell n. New York City Dept, of Social Services, 436 U. S. 658, 709-710, n. 6 (1978) (Powell, J., concurring), particularly in a constitutional case. 15 In Serfass v. United States, 420 U. S. 377 (1975), the petitioner sought to have the point of attachment moved forward to the filing of pretrial motions. The Court’s refusal to fix the attachment of jeopardy at that stage of the litigation did not require any consideration of the policies underlying the rule assumed in Downum and reaffirmed today. CRIST v. BRETZ 49 28 Powell, J., dissenting ysis and apparently with little awareness of history. In view, however, of the consistency with which federal courts have assumed without question that the swearing of the jury triggers jeopardy, I would accept this as the established supervisory rule within the federal system. But the acceptance of a supervisory rule, primarily on grounds of long tenure and convenience, is no justification for elevating it to constitutional doctrine. We should be hesitant to constitutionalize a rule that derives no support from the Framers’ understanding of the English practice from which the Double Jeopardy Clause was derived, and which is supported by no doctrinal reasoning that reaches constitutional dimension. Restraint is doubly indicated with respect to this rule since it is applied only in jury trials. Where a criminal case is tried to the court, jeopardy does not attach until “the court begins to hear evidence.” Serfass v. United States, 420 U. S. 377, 388 (1975). No compelling reason has been suggested today, or in earlier decisions of this Court, why the time when jeopardy attaches should be different depending upon whether the defendant’s “valued right” is asserted in a jury trial rather than a bench trial. I turn next to an examination of the jury trial rule in light of the double jeopardy policies it is now belatedly thought to advance. II Three aspects of criminal process ordinarily precede the initial introduction of evidence in a jury trial: motions, jury selection, and opening statements. Defendants are vitally interested in each, yet it is far from clear that any should trigger the attachment of jeopardy. Defendants may, and sometimes must, see, e. g., Fed. Rule Crim. Proc. 12, move for various rulings on the indictment and the admissibility of evidence before trial. These motions, in practical terms, may decide the defendant’s case. They 50 OCTOBER TERM, 1977 437 U.S. Powell, J., dissenting sometimes may require a devotion of time, energies, and resources exceeding that necessary for the trial itself. Yet it has never been held that jeopardy attaches as of the making or deciding of pretrial motions. See Serfass v. United States, supra. Appellee does not contend otherwise. It is clear, then, that the central concern of the Double Jeopardy Clause cannot be regarded solely as protecting against repeated expenditures of the defendant’s efforts and resources. Opening statements may be made in both bench and jury trials.16 In either type of trial, statements by counsel or questions by the court may prompt the prosecutor to abort— by dismissing the indictment or otherwise—the proceedings with the view to reindicting the defendant and commencing anew. The prosecutor also may simply request a continuance to gain time to meet some unexpected defense stratagem, although such a motion rarely would prevail. In any event, delay or postponement occasioned during or as a result of the opening-statement phase of a trial would be equally adverse to the defendant without regard to whether he were being tried by the court or a jury. The Due Process Clause would protect such a defendant in either case against prosecutorial abuse. Thus, with respect to the opening-statement phase of a criminal trial, there appears to be no difference of substance between jury and bench trials in terms of serving double jeopardy policies. The situation does differ in some respects where a jury is selected, and the defendant—by voir dire and challenges— participates in the selection of the factfinder. It is not unusual for this process to entail a major effort and extend over a protracted period. But, as in the case of pretrial 16 Apparently, defense counsel often choose to reserve their opening statements until the close of the prosecution’s case. Tr. of Oral Arg. 10, 15-17; Brief on Reargument for United States as Amicus Curiae 23 n. 25. Where this course is followed, there will be no early disclosure of defense strategy. CRIST v. BRETZ 51 28 Powell, J., dissenting motions, expenditure of effort alone is not sufficient to trigger the attachment of jeopardy.17 The federal rule of attachment in jury trials offers no basis for a double jeopardy claim if the prosecutor—dissatisfied by the jury selection process—is successful in dismissing the prosecution before the last juror is seated, or indeed before the whole panel is sworn. A defendant’s protection against denial or abuse of his rights in this respect lies in the Due Process Clause. Moreover, the Double Jeopardy Clause cannot be viewed as a guarantee of the defendant’s claim to a factfinder perceived as favorably inclined toward his cause. That interest does not bar pretrial reassignment of his case from one judge to another, even though he may have waived jury trial on the belief that the original judge viewed his case favorably. Thus, the Double Jeopardy Clause interest in having his “trial completed by a particular tribunal,” Wade v. Hunter, 336 U. S., at 689, must refer to some interest other than retaining a factfinder thought to be disposed favorably toward defendant. The one event that can distinguish one factfinder from another in the eyes of the law in general, and the Double Jeopardy Clause in particular, is the beginning of the factfinder’s work. As the Court stated in Green, “a defendant is placed in jeopardy once he is put to trial before” a factfinder. 355 U. S., at 188 (emphasis added). When the court or jury has undertaken its constitutional duty—the hearing of evidence— the trial quite clearly is under way, and the prosecution’s case has begun to unfold before the trier of fact. Cf. United States v. Scott, post, at 101. As testimony commences, the evidence of the alleged criminal conduct is presented to the 17 At least one commentator has* proposed fixing jeopardy’s attachment at the start of voir dire, in order to protect the defendant’s interest in each juror, as selected. Schulhofer, Jeopardy and Mistrials, 125 U. Pa. L. Rev. 449, 513 (1977). This proposal, however, has no historical foundation nor any clear grounding in the concerns of the Double Jeopardy Clause. 52 OCTOBER TERM, 1977 Powell, J., dissenting 437 U. S. factfinder and becomes a matter of public record. The defendant’s public embarrassment and anxiety begin. From this point on, retrial will mean repeating painful and embarrassing testimony, together with the possibility that the earlier “trial run” will strengthen the prosecution’s case. At a retrial, for example, prosecution witnesses may be better prepared for the rigors of cross-examination. Thus, the defendant has a strong interest in taking his case to the first jury, once witnesses testify. Carsey v. United States, 129 U. S. App. D. C. 205, 208-209, 392 F. 2d 810, 813-814 (1967) (Leventhal, J., concurring). The rationale of the Double Jeopardy Clause is implicated once this threshold is crossed, but not before. That this is the crucial time for Double Jeopardy Clause purposes is evident from the attachment rule in bench trials. Once the judge has embarked upon his factfinding mission, the defendant is justified in concluding that his ordeal has begun; he is in the hands of his judge and may expect the matter to proceed to a finish. This same principle should apply in jury trials. Thus, Montana’s rule fixing the attachment of jeopardy at the swearing of the first witness is consonant with the central concerns of the Double Jeopardy Clause. It furnishes a clear line of demarcation for the attachment of jeopardy, and it places that line in advance of the point at which real jeopardy—in Fifth Amendment terms—can be said to begin. Ill Even if I were to conclude that the Fifth Amendment— merely by virtue of long, unreasoned acceptance—required attachment of jeopardy at the swearing of the jury, I would not hold that the Fourteenth Amendment necessarily imposes that requirement upon the States. This issue would turn on the answer to the question whether jeopardy’s attachment at that point is fundamental to the guarantees of the Double Jeopardy Clause. Apodaca v. Oregon, 406 U. S. 404, 373 (1972) (Powell, J., concurring in judgment); Ludwig v. CRIST v. BRETZ 53 28 Powell, J., dissenting Massachusetts, 427 U. S. 618, 632 (1976) (Powell, J., concurring). As the previous discussion makes clear, the jury trial rule accorded constitutional status by the Court today implicates no rights that have been identified as fundamental in a constitutional sense. There is no basis for incorporating it “jot-for-jot” into the Fourteenth Amendment. See Duncan v. Louisiana, 391 U. S. 145,181 (1968) (Harlan, J., dissenting). IV Aside from paying cryptic homage to the hitherto unexplained “valued right” to a particular jury, the Court does not even attempt to justify its holding that the Fifth Amendment mandates the rule of attachment that it adopts. It identifies no policy of the Double Jeopardy Clause, and no interests of a fair system of criminal justice, that elevate this “right” to constitutional status. The Court’s rule is not even a “linedrawing” that finds support in logic or significant convenience. I perceive no reason for this Court to impose what, in effect, is no more than a supervisory rule of practice upon the courts of every State in the Union. 54 OCTOBER TERM, 1977 Syllabus 437 U. S. SANABRIA v. UNITED STATES CERTIORARI TO THE UNITED STATES COURT OF APPEALS FOR THE FIRST CIRCUIT No. 76-1040. Argued November 8, 1977—Decided June 14, 1978 Title 18 U. S. C. § 1955 (1976 ed.) makes it a federal offense for five or more persons to conduct an “illegal gambling business” in violation of the law of the place where the business is located. Petitioner, along with several others, was indicted for violating § 1955 in a single count charging that the defendants’ gambling business involved numbers betting and betting on horse races in violation of a specified Massachusetts statute. The Government’s evidence at trial in the District Court showed that the defendants had been engaged in both horse betting and numbers betting. At the close of the Government’s case defense counsel argued that the Government had failed to prove a violation of the Massachusetts statute because that statute did not prohibit numbers betting but only horse betting. After the defendants had rested, the trial judge granted their motion to exclude all evidence of numbers betting and then granted a motion to acquit petitioner because of lack of evidence of his connection with the horse-betting business. The case against the remaining defendants went to the jury, and they were all convicted. The Government appealed under 18 U. S. C. § 3731 (1976 ed.) from the order excluding the numbers-betting evidence and from the judgment acquitting petitioner, and sought a new trial on the portion of the indictment relating to numbers betting. The Court of Appeals held that it had jurisdiction of the appeal, taking the view that, although § 3731, by its terms, authorizes the Government to appeal only from orders “dismissing an indictment ... as to any one or more counts,” the word “counts” refers to any discrete basis for imposing criminal liability, that since the horsebetting and numbers allegations were discrete bases for liability duplici-tously joined in a single count, the District Court’s action constituted a “dismissal” of the numbers “charge” and an acquittal for insufficient evidence on the horse-betting charge, and that therefore § 3731 authorized an appeal from the “dismissal” of the numbers charge. The court went on to hold that the Double Jeopardy Clause of the Fifth Amendment did not bar a retrial, because petitioner had voluntarily terminated the proceedings on the numbers portion of the count by moving, in effect, to dismiss it. The court vacated the judgment of acquittal, and remanded for a new trial on the numbers charge. Held: SANABRIA v. UNITED STATES 55 54 Syllabus 1. A retrial on, the numbers theory of liability is barred by the Double Jeopardy Clause. Pp. 63-74. (a) The Court of Appeals erroneously characterized the District Court’s action as a “dismissal” of the numbers theory. There was only one count charged, the District Court did not order language in the indictment stricken, and the indictment was not amended, but the judgment of acquittal was entered on the entire count and found petitioner not guilty of violating § 1955 without specifying that it did so only with respect to one theory of liability. Pp. 65-68. (b) To the extent that the District Court found the indictment’s description of the offense too narrow to warrant admission of certain evidence, the court’s ruling was an erroneous evidentiary ruling, which led to an acquittal for insufficient evidence, and that judgment of acquittal, however erroneous, bars further prosecution on any aspect of the count and hence bars appellate review of the trial court’s error. Pp. 68-69. (c) Even if it could be said that the District Court “dismissed” the numbers allegation, a retrial on that theory would subject petitioner to a second trial on the “same offense” of which he was acquitted. Under § 1955 participation in a single gambling business is but a single offense, no matter how many state statutes the enterprise violated, and with regard to this single gambling business petitioner was acquitted. The Government having charged only a single gambling business, the discrete violations of state law that that business may have committed are not severable in order to avoid the Double Jeopardy Clause’s bar of retrials for the “same offense.” Pp. 69-74. 2. Once the defendant has been acquitted, no matter how “egregiously erroneous” the legal rulings leading to the judgment of acquittal might be, there is no exception to the constitutional rule forbidding successive trials for the same offense. Fong Foo v. United States, 369 U. S. 141. Thus here, while the numbers evidence was erroneously excluded, the judgment of acquittal produced thereby is final and unreviewable. Lee n. United States, 432 U. S. 23; Jefjers v. United States, 432 U. S. 137, distinguished. Pp. 75-78. 548 F. 2d 1, reversed. Marshall, J., delivered the opinion of the Court, in which Burger, C. J., and Brennan, Stewart, and Powell, JJ., joined; in all but n. 23 of which Stevens, J., joined; and in Parts I, II-A, and III of which White, J., joined. Stevens, J., filed a concurring opinion, post, p. 78. Blackmun, J., filed a dissenting opinion, in which Rehnquist, J., joined, post, p. 80. 56 OCTOBER TERM, 1977 Opinion of the Court 437 U. 8. Francis J. DiMento argued the cause and filed briefs for petitioner. Frank H. Easterbrook argued the cause for the United States pro hac vice. With him on the brief were Solicitor General McCree, Assistant Attorney General Civiletti, and Sidney M. Glazer. Mr. Justice Marshall delivered the opinion of the Court.* The issue presented is whether the United States may appeal in a criminal case from a midtrial ruling resulting in the exclusion of certain evidence and from a subsequently entered judgment of acquittal. Resolution of this issue depends on the application of the Double Jeopardy Clause of the Fifth Amendment to the somewhat unusual facts of this case. I Petitioner was indicted, along with several others, for violating 18 U. S. C. § 1955 (1976 ed.), which makes it a federal offense to conduct, finance, manage, supervise, direct, or own all or part of an “illegal gambling business.” § 1955 (a). Such a business is defined as one that is conducted by five or more persons in violation of the law of the place where the business is located and that operates for at least 30 days or earns at least $2,000 in any one day. § 1955 (b)(1).1 The *Mr. Justice White joins Parts I, II-A, and III of this opinion. 1 Title 18 U. 8. C. § 1955 (1976 ed.) provides in relevant part: “Prohibition of illegal gambling businesses. “(a) Whoever conducts, finances, manages, supervises, directs, or owns all or part of an illegal gambling business shall be fined not more than $20,000 or imprisoned not more than five years, or both. “(b) As used in this section— “(1) ‘illegal gambling business’ means a gambling business which— “ (i) is a violation of the law of a State or political subdivision in which it is conducted; “ (ii) involves five or more persons who conduct, finance, manage, supervise, direct, or own all or part of such business; and SANABRIA v. UNITED STATES 57 54 Opinion of the Court single-count indictment here charged in relevant part that the defendants’ gambling business involved “accepting, recording and registering bets and wagers on a parimutual [sic] number pool and on the result of a trial and contest of skill, speed, and endurance of beast,” and that the business “was a violation of the laws of the Commonwealth of Massachusetts, to wit, M. G. L. A. Chapter 271, Section 17.” 2 The Government’s evidence at trial showed the defendants to have been engaged primarily in horse betting and numbers betting. At the close of the Government’s case, petitioner’s counsel, who represented 8 of the 11 defendants, moved for a judgment of acquittal as to all of his clients. Joined by counsel for other defendants, he argued, inter alia, that the “(iii) has been or remains in substantially continuous operation for a period in excess of thirty days or has a gross revenue of $2,000 in any single day. “(2) 'gambling’ includes but is not limited to pool-selling, bookmaking, maintaining slot machines, roulette wheels or dice tables, and conducting lotteries, policy, bolita or numbers games, or selling chances therein. "(3) 'State’ means any State of the United States, the District of Columbia, the Commonwealth of Puerto Rico, and any territory or possession of the United States.” 2 The indictment alleged in full: "From on or about June 1, 1971 and continuing thereafter up to and including November 13, 1971 at Revere, Massachusetts within the District of Massachusetts, [the defendants] did unlawfully, knowingly, and wilfully conduct, finance, manage, supervise, direct and own all and a part of an illegal gambling business, to wit, accepting, recording and registering bets and wagers on a parimutual [sic] number pool and on the result of a trial and contest of skill, speed, and endurance of beast, said illegal gambling business; (i) was a violation of the laws of the Commonwealth of Massachusetts, to wit, M. G. L. A. Chapter 271, Section 17, in which place said gambling business was being conducted; (ii) involved five and more persons who conducted, financed, managed,'supervised, directed and owned all and a part of said business; (iii) had been in substantially continuous operation for a period in excess of thirty days and had a gross revenue of two thousand dollars ($2,000) in any single day; all in violation of Title 18, United States Code, Sections 1955 and 2.” 58 OCTOBER TERM, 1977 Opinion of the Court 437 U. S. Government had failed to prove that there was a violation of the state statutory section as alleged in the indictment, since Mass. Gen. Laws Ann., ch. 271, § 17 (West 1970), as construed by the state courts, did not prohibit numbers betting but applied only to betting on “games of competition” such as horse races. The Government responded that “violation of the State law is a jurisdictional element of [the federal] statute” and that “not every [defendant] must be found to be violating this State law.” The District Court accepted the Government’s theory and denied the defendants’ motion, stating that “a defendant to be convicted must [only] be found to have joined in [the illegal] enterprise in some way.” Petitioner’s counsel then sought clarification of whether “the numbers pool allegation [was] still in the case.” The court indicated that it was, because counsel had not presented any state-court authority for the proposition that § 17 did not include numbers betting. The court also expressed the view, however, that if petitioner’s counsel were correct, “we would have to exclude ... all of the evidence that has to do with bets o[n] numbers.” The Government demurred, arguing that exclusion of the numbers evidence would “not necessarily follow” from acceptance of petitioner’s theory.3 Taking his lead from the court, petitioner’s counsel next moved “to strike or limit the evidence.” The motion was denied. After the defendants had rested, the trial judge announced that he was reversing his earlier ruling on the motion to exclude evidence, because he had discovered a Massachusetts 3 When the District Judge asked why exclusion of the numbers evidence “would not necessarily follow,” the Government responded: “Because the Defendants have been charged with operating a gambling business, which is in violation of State law. Now, there’s no question that the horse race aspect of it is in violation of State law. There are other aspects to the bets as well, but the violation of State law is merely a jurisdictional element which must be satisfied prior to the initiation of Federal prosecution.” SANABRIA v. UNITED STATES 59 54 Opinion of the Court case holding that numbers betting was not prohibited by § 17, but only by § 7 of ch. 271.4 The court then struck all evidence of numbers betting, apparently because it believed such action to be required by the indictment’s failure to set forth the proper section.5 At this point counsel moved for a judgment of acquittal as to petitioner alone, arguing that there was no evidence of his connection with horse-betting activities. The Government did not disagree that the evidence was insufficient to show petitioner’s involvement with a horse-betting operation, but repeated its earlier argument relating to the “jurisdictional” nature of the state-law violation. The court rejected this contention, stating that the offense had “to be established in the terms that you [the Government] charged it, which was as a violation of § 17” and that petitioner had to be “connected with this operation, and by that I mean a horse operation.” The court concluded: “I don’t think you’ve done it.” It then granted petitioner’s motion for a judgment of acquittal6 and entered an order embodying this ruling later that day.7 The next day the Government moved the court to reconsider both “its ruling . . . striking . . . evidence concerning the operation of an illegal . . . numbers pool” and “its decision granting defendant Thomas Sanabria’s motion for judgement 4 Commonwealth v. Boyle, 346 Mass. 1, 189 N. E. 2d 844 (1963). 5 The Government did not at this time argue, as it had previously, see n. 3, supra, that the numbers evidence was relevant to show "other aspects” of the bets even if it could not be used to prove that the business violated state law. Instead, it urged that the numbers evidence was admissible as proof of "similar acts.” 6 Petitioner has consistently maintained that he properly moved to exclude the numbers evidence as irrelevant to the indictment’s characterization of the gambling business; that the District Court properly granted the evidentiary motion, see Tr. of Oral Arg. 12; and that the District Court properly granted petitioner’s motion for a judgment of acquittal after excluding the numbers evidence on the grounds of insufficient evidence. 7 The text of the judgment is quoted injra, at 67. 60 OCTOBER TERM, 1977 Opinion of the Court 437U.S. [sic] of acquittal.” 8 Prompted by the Government’s arguments in support of reconsideration, the court asked defense counsel why he had not raised the objection to the indictment’s citation of § 17 earlier and what prejudice resulted to petitioner from the failure to cite the proper section. Counsel responded that the objection had not “ripened” until, at the end of the Government’s case, the court was asked to take judicial notice of § 17, and that he need not and did not allege actual prejudice. The court denied the motions to reconsider, but indicated that, had it granted the motion to restore the numbers evidence, it also would have vacated the judgment of acquittal.9 The case against the remaining 10 defendants went to the jury on a theory that the gambling business was engaged in horse betting; all were convicted. The Government filed a timely appeal “from [the] decision 8 In support of these motions, the Government argued that the failure to cite Mass. Gen. Laws Ann., ch. 271, § 7 (West 1970), in the indictment was a technical defect causing no prejudice to the defendants and subject to correction during trial under Fed. Rule Crim. Proc. 7. See n. 11, infra. If the numbers evidence were restored to the case, the Government argued, vacating the judgment of acquittal would be proper, since it had resulted solely from the erroneous exclusion of evidence and since no new trial would be necessary in view of the fact that the jury had not been discharged. 9 The trial court explained its reasoning as follows: “If the other motion had been granted, I think, probably, the Motion to Reconsider the Acquittal of Sanabria would be allowed under these new decisions: Wilson, which is in 420 US 332; Jenkins, 420 US 358; and Serfass at 420 US 377, all decided the last term. All of those seem to say if a judgment of acquittal or judgment of dismissal is entered on legal grounds as opposed to containing or importing a finding of fact and the reversal of that decision would not require a new trial, then'it may be reversed. “In Fong Foo [v. United States, 369 U. S. 141 (1962)] the jury had been discharged, and it would have been necessary to draw a new jury and start a new trial, and in Jenkins they specifically distinguished Fong Foo from the Wilson-Jenkins-Serfass group . . , SANABRIA v. UNITED STATES 61 54 Opinion of the Court and order . . . excluding evidence and entering a judgment of acquittal... and ... denying the Motion for Reconsideration.” Conceding that there could be no review of the District Court’s ruling that there was insufficient evidence of petitioner’s involvement with horse betting, the Government sought a new trial on the portion of the indictment relating to numbers betting. The Court of Appeals for the First Circuit held first that it had jurisdiction of the appeal. Although the jurisdictional statute, 18 U. S. C. § 3731 (1976 ed.), by its terms authorizes the Government to appeal only from orders “dismissing an indictment ... as to any one or more counts.”10 the word “count” was “interpret[ed] ... to refer to any discrete basis for the imposition of criminal liability.” 548 F. 2d 1, 5 (1976). Viewing the horse-betting and numbers allegations as “discrete bas[es] of criminal liability” duplicitously joined in a single count, the court characterized the District Court’s action as a “dismissal” of the numbers “charge” and an acquittal for insufficient evidence on the horse-betting charge. Id., at 4-5, and n. 4. It concluded that § 3731 authorized an appeal from the “dismissal” of the numbers charge, “if the double jeopardy clause does not bar a future prosecution on this charge.” 548 F. 2d, at 5. Consistent with its above analysis, the court found that petitioner had voluntarily terminated the proceedings on the numbers portion of the count by moving, in effect, to dismiss it. Since the “dismissal” imported no ruling on petitioner’s 10 Another provision of § 3731 authorizes the Government to appeal from orders “suppressing or excluding evidence . . . not made after the defendant has been put in jeopardy and before the verdict or finding on [the] indictment.” The Government does not contend that the ruling excluding numbers evidence was ‘appealable under this provision. By its plain terms, moreover, this second paragraph of § 3731 does not authorize this appeal, since the ruling excluding evidence occurred after the defendant had been put in jeopardy and before verdict. Cf. United States v. Morrison, 429 U. S. 1 (1976). 62 OCTOBER TERM, 1977 Opinion of the Court 437 IT. S. “criminal liability as such,” and since petitioner’s motion was not attributable to “prosecutorial or judicial overreaching,” the court applied the rule permitting retrials after a prosecution is terminated by a defendant’s request for a mistrial. Id., at 7-8, citing United States v. Dinitz, 424 U. S. 600 (1976). There being no double jeopardy bar to a new trial, the court went on to resolve the merits of the appeal in the Government’s favor. It held, based on an intervening First Circuit decision,11 that the District Court had erred in “dismissing” the numbers theory. Accordingly, the judgment of acquittal was “vacated” and the case “remanded so that the government may try defendant on that portion of the indictment that charges a violation of § 1955 based upon numbering [sic] activities.” 548 F. 2d, at 8. We granted certiorari, 433 U. S. 907 (1977),12 limiting our review to the related issues of appealability and double jeopardy.13 We now reverse. 11 United States v. Morrison, 531 F. 2d 1089, 1094, cert, denied, 429 U. S. 837 (1976). Morrison held a failure to cite Mass. Gen. Laws Ann., ch. 271, § 7 (West 1970), in a similarly worded indictment to be harmless error. Based on Morrison, the court below concluded that the indictment was sufficient to give “notice that numbers activity was a basis upon which the government sought to establish criminal liability under § 1955.” 548 F. 2d, at 4. 12 The petition for certiorari was filed one day out of time. The time requirement of this Court’s Rule 22 (2) is not jurisdictional, Schacht n. United States, 398 IT. S. 58, 63-65 (1970), and petitioner has filed a motion, supported by affidavits, seeking waiver of this requirement. We now grant petitioner’s motion. 13 The petition for certiorari presented four questions for review, the first three relating to whether the Government’s appeal was authorized by statute and not barred by the Double Jeopardy Clause. The fourth question sought review of the Court of Appeals’ ruling that the indictment gave sufficient notice of the Government’s intent to rely on evidence of numbers betting. Our order limited the grant of certiorari to the first three questions. 433 U. S. 907 (1977). Accordingly, we must assume that the District Court erred in ruling that the indictment did not SANABRIA v. UNITED STATES 63 54 Opinion of the Court II In United States v. Wilson, 420 U. S. 332 (1975), we found that the primary purpose of the Double Jeopardy Clause was to prevent successive trials, and not Government appeals per se. Thus we held that, where an indictment is dismissed after a guilty verdict is rendered, the Double Jeopardy Clause did not bar an appeal since the verdict could simply be reinstated without a new trial if the Government were successful.14 That a new trial will follow upon a Government appeal does not necessarily forbid it, however, because in limited circumstances a second trial on the same offense is constitutionally permissible.15 Appealability in this case therefore turns on whether the new trial ordered by the court below would violate the command of the Fifth Amendment that no “person [shall] be subject for the same offence to be twice put in jeopardy of life or limb.” 16 encompass the numbers allegation because of its failure to cite Mass. Gen. Laws Ann., ch. 271, § 7 (West 1970). 14 United States v. Jenkins, 420 U. S. 358 (1975), by contrast, held that appeal of an order dismissing an indictment after jeopardy had attached, but before verdict, was barred because a successful appeal would require “further proceedings . . . devoted to the resolution of factual issues going to the elements of the offense charged.” Id., at 370. See Lee v. United States, 432 U. S. 23, 29-30 (1977). 15 A new trial is permitted, e. g., where the defendant successfully appeals his conviction, United States v. Ball, 163 U. S. 662, 672 (1896); where a mistrial is declared for a “manifest necessity,” Wade v. Hunter, 336 U. S. 684 (1949); where the defendant requests a mistrial in the absence of prosecutorial or judicial overreaching, United States v. Dinitz, 424 U. S. 600 (1976); or where an indictment is dismissed at the defendant’s request in circumstances functionally equivalent to a mistrial, Lee v. United States, supra. See also Jeffers v. United States, 432 U. S. 137 (1977). 16 We have on several occasions observed that the jurisdictional statute authorizing Government appeals, 18 U. S. C. §3731 (1976 ed.), was “ 'intended to remove all statutory barriers’ ” to appeals from orders terminating prosecutions. United States v. Martin Linen Supply Co., 430 U. S. 564, 568 (1977), quoting United States n. Wilson, 420 U. S. 332, 337 (1975). We therefore turn immediately to the constitutional issues. 64 OCTOBER TERM, 1977 Opinion of the Court 437U.S. In deciding whether a second trial is permissible here, we must immediately confront the fact that petitioner was acquitted on the indictment. That “‘[a] verdict of acquittal... [may] not be reviewed ... without putting [the defendant] twice in jeopardy, and thereby violating the Constitution,’ ” has recently been described as “the most fundamental rule in the history of double jeopardy jurisprudence.” United States v. Martin Linen Supply Co., 430 U. S. 564, 571 (1977), quoting United States v. Ball, 163 U. S. 662, 671 (1896). The fundamental nature of this rule is manifested by its explicit extension to situations where an acquittal is “based upon an egregiously erroneous foundation.” Fong Foo v. United States, 369 U. S. 141,143 (1962); see Green v. United States, 355 U. S. 184, 188 (1957). In Fong Foo the Court of Appeals held that the District Court had erred in various rulings and lacked power to direct a verdict of acquittal before the Government rested its case.17 We accepted the Court of Appeals’ holding that the District Court had erred, but nevertheless found that the Double Jeopardy Clause was “violated when the Court of Appeals set aside the judgment of acquittal and directed that petitioners be tried again for the same offense.” 369 U. S., at 143. Thus when a defendant has been acquitted at trial he may not be retried on the same offense, even if the legal rulings underlying the acquittal were erroneous. The Government does not take issue with these basic principles. Indeed, it concedes that the acquittal for insufficient evidence on what it refers to as the horse-betting theory of liability is unreviewable and bars a second trial on that charge.18 The disputed question, however, is whether a retrial 17 In re United States, 286 F. 2d 556 (CAI 1961). 18 It is without constitutional significance that the court entered a judgment of acquittal rather than directing the jury to bring in a verdict of acquittal or giving it erroneous instructions that resulted in an acquittal. United States v. Martin Linen Supply Co., supra, at 567 n. 5, 573; United States v. Sisson, 399 U. S. 267, 290 (1970).- SANABRIA v. UNITED STATES 65 54 Opinion of the Court on the numbers theory of liability would be on the “same offense” as that on which petitioner has been acquitted. The Government contends, in accordance with the reasoning of the Court of Appeals, that the numbers theory was dismissed from the count before the judgment of- acquittal was entered and therefore that petitioner was not acquitted of the numbers theory. Petitioner responds that the District Court did not “dismiss” anything but rather struck evidence and acquitted petitioner on the entire count; further, assuming arguendo that there was a “dismissal” of the numbers theory, he urges that a retrial on this theory would nevertheless be barred as a second trial on the same statutory offense. We first consider whether the Court of Appeals correctly characterized the District Court’s action as a “dismissal” of the numbers theory. A In the Government’s view, the numbers theory was “dismissed” from the case as effectively as if the Government had actually charged the crime in two counts and the District Court had dismissed the numbers count. The first difficulty this argument encounters is that the Government did not in fact charge this offense in two counts. Legal consequences ordinarily flow from what has actually happened, not from what a party might have done from the vantage of hindsight. See Central Tablet Mfg. Co. v. United States, 417 U. S. 673, 690 (1974).19 The precise manner in which an indictment 19 The difficulty in allowing a defendant’s rights to turn on what the Government might have done is illustrated by considering that, had the Government alleged each “theory of liability” in a separate count, the indictment would have been subject to objection on grounds of multiplicity, the charging of a single offense in separate counts. See n. 20, infra. The Government might then have been forced to elect on which count it would proceed against petitioner, United States v. Universal C. I. T. Credit Corp., 344 U. S. 218 (1952), and probably would have chosen to proceed on the numbers theory as to which its evidence was apparently stronger. In that 66 OCTOBER TERM, 1977 Opinion of the Court 437U.S. is drawn cannot be ignored, because an important function of the indictment is to ensure that, “in case any other proceedings are taken against [the defendant] for a similar offence, . . . the record [will] sho[w] with accuracy to what extent he may plead a former acquittal or conviction.” Cochran v. United States, 157 U. S. 286, 290 (1895), quoted with approval in Russell v. United States, 369 U. S. 749, 764 (1962); Hagner v. United States, 285 U. S. 427, 431 (1932).20 With regard to the one count that was in fact charged, as to which petitioner has been at least formally acquitted, we are not persuaded that it is correct to characterize the trial court’s action as a “dismissal” of a discrete portion of the count. While form is not to be exalted over substance in determining the double jeopardy consequences of a ruling terminating a prosecution, Serfass v. United States, 420 U. S. 377, 392-393 (1975); United States v. Jorn, 400 U. S. 470, 478 n. 7 (1971); United States v. Goldman, 277 U. S. 229, 236 (1928), neither is it appropriate entirely to ignore the form of order entered by the trial court, see United States v. Barber, 219 U. S. 72, 78 (1911). Here the District Court issued only two orders, one excluding certain evidence and the other entering a judgment of acquittal on the single count charged. No language in the indictment was ordered to be stricken, compare United States v. Alberti, 568 F. 2d 617, 621 (CA2 1977), nor was the indictment amended. The judgment of acquittal was entered on the entire count and found petitioner not guilty of event, however, petitioner could not have been acquitted of the horsebetting count, and the instant problem would not have arisen. 20 The Court of Appeals erred in its apparent view that the Government should have drawn the indictment in two counts because the single count was duplicitous. 548 F. 2d, at 5 n. 4. Only a single gambling business was alleged, and hence only a single offense. See infra, at 70-71. A single offense should normally be charged in one count rather than several, even if different means of committing the offense are alleged. See Fed. Rule Crim. Proc. 7 (c)(1); Advisory Committee’s Notes on Fed. Rule Crim. Proc. 7, 18 U. S. C. App., p. 1413 (1976 ed.); n. 19, supra. SANABRIA v. UNITED STATES 67 54 Opinion of the Court the crime of violating 18 U. S. C. § 1955 (1976 ed.), without specifying that it did so only with respect to one theory of liability: “The defendant having been set to the bar to be tried for the offense of unlawfully engaging in an illegal gambling business, in violation of Title 18, United States Code, Sections 1955 and 2, and the Court having allowed defendant’s motion for judgment of acquittal at the close of government’s evidence, “It is hereby Ordered that the defendant Thomas Sanabria be, and he hereby is, acquitted of the offense charged, and it is further Ordered that the defendant Thomas Sanabria is hereby discharged to go without day.” The Government itself characterized the District Court’s ruling from which it sought to appeal as “a decision and order . . . excluding evidence and entering a judgment of acquittal.” Notice of Appeal.21 Similar language appears in 21 The Court of Appeals might have been warranted in dismissing the appeal for failure of the notice to specify the only arguably appealable ruling rendered below. The court believed that “ [t J he critical ruling by the district court was that the indictment failed to charge a violation of § 1955 on a numbers theory.” 548 F. 2d, at 5 n. 5. But this “critical ruling,” which the court below concluded was a “dismissal,” is not set forth in the notice of appeal. Since the Government is not authorized to appeal from all adverse rulings in criminal cases, it is especially important that it specify precisely what it claims to have been the appealable ruling. The Court of Appeals, however, must have concluded that the notice was sufficient to bring up for review the legal ruling preceding the order excluding evidence. A mistake in designating the judgment appealed from is not always fatal, so long as the intent to appeal from a specific ruling can fairly be inferred by probing the notice and the other party was not misled or prejudiced. Daily Mirror, Inc. v. New York News, Inc., 533 F. 2d 53 (CA2 1976) (per curiam); Jones v. Nelson, 484 F. 2d 1165 (CAIO 1973). The Government’s “Designation of Issue [szc] on Appeal,” apparently filed after the notice, did set forth that “[t]he trial judge erred in ruling that M. G. L. A. Chapter 271, Section 17 does not encompass an illegal numbers operation and as a result erred in granting the Motion to Strike and the Motion for Judgment of Acquittal.” 68 OCTOBER TERM, 1977 Opinion of the Court 437U.S. its motion for reconsideration filed in the District Court. Indeed, the view that the trial court “dismissed” as to one “discrete basis of liability” appears to have originated in the opinion below. Thus, not only defense counsel and the trial court but the Government as well seemed in agreement that the trial court had made an evidentiary ruling based on its interpretation of the indictment. We must assume that the trial court’s interpretation of the indictment was erroneous. See n. 13, supra. But not every erroneous interpretation of an indictment for purposes of deciding what evidence is admissible can be regarded as a “dismissal.” Here the District Court did not find that the count failed to charge a necessary element of the offense, cf. Lee v. United States, 432 U. S. 23 (1977); rather, it found the indictment’s description of the offense too narrow to warrant the admission of certain evidence. To this extent, we believe the ruling below is properly to be characterized as an erroneous evidentiary ruling,22 which led to an acquittal for insufficient 22 The District Court’s interpretation of the indictment as not encompassing a charge that the gambling business engaged in numbers betting in violation of state law did not by itself require that numbers evidence be excluded. Even if the indictment had charged only that the defendants had conducted an illegal gambling business engaged in horse-betting activities in violation of state law, evidence relating to numbers betting would have been admissible, absent actual surprise or prejudice, to show the defendants’ connection with “all or part of [that] illegal gambling business.” 18 U. S. C. § 1955 (a) (1976 ed.). As the Government repeatedly argued to the District Court, the violation of state law is a jurisdictional element which need only be proved with respect to the business. The District Court’s erroneous assumption that the numbers evidence had to be excluded may have resulted in part from the Government’s failure to repeat in full its earlier argument, see supra, at 58, when the judge ruled that § 17 did not encompass numbers betting, see supra, at 58-59. See n. 5, supra. Had the numbers evidence not been excluded, the judgment of acquittal would not have been entered, even if the court adhered to its ruling on the scope of the indictment, and the case would have gone to the jury, presumably with instructions that the jurors had to find the SANABRIA v. UNITED STATES 69 54 Opinion of the Court evidence. That judgment of acquittal, however erroneous, bars further prosecution on any aspect of the count and hence bars appellate review of the trial court’s error. United States v. Martin Linen Supply Co., 430 U. S., at 571; Fong Foo v. United States, 369 IL S. 141 (1962); Green v. United States, 355 U. S., at 188; United States v. Ball, 163 U. S., at 671. B Even if the Government were correct that the District Court “dismissed” the numbers allegation, in our view a retrial on that theory would subject petitioner to a second trial on the “same offense” of which he has been acquitted.23 It is Congress, and not the prosecution, which establishes and defines offenses. Few, if any, limitations are imposed by the Double Jeopardy Clause on the legislative power to define offenses. Brown v. Ohio, 432 U. S. 161, 165 (1977). But once Congress has defined a statutory offense by its prescription of the “allowable unit of prosecution,” United States v. Universal C. I. T. Credit Corp., 344 IL S. 218, 221 gambling business to have engaged in horse betting, and the defendants to have conducted “all or part” of that gambling business. 23 We agree with the Court of Appeals, see supra, at 61, that there is no statutory barrier to an appeal from an order dismissing only a portion of a count. One express purpose of 18 U. S. C. § 3731 (1976 ed.) is to permit appeals from orders dismissing indictments “as to any one or more counts.” A “count” is the usual organizational subunit of an indictment, and it would therefore appear that Congress intended to authorize appeals from any order dismissing an indictment in whole or in part. Congress could hardly have meant appealability to depend on the initial decision of a prosecutor to charge in one count what could also have been charged in two, a decision frequently fortuitous for purposes of the interests served by §3731. To so rule would import an empty formalism into a statute expressly designed to eliminate “[t]echnical distinctions in pleadings as limitations on appeals by the United States.” H. R. Conf. Rep. No. 91-1768, p. 21 (1970); accord, S. Rep. No. 91-1296, p. 5 (1970). We note that the only Court- of Appeals other than the court below that has considered this question reached a similar result. United States v. Alberti, 568 F. 2d 617 (CA2 1977). 70 OCTOBER TERM, 1977 Opinion of the Court 437 U. S. (195 2) ; Bell v. United States, 349 U. S. 81 (1955) ; Braverman v. United States, 317 U. S. 49 (1942) ; In re Nielsen, 131 U. S. 176 (1889), that prescription determines the scope of protection afforded by a prior conviction or acquittal. Whether a particular course of conduct involves one or more distinct “offenses” under the statute depends on this congressional choice.24 The allowable unit of prosecution under § 1955 is defined as participation in a single “illegal gambling business.” Congress did not assimilate state gambling laws per se into the federal penal code, nor did it define discrete acts of gambling as independent federal offenses. See H. R. Rep. No. 91-1549, p. 53 (1970). See also lannelli v. United States, 420 U. S. 770, 784-790 ( 1975). The Government need not prove that the defendant himself performed any act of gambling prohibited by state law.25 It is participation in the gambling business that is a federal offense, and it is only the gambling business that must violate state law.26 And, as the Government recog- 24 See Note, Twice in Jeopardy, 75 Yale L. J. 262, 268, 302-310 (1965). Because only a single violation of a single statute is at issue here, we do not analyze this case under the so-called “same evidence” test, which is frequently used to determine whether a single transaction may give rise to separate prosecutions, convictions, and/or punishments under separate statutes. See, e. g., Gavieres. v. United States, 220 U. S. 338, 342 (1911); Blockburger v. United States, 284 U. S. 299 (1932); Gore v. United States, 357 U. S. 386 (1958); lannelli v. United States, 420 U. S. 770 (1975). See also Brown v. Ohio, 432 U. S. 161, 166-167, n. 6 (1977); United States v. Jones, 533 F. 2d 1387 (CA6 1976), cert, denied, 431 U. S. 964 (1977). Nor is the case controlled by decisions permitting prosecution under statutes defining as the criminal offense a discrete act, after a prior conviction or acquittal of a distinguishable discrete act that is a separate violation of the statute. See, e. g., Ebeling v. Morgan, 237 U. S. 625 (1915); Burton n. United States, 202 U. S. 344 (1906). Cf. Ladner n. United States, 358 U. S. 169 (1958); Bell n. United States, 349 U. S. 81 (1955). 25 United States v. Hawes, 529 F. 2d 472, 478 (CA5 1976). 26 Numerous cases have recognized that 18 U. S. C. § 1955 (1976 ed.) proscribes any degree of participation in an illegal gambling business, SANABRIA v. UNITED STATES 71 54 Opinion of the Court • nizes, under § 1955 participation in a single gambling business is but a single offense, “no matter how many state statutes the enterprise violated.” Brief for United States 31. The Government’s undisputed theory of this case is that there was a single gambling business, which engaged in both horse betting and numbers betting. With regard to this single business, participation in which is concededly only a single offense, we have no doubt that petitioner was truly acquitted. We have recently defined an acquittal as “ ‘a resolution, correct or not, of some or all of the factual elements of the offense charged.’ ” Lee v. United States, 432 U. S., at 30 n. 8, quoting United States v. Martin Linen Supply Co., supra, at 571. Petitioner was found not guilty for a failure of proof on a key “factual element of the offense charged”: that he was “connected with” the illegal gambling business. See supra, at 59.27 Had the Government charged only that the business except participation as a mere bettor. See, e. g., United States v. DiMuro, 540 F. 2d 503, 507-508 (CAI 1076), cert, denied, 429 U. S. 1038 (1977); United States v. Leon, 534 F. 2d 667,676 (CA6 1976); United States v. Brick, 502 F. 2d 219, 225 n. 17 (CA8 1974); United States v. Smaldone, 485 F. 2d 1333, 1351 (CAIO '1973), cert, denied, 416 U. S. 936 (1974); United States v. Hunter, 478 F. 2d 1019, 1021-1022 (CA7), cert, denied, 414 U. S. 857 (1973); United States v. Ceraso, 467 F. 2d 653, 656 (CA3 1972); United States v. Becker, 461 F. 2d 230, 232-233 (CA2 1972), vacated on other grounds, 417 U. S. 903 (1974). Similarly, the Government need not prove that each defendant participated in an illegal gambling business for more than 30 days (or grossed more than $2,000 in a single day), but only that the business itself existed for more than 30 days (or met the earnings criteria). United States v. Graham, 534 F. 2d 1357, 1359 (CA9 1976) (per curiam); United States v. MarrifiM, 515 F. 2d 877, 880-881 (CA5 1975); United States v. Schaefer, 510 F. 2d 1307, 1312 (CA8), cert, denied sub nom. Del Pietro v. United States, 421 U. S. 975 (1975); United States v. Smaldone, supra, at 1351; see United States v. DiMario, 473 F. 2d 1046, 1048 (CA6), cert, denied, 412 U. S. 907 (1973). 27 The court’s finding that petitioner was not “connected with” the gambling business necessarily meant that he was found not to conduct, finance, manage, supervise, direct, or own it. See 18 U. S. C. § 1955 (a) (1976 ed.). 72 OCTOBER TERM, 1977 Opinion of the Court 437U.S. was engaged in horse betting and had petitioner been acquitted, his acquittal would bar any further prosecution for participating in the same gambling business during the same time period on a numbers theory.28 That the trial court disregarded the Government’s allegation of numbers betting does not render its acquittal on the horse-betting theory any less an acquittal on the “offense” charged. “The Double Jeopardy Clause is not such a fragile guarantee that ... its limitations [can be avoided] by the simple expedient of dividing a single crime into a series of temporal or spatial units,” Brown v. Ohio, 432 U. S., at 169, or, as we hold today, into “discrete bases of liability” not defined as such by the legislature. See id., at 169 n. 8.29 While recognizing that only a single violation of the statute is alleged under either theory,30 the Government nevertheless contends that separate counts would have been proper, and that an acquittal of petitioner on a horse-betting count would not bar another prosecution on a numbers count. Brief for United States 33. Although there may be circumstances in which this is true, petitioner here was acquitted for insufficient proof of an element of the crime which both such counts would share—that he was “connected with” the single gambling business. See supra, at 59. This finding of fact stands as an 28 See 1 C. Wright, Federal Practice and Procedure § 125, p. 241 (1969). See also United States v. Sabella, 272 F. 2d 206, 211 (CA2 1959) (Friendly, J.); Hanj v. United States, 235 F. 2d 710, 715 (CA8), cert, denied, 352 U. S. 880 (1956). 29 See also United States n. Jackson, 560 F. 2d 112, 121 n. 9 (CA2 1977) (Government may not, under Double Jeopardy Clause, “fragment what is in fact a single crime into its components”). 30 The Government concedes that it was required to bring all “theories of liability” in a single trial, and that only a single punishment could be imposed upon conviction on more than one such theory. Brief for United States 31, 33. SANABRIA v. UNITED STATES 73 54 Opinion of the Court absolute bar to any further prosecution for participation in that business.31 The Government having charged only a single gambling business, the discrete violations of state law which that business may have committed are not severable in order to avoid the Double Jeopardy Clause’s bar on retrials for the “same offense.” 32 Indeed, the Government’s argument that these are discrete bases of liability warranting reprosecution following a final judgment of acquittal on one such “discrete basis” is quite similar to an unsuccessful argument that it presented in Braverman v. United States, 317 U. S. 49 (1942). Braverman had been convicted of and received consecutive sentences on four separate counts of conspiracy, each count alleging a conspiracy to violate a separate substantive provision of the federal narcotics laws. The Government conceded that only a single conspiracy existed, as it concedes here that only a single gambling business existed; nonetheless, it urged that separate punishments were appropriate because the single conspiracy had several discrete objects. We firmly rejected that argument: “[T]he precise nature and extent of the conspiracy must be determined by reference to the agreement which embraces and defines its objects. Whether the object of a single agreement is to commit one or many crimes, it is in 31 It is true that no factual determination was made that petitioner had not engaged in numbers betting. Thus, there would be no collateralestoppel bar to a prosecution of petitioner for a different offense in which his liability would depend on proof of that fact. Cf. Ashe v. Swenson, 397 U.S.436 (1970). 32 A single gambling business theoretically may violate as many laws as a State has prohibiting gambling, and § 1955 specifies six means by which a defendant may illegally participate in such a business, i. e., by conducting, financing, managing, supervising, directing, or owning it. If we were to accept the Government’s theory, each of these could be varied, one at a time, to charge a separate count on which a defendant could be reprosecuted following acquittals on any of the others. 74 OCTOBER TERM, 1977 Opinion of the Court 437U.S. either case that agreement which constitutes the conspiracy which the statute punishes. The one agreement cannot be taken to be several agreements and hence several conspiracies because it envisages the violation of several statutes rather than one.” Id., at 53. The same reasoning must also apply where the essence of the crime created by Congress is participation in a “business,” rather than participation in an “agreement.” 33 The Double Jeopardy Clause is no less offended because the Government here seeks to try petitioner twice for this single offense, instead of seeking to punish him twice as it did in Braverman.34 “If two offenses are the same ... for purposes of barring consecutive sentences at a single trial, they necessarily will be the same for purposes of barring successive prosecutions.” Brown v. Ohio, supra, at 166. Accordingly, even if the numbers allegation were “dismissed,” we conclude that a subsequent trial of petitioner for conducting the same illegal gambling business as that at issue in the first trial would subject him to a second trial on the “same offense” of which he was acquitted. 83 If two different gambling businesses were alleged and proved, separate convictions and punishments would be proper. See American Tobacco Co. v. United States, 328 U. S. 781, 787-788 (1946) (holding Braverman inapplicable where two distinct conspiracies alleged). It is not always easy to ascertain whether one or more gambling businesses have been proved under § 1955. See, e. g., United States v. DiMuro, 540 F. 2d, at 508-509; United States v. Bobo, 477 F. 2d 974, 988 (CA4 1973). No such difficulties are presented here because both sides agree that only a single gambling business existed. 34 United States v. Tanner, 471 F. 2d 128, 141 n. 21 (CA7), cert, denied, 409 U. S. 949 (1972); see United States v. Mayes, 512 F. 2d 637, 652 (CA6), cert, denied, 422 U. S. 1008 (1975); United States v. Young, 503 F. 2d 1072, 1075 (CA3 1974); United States v. Cohen, 197 F..2d 26 (CA3 1952). See also Short n. United States, 91 F. 2d 614 (CA4 1937); Powe v. United States, 11 F. 2d 598 (CA5 1926); United States v. Weiss, 293 F. 992 (ND Ill. 1923). SANABRIA v. UNITED STATES 75 54 Opinion of the Court III The only question remaining is whether any of the exceptions to the constitutional rule forbidding successive trials on the same offense, see n. 15, supra, apply here. The short answer to this question is that there is no exception permitting retrial once the defendant has been acquitted, no matter how “egregiously erroneous,” Fong Foo v. United States, 369 U. S., at 143, the legal rulings leading to that judgment might be. The Government nevertheless argues, relying principally on Lee v. United States, 432 U. S. 23 (1977), and Jeffers v. United States, 432 U. S. 137 (1977), that petitioner waived his double jeopardy rights by moving to “dismiss” the numbers allegation and by not objecting to the form of the allegation prior to trial. In Lee we held a retrial permissible because the District Court’s midtrial decision granting the defendant’s motion to dismiss the indictment for failure to state an offense was “functionally indistinguishable from a declaration of mistrial” at the defendant’s request. 432 U. S., at 31. The mistrial analogy relied on in Lee is manifestly inapposite here. Although jeopardy had attached in Lee, no verdict had been rendered; indeed, petitioner conceded that “the District Court’s termination of the first trial was not an acquittal,” id., at 30 n. 8. Here, by contrast, the trial proceeded to verdict, and petitioner was acquitted. While in Lee the trial court clearly did contemplate a reprosecution when it granted defendant’s motion, id., at 30-31, neither petitioner’s motion here nor the trial court’s rulings contemplated a second trial—nor could they have, since only a single offense was involved and petitioner went to judgment on that offense. Where a trial terminates with a judgment of acquittal, as here, “double jeopardy principles governing the permissibility of retrial after a declaration of mistrial,” Lee v. United States, 432 U. 8., at 31, have no bearing. Nor does Jeffers support the Government’s position. The 76 OCTOBER TERM, 1977 Opinion of the Court 437U.S. defendant there was first tried and convicted of conspiring to distribute narcotics in violation of 21 U. S. C. § 846. Eight Members of the Court agreed that his subsequent trial for conducting a continuing criminal enterprise in violation of 21 U. S. C. § 848 during the same time period was on the “same offense,” since the § 846 violation was a lesser included offense to the § 848 violation. Prior to the first trial, however, Jeffers had specifically opposed the Government’s effort to try both indictments together, in part on the ground that they involved distinct offenses. 432 U. S., at 144 n. 8. Reasoning that Jeffers necessarily contemplated a second trial, four Members of the Court found that he had “electfed] to have the two offenses tried separately,” id., at 152, and, by not raising the potential double jeopardy problem, had waived any objection on that ground to successive trials, id., at 152-154.35 The instant case presents quite a different situation. Petitioner’s counsel never argued that horse betting and numbers were distinct offenses,36 a fortiori did not argue for or contemplate 35 While holding that Jeffers could be subjected to a second trial, these four Justices were of the view that the total punishment imposed on Jeffers could not be in excess of that authorized for a single violation of 21 U. S. C. § 848. They relied in part on the fact that Jeffers, who had argued in the District Court that the two statutes involved distinct offenses, had “never affirmatively argued that the difference in the two statutes was so great as to authorize separate punishments . . . .” 432 U. S., at 154 n. 23. They were joined in voting to vacate the excess punishment by the four Justices who believed that Jeffers could not be constitutionally subjected to another trial. Mr. Justice White believed that Jeffers could be subjected to both a second trial and separate punishments. 33 That no such argument was made as to the numbers and horse-betting allegations is highlighted by the fact that petitioner’s counsel did argue on behalf of another defendant that evidence relating to that defendant’s betting on dog races should be excluded because “the theory of the Government’s case is that this is a horse and numbers business. . . . [The dog betting] stands by itself as a separate business, SANABRIA v. UNITED STATES 77 54 Opinion of the Court separate trials on each theory, and a multo fortiori did not “elect” to undergo successive trials. Finally, we agree with the Court of Appeals that this case does not present the hypothetical situation on which we reserved judgment in Serfass v. United States, of “ ‘a defendant who is afforded an opportunity to obtain a determination of a legal defense prior to trial and nevertheless knowingly allows himself to be placed in jeopardy before raising the defense.’ ” 420 U. S., at 394, quoting Solicitor General; see 548 F. 2d, at 7. Petitioner did not have a “legal defense” to the single offense charged: participating in an illegal gambling business in violation of § 1955. Unlike questions of whether an indictment states an offense, a statute is unconstitutional, or conduct set forth in an indictment violates the statute, what proof may be presented in support of a valid indictment and the sufficiency of that proof are not “legal defenses” required to be or even capable of being resolved before trial. In all of the former instances, a ruling in the defendant’s favor completely precludes conviction, at least on that indictment. Here, even if the numbers language had been struck before trial, there was no “legal” reason why petitioner could not have been convicted on this indictment, as were his 10 codefendants. The acquittal resulted from the insufficiency of the Government’s proof at trial to establish petitioner’s connection with the gambling business, as the trial judge erroneously understood it to have been charged. The Government’s real quarrel is with the judgment of acquittal. While the numbers evidence was erroneously excluded, the judgment of acquittal produced thereby is final and unreviewable. Neither 18 U. S. C. § 3731 (1976 ed.) nor and . . . the Government [must] prove one business here. It’s like having multiple conspiracy.” Record 28-29. The motion for exclusion was denied because the District Court found that dog betting was part of the single gambling business shown to have been conducted from the office at 63 Bickford Avenue. Id., at 29-30. 78 OCTOBER TERM, 1977 Stevens, J., concurring 437 U. S. the Double Jeopardy Clause permits the Government to obtain relief from all of the adverse rulings—most of which result from defense motions—that lead to the termination of a criminal trial in the defendant’s favor. See United States v. Wilson, 420 U. 8., at 351-352; S. Rep. No. 91-1296, p. 2 (1970). To hold that a defendant waives his double jeopardy protection whenever a trial court error in his favor on a midtrial motion leads to an acquittal would undercut the adversary assumption on which our system of criminal justice rests, see Jeffers v. United States, 432 U. S., at 159-160 (Stevens, J., dissenting in part and concurring in judgment in part), and would vitiate one of the fundamental rights established by the Fifth Amendment. The trial court’s rulings here led to an erroneous resolution in the defendant’s favor on the merits of the charge. As Fong Foo v. United States makes clear, the Double Jeopardy Clause absolutely bars a second trial in such circumstances. The Court of Appeals thus lacked jurisdiction of the Government’s appeal. Accordingly, the judgment of the Court of Appeals is Reversed. Mr. Justice Stevens, concurring. Although I join the text of the Court’s opinion, I cannot agree with the dictum in footnote 23. It is true “that there is no statutory barrier to an appeal from an order dismissing only a portion of a count,” ante, at 69 n. 23, but it is equally true that there is no statutory authority for such an appeal. It necessarily follows—at least if we are faithful to the concept that federal courts have only such jurisdiction as is conferred by Congress—that the Court of Appeals had no jurisdiction of this appeal. The Criminal Appeals Act, 18 U. S. C. §3731 (1976 ed.), authorizes the United States to appeal an order of a district SANABRIA v. UNITED STATES 79 54 Stevens, J., concurring court “dismissing an indictment or information as to any one or more counts, except that no appeal shall lie where the double jeopardy clause of the United States Constitution prohibits further prosecution.” (Emphasis added.) By its plain terms, this statute does not encompass the present case. Putting to one side the question whether an acquittal may properly be regarded as an order “dismissing an indictment” within the meaning of the statute, see United States v. Martin Linen Supply Co., 430 U. S. 564, 576 (Stevens, J., concurring), the statutory grant of appellate jurisdiction is still unequivocally limited to review of a dismissal “as to any one or more counts.” The statute does not refer to “subunit[s] of an indictment” or “portion[s] of a count,” ante, at 69 n. 23, but only to “counts,” a well-known and unambiguous term of art. Prior to the amendment of § 3731 in 1971, this Court’s rule of statutory interpretation was that “the Criminal Appeals Act [should be] strictly construed against the Government’s right of appeal, Carroll v. United States, 354 U. S. 394, 399-400 (1957).” Will v. United States, 389 U. S. 90, 96-97. The Court’s present pattern of interpretation of § 3731, as exemplified by Martin Linen, supra, does more than simply abandon this approach; it reverses direction entirely and reads the statute in whatever manner would favor a Government appeal. It is, of course, true that the legislative history of the Act indicates that Congress intended § 3731 “to be liberally construed,” S. Rep. No. 91-1296, p. 18 (1970), but this expression of legislative intent does not give us a license to ignore the words of the statute. In fact, the Court does not even suggest that the language “one or more counts” is ambiguous; instead it argues that the words cannot be given their proper meaning because the Act was intended “to eliminate ‘[technical distinctions in pleadings ....’” Ante, at 69 n. 23. This argument has a hollow ring in light of the Court’s prior assertion 80 OCTOBER TERM, 1977 Blackmun, J., dissenting 437 U. S. that “[t]he precise manner in which an indictment is drawn cannot be ignored, because an important function of the indictment is to ensure that, ‘in case any other proceedings are taken against [the defendant] for a similar offence, . . . the record [will] show with accuracy to what extent he may plead a former acquittal or conviction.’ ” Ante, at 65-66. Furthermore, in my judgment, a rule that the Government may appeal from the “dismissal” of a portion of a count, provided that the portion establishes a “discrete basis of liability,” fosters rather than eliminates technical distinctions and encourages exactly the sort of nearsighted parsing of indictments that the amendment was intended to discourage. I cannot, therefore, join that portion of the Court’s decision which states that the Criminal Appeals Act permits an appeal from only a portion of a count. It clearly does not, and for that reason, as well as for the reasons stated in the text of the Court’s opinion, the Court of Appeals’ decision must be reversed. Mr. Justice Blackmun, with whom Mr. Justice Rehnquist joins, dissenting. This case, of course, is an odd and an unusual one, factually and procedurally. Because it is, the case will afford little guidance as precedent in the Court’s continuing struggle to create order and understanding out of the confusion of the lengthening list of its decisions on the Double Jeopardy Clause. I would have thought, however, that the principles enunciated late last Term in Lee v. United States, 432 U. S. 23 (1977)—which I deem a more difficult case for the Government than this one—had application to the facts here. I do not share the Court’s distinction of Lee, ante, at 75, and I do not agree that Lee is “manifestly inapposite.” Here, as in Lee, there is misdescription by the trial court of the nature of its order, and, as in Lee, the defendant-petitioner’s maneu- SANABRIA v. UNITED STATES 81 54 Blackmun, J., dissenting vers should result in a surrender of his right to receive a verdict by the jury that had been drawn. Further, it appears to me that petitioner has succeeded in having the indictment read one way in the trial court, and another way here, as the situation required. I would affirm the judgment of the Court of Appeals. 82 OCTOBER TERM, 1977 Syllabus 437 U. S. UNITED STATES v. SCOTT CERTIORARI TO THE UNITED STATES COURT OF APPEALS FOR THE SIXTH CIRCUIT No. 76-1382. Argued February 21, 1978—Decided June 14, 1978 Respondent, indicted for federal drug offenses, moved before trial and twice during trial for dismissal of two counts of the indictment on the ground that his defense had been prejudiced by preindictment delay. At the close of all the evidence the trial court granted respondent’s motion. The Government sought to appeal the dismissals under 18 U. S. C. §3731 (1976 ed.), which allows the United States to appeal from a district court’s dismissal of an indictment except where the Double Jeopardy Clause of the Fifth Amendment prohibits further prosecution. The Court of Appeals, concluding that that Clause barred further prosecution, dismissed the appeal, relying on United States v. Jenkins, 420 U. S. 358. In that case the Court, following the principle underlying the Double Jeopardy Clause that the Government with all its resources and power should not be allowed to make repeated attempts to convict an individual for an alleged offense, held that, whether or not a dismissal of an indictment after jeopardy had attached amounted to an acquittal on the merits, the Government had no right to appeal because “further proceedings of some sort, devoted to the resolution of factual issues going to the elements of the offense charged, would have been required upon reversal and remand.” Held: Where a defendant himself seeks to have his trial terminated without any submission to either judge or jury as to his guilt or innocence, an appeal by the Government from his successful effort to do so does not offend the Double Jeopardy Clause, and hence is not barred by 18 U. S. C. §3731 (1976 ed.). United States v. Jenkins, supra, overruled. Pp. 87-101. (a) The successful appeal of a judgment of conviction, except on the ground of insufficiency of the evidence to support the verdict, Burks v. United States, ante, p. 1, does not bar further prosecution on the same charge. A judgment of acquittal, whether based on a jury verdict of not guilty or on a ruling by the court that the evidence is insufficient to convict, may not be appealed and terminates the prosecution when a second trial would be necessitated by a reversal. Pp. 87-92. (b) Where no final determination of guilt or innocence has been made a trial judge may declare a mistrial on the motion of the prosecution or UNITED STATES v. SCOTT 83 82 Syllabus upon his own initiative only if “there is a manifest necessity for the act, or the ends of public justice would otherwise be defeated,” United States v. Perez, 9 Wheat. 579, 580, but where a defendant successfully seeks to avoid his trial prior to its conclusion by a motion for a mistrial, the Double Jeopardy Clause is not offended by a second prosecution. Such a motion by the defendant is deemed to be a deliberate election on his part to forgo his valued right to have his guilt or innocence determined by the first trier of fact. United States v. Dinitz, 424 U. S. 600, 609. Pp. 92-94. (c) At least in some cases, the dismissal of an indictment after jeopardy has “attached” may be treated on the same basis as the declaration of a mistrial even though a successful Government appeal would require further trial court proceedings leading to the factual resolution of the issue of guilt or innocence, see Lee v. United States, 432 U. S. 23; and the Court’s growing experience with Government appeals calls for a re-examination of the rationale in Jenkins in light of Lee; United States v. Martin Linen Supply Co., 430 U. S. 564; and other recent expositions of the Double Jeopardy Clause. Pp. 94r-95. (d) In a situation such as the instant one, where a defendant chooses to avoid conviction, not because of his assertion that the Government has failed to make out a case against him, but because of a legal claim that the Government’s case against him must fail even though it might satisfy the trier of fact that he was guilty beyond a reasonable doubt, the defendant by deliberately choosing to seek termination of the trial suffers no injury cognizable under the Double Jeopardy Clause if the Government is permitted to appeal from such a trial-court ruling favoring the defendant. The Double Jeopardy Clause, which guards against Government oppression, does not relieve a defendant of the consequences of his voluntary choice. Pp. 95-101. 544 F. 2d 903, reversed and remanded. Rehnquist, J., delivered the opinion of the Court, in which Burger, C. J., and Stewart, Blackmun, and Powell, JJ., joined. Brennan, J., filed a dissenting opinion, in which White, Marshall, and Stevens, JJ., joined, post, p. 101. Deputy Solicitor General Frey argued the cause for the United States. With him on the brief were Acting Solicitor General Friedman, Assistant Attorney General Civiletti, Frank H. Easterbrook, and Sidney M. Glazer. 84 OCTOBER TERM, 1977 Opinion of the Court 437U.S. William C. Marietti argued the cause for respondent. With him on the brief was Alexis J. Rogoski. Mr. Justice Rehnquist delivered the opinion of the Court. On March 5, 1975, respondent, a member of the police force in Muskegon, Mich., was charged in a three-count indictment with distribution of various narcotics. Both before his trial in the United States District Court for the Western District of Michigan, and twice during the trial, respondent moved to dismiss the two counts of the indictment which concerned transactions that took place during the preceding September, on the ground that his defense had been prejudiced by preindictment delay. At the close of all the evidence, the court granted respondent’s motion. Although the court did not explain its reasons for dismissing the second count, it explicitly concluded that respondent had “presented sufficient proof of prejudice with respect to Count I.” App. to Pet. for Cert. 8a. The court submitted the third count to the jury, which returned a verdict of not guilty. The Government sought to appeal the dismissals of the first two counts to the United States Court of Appeals for the Sixth Circuit. That court, relying on our opinion in United States v. Jenkins, 420 U. S. 358 (1975), concluded that any further prosecution of respondent was barred by the Double Jeopardy Clause of the Fifth Amendment, and therefore dismissed the appeal. 544 F. 2d 903 (1976). The Government has sought review in this Court only with regard to the dismissal of the first count. We granted certiorari to give further consideration to the applicability of the Double Jeopardy Clause to Government appeals from orders granting defense motions to terminate a trial before verdict. We now reverse. I The problem presented by this case could not have arisen during the first century of this Court’s existence. The Court has long taken the view that the United States has no right of UNITED STATES v. SCOTT 85 82 Opinion of the Court appeal in a criminal case, absent explicit statutory authority. United States v. Sanges, 144 U. S. 310 (1892). Such authority was not provided until the enactment of the Criminal Appeals Act, Act of Mar. 2, 1907, ch. 2564, 34 Stat. 1246, which permitted the United States to seek a writ of error in this Court from any decision dismissing an indictment on the basis of “the invalidity, or construction of the statute upon which the indictment is founded.” Our consideration of Government appeals over the ensuing years ordinarily focused upon the intricacies of the Act and its amendments.1 In 1971, however, Congress adopted the current language of the Act, permitting Government appeals from any decision dismissing an indictment, “except that no appeal shall lie where the double jeopardy clause of the United States Constitution prohibits further prosecution.” 18 U. S. C. § 3731 (1976 ed.). Soon thereafter, this Court remarked in a footnote with more optimism than prescience that “[t]he end of our problems with this Act is finally in sight.” United States v. Weller, 401 U. S. 254, 255 n. 1 (1971). For in fact the 1971 amendment did not end the debate over appeals by the Government in criminal cases; it simply shifted the focus of the debate from issues of statutory construction to issues as to the scope and meaning of the Double Jeopardy Clause. In our first encounter with the new statute, we concluded that “Congress intended to remove all statutory barriers to Government appeals and to allow appeals whenever the Constitution would permit.” United States v. Wilson, 420 U. S. 332, 337 (1975). Since up to that point Government appeals had been subject to statutory restrictions independent of the Double Jeopardy Clause, our previous cases construing the statute proved to be of little assistance in determining when the Double Jeopardy Clause of the Fifth Amendment would 1A thorough account of the enactment and development of the Act is set out in Mr. Justice Harlan’s opinion for the Court in United States v. Sisson, 399 U. S. 267, 291-296 (1970). 86 OCTOBER TERM, 1977 Opinion of the Court 437U.S. prohibit further prosecution. A detailed canvass of the history of the double jeopardy principles in English and American law led us to conclude that the Double Jeopardy Clause was primarily “directed at the threat of multiple prosecutions,” and posed no bar to Government appeals “where those appeals would not require a new trial.” Id., at 342. We accordingly held in Jenkins, supra, at 370, that, whether or not a dismissal of an indictment after jeopardy had attached amounted to an acquittal on the merits, the Government had no right to appeal, because “further proceedings of some sort, devoted to the resolution of factual issues going to the elements of the offense charged, would have been required upon reversal and remand.” 2 If Jenkins is a correct statement of the law, the judgment of the Court of Appeals relying on that decision, as it was bound to do, would in all likelihood have to be affirmed.3 Yet, though our assessment of the history and meaning of the Double Jeopardy Clause in Wilson, Jenkins, and Serfass v. United States, 420 U. S. 377 (1975), occurred only three Terms ago, our vastly increased exposure to the various facets of the Double Jeopardy Clause has now convinced us that Jenkins 2 The rule established in Wilson and Jenkins was later described in the following terms: “[Dismissals (as opposed to mistrials) if they occurred at a stage of the proceeding after which jeopardy had attached, but prior to the factfinder’s conclusion as to guilt or innocence, were final so far as the accused defendant was concerned and could not be appealed by the Government because retrial was barred by double jeopardy. This made the issue of double jeopardy turn very largely on temporal considerations—if the Court granted an order of dismissal during the factfinding stage of the proceedings, the defendant could not be reprosecuted, but if the dismissal came later, he could.” Lee v. United States, 432 U. S. 23, 36 (1977) (Rehnquist, J., concurring). 3 The Government contends here that the District Court in Jenkins entered a judgment of acquittal in favor of Jenkins, but our opinion in that case recognized that it could not be said with certainty whether this was the case. See Jenkins, 420 U. S., at 367. UNITED STATES v. SCOTT 87 82 Opinion of the Court was wrongly decided. It placed an unwarrantedly great emphasis on the defendant’s right to have his guilt decided by the first jury empaneled to try him so as to include those cases where the defendant himself seeks to terminate the trial before verdict on grounds unrelated to factual guilt or innocence. We have therefore decided to overrule Jenkins, and thus to reverse the judgment of the Court of Appeals in this case. II The origin and history of the Double Jeopardy Clause are hardly a matter of dispute. See generally Wilson; supra, at 339-340; Green v. United States, 355 U. S. 184, 187-188 (1957); id., at 200 (Frankfurter, J., dissenting). The constitutional provision had its origin in the three common-law pleas of autrefois acquit, autrefois convict, and pardon. These three pleas prevented the retrial of a person who had previously been acquitted, convicted, or pardoned for the same offense. As this Court has described the purpose underlying the prohibition against double jeopardy: “The underlying idea, one that is deeply ingrained in at least the Anglo-American system of jurisprudence, is that the State with all its resources and power should not be allowed to make repeated attempts to convict an individual for an alleged offense, thereby subjecting him to embarrassment, expense and ordeal and compelling him to live in a continuing state of anxiety and insecurity, as well as enhancing the possibility that even though innocent he may be found guilty.” Green, supra, at 187-188. These historical purposes are necessarily general in nature, and their application has come to abound in often subtle distinctions which cannot by any means all be traced to the original three common-law pleas referred to above. Part of the difficulty arises from the development of other protections for criminal defendants in the years since the 88 OCTOBER TERM, 1977 Opinion of the Court 437 U. S. adoption of the Bill of Rights. At the time the Fifth Amendment was adopted, its principles were easily applied, since most criminal prosecutions proceeded to final judgment, and neither the United States nor the defendant had any right to appeal an adverse verdict. See Act of Sept. 24, 1789, ch. 20, § 22, 1 Stat. 84. The verdict in such a case was unquestionably final, and could be raised in bar against any further prosecution for the same offense. Soon thereafter, Congress made provision for review of certain criminal cases by this Court, but only upon a certificate of division from the circuit court, and not at the instigation of the defendant. Act of Apr. 29, 1802, ch. 31, § 6, 2 Stat. 159. It was not until 1889 that Congress permitted criminal defendants to seek a writ of error in this Court, and then only in capital cases. Act of Feb. 6, 1889, ch. 113, § 6, 25 Stat. 656.4 Only then did it become necessary for this Court to deal with the issues presented by the challenge of verdicts on appeal. And, in the very first case presenting the issues, United States v. Ball, 163 U. S. 662 (1896), the Court established principles that have been adhered to ever since. Three persons had been tried together for murder; two were convicted, the other acquitted. This Court reversed the convictions, finding the indictment fatally defective, Ball v. United States, 140 U. S. 118 (1891), whereupon all three defendants were tried again. This time all three were convicted and they again sought review here. This Court held that the Double Jeopardy Clause precluded further prosecution of the defendant who had been acquitted at the original trial5 but that it posed no such 4 Two years later, review was provided for all “infamous” crimes. Act of Mar. 3, 1891, ch. 517, § 5, 26 Stat. 827. 5 The Court thereby rejected the English rule set out in Vaux’s Case, 4 Co. Rep. 44a, 76 Eng. Rep. 992 (K. B. 1590), which refused to recognize a plea of autrefois acquit where the initial indictment had been insufficient to support a conviction. Again, this ruling provided a greater measure of protection for criminal defendants than had been known at the time of the UNITED STATES v. SCOTT 89 82 Opinion of the Court bar to the prosecution of those defendants who had been convicted in the earlier proceeding. The Court disposed of their objection almost peremptorily: “Their plea of former conviction cannot be sustained, because upon a writ of error sued out by themselves the judgment and sentence against them were reversed, and the indictment ordered to be dismissed. ... [I]t is quite clear that a defendant, who procures a judgment against him upon an indictment to be set aside, may be tried anew upon the same indictment, or upon another indictment, for the same offence of which he had been convicted.” 163 U. S., at 671-672. Although Ball firmly established that a successful appeal of a conviction precludes a subsequent plea of double jeopardy, the opinion shed no light on whether a judgment of acquittal could be reversed on appeal consistently with the Double Jeopardy Clause. Because of the statutory restrictions upon Government appeals in criminal cases, this Court in the years after Ball was faced with that question only in unusual circumstances, such as were present in Kepner v. United States, 195 U. S. 100 (1904). That case arose out of a criminal prosecution in the Philippine Islands, to which the principles of the Double Jeopardy Clause had been expressly made applicable by Act of Congress. Although the defendant had been acquitted in his original trial, traditional Philippine procedure provided for a trial de novo upon appeal. This Court, in reversing the resulting conviction, remarked: “The court of first instance, having jurisdiction to try the question of the guilt or innocence of the accused, found Kepner not guilty; to try him again upon the merits, even adoption of the Constitution. A contrary ruling would have altered this Court’s task in such cases as Lee v. United States, 432 U. S. 23 (1977), and Illinois v. Somerville, 410 U. S. 458 (1973). 90 OCTOBER TERM, 1977 Opinion of the Court 437 U. S. in an appellate court, is to put him a second time in jeopardy for the same offense . . . .” Id., at 133.6 More than 50 years later, in Fong Foo v. United States, 369 U. S. 141 (1962), this Court reviewed the issuance of a writ of mandamus by the Court of Appeals for the First Circuit instructing a District Court to vacate certain judgments of acquittal. Although indicating its agreement with the Court of Appeals that the judgments had been entered erroneously, this Court nonetheless held that a second trial was barred by the Double Jeopardy Clause. Id., at 143. Only last Term, this Court relied upon these precedents in United States v. Martin Linen Supply Co., 430 U. S. 564 (1977), and held that the Government could not appeal the granting of a motion to acquit pursuant to Fed. Rule Crim. Proc. 29 where a second trial would be required upon remand. The Court, quoting language in Ball, supra, at 671, stated: “Perhaps the most fundamental rule in the history of double jeopardy jurisprudence has been that ‘[a] verdict of acquittal . . . could not be reviewed, on error or otherwise, without putting [a defendant] twice in jeopardy, and thereby violating the Constitution.’ ” 430 U. S., at 571. These, then, at least, are two venerable principles of double jeopardy jurisprudence. The successful appeal of a judgment of conviction, on any ground other than the insufficiency of 6 In so doing, the Court rejected the contention of Mr. Justice Holmes in dissent that “there is no rule that a man may not be tried twice in the same case.” 195 U. S., at 134. He went on to say: “If a statute should give the right to take exceptions to the Government, I believe it would be impossible to maintain that the prisoner would be protected by the Constitution from being tried again. He no more would be put in jeopardy a second time when retried because of a mistake of law in his favor, than he would be when retried for a mistake that did him harm.” Id.. at 135. Mr. Justice Holmes’ concept of continuing jeopardy would have greatly simplified the matter of Government appeals, but it has never been accepted by a majority of this Court. See Jenkins, 420 IT. S., at 358. UNITED STATES v. SCOTT 91 82 Opinion of the Court the evidence to support the verdict, Burks n. United States, ante, p. 1, poses no bar to further prosecution on the same charge. A judgment of acquittal, whether based on a jury verdict of not guilty or on a ruling by the court that the evidence is insufficient to convict, may not be appealed and terminates the prosecution when a second trial would be necessitated by a reversal.7 What may seem superficially to be a disparity in the rules governing a defendant’s liability to be tried again is explainable by reference to the underlying purposes of the Double Jeopardy Clause. As Kepner and Fong Foo illustrate, the law attaches particular significance to an acquittal. To permit a second trial after an acquittal, however mistaken the acquittal may have been, would present an unacceptably high risk that the Government, with its vastly superior resources, might wear down the defendant so that “even though innocent he may be found guilty.” Green, 355 U. S., at 188. On the other hand, to require a criminal defendant to stand trial again after he has successfully invoked a statutory right of appeal to upset his first conviction is not an act of governmental oppression of the sort against which the Double Jeopardy Clause was intended to protect. The common sense of the matter is most pithily, if not most elegantly, expressed in the words of Mr. Justice McLean on circuit in United States v. Keen, 26 F. Cas. 686 (No. 15,510) 7 In Jenkins we had assumed that a judgment of acquittal could be appealed where no retrial would be needed on remand: “When this principle is applied to the situation where the jury returns a verdict of guilt but the trial court thereafter enters a judgment of acquittal, an appeal is permitted. In that situation a conclusion by an appellate court that the judgment of acquittal was improper does not require a criminal defendant to submit to a second trial; the error can be corrected on remand by the entry of a judgment on the verdict.” Id., at 365. Despite the Court’s heavy emphasis on the finality of an acquittal in Martin Linen and Sanabria v. United States, ante, p. 54, neither decision explicitly repudiates this assumption. Sanabria, ante, at 75; Martin Linen, 430 U. S., at 569-570. 92 OCTOBER TERM, 1977 Opinion of the Court 437U.S. ((DC Ind. 1839). He vigorously rejected the view that the Double Jeopardy Clause prohibited any new trial after the setting aside of a judgment of conviction against the defendant or that it “guarantees to him the right of being hung, to protect him from the danger of a second trial.” Id., at 690. Ill Although the primary purpose of the Double Jeopardy Clause was to protect the integrity of a final judgment, see Crist v. Bretz, ante, at 33, this Court has also developed a body of law guarding the separate but related interest of a defendant in avoiding multiple prosecutions even where no final determination of guilt or innocence has been made. Such interests may be involved in two different situations: the first, in which the trial judge declares a mistrial; the second, in which the trial judge terminates the proceedings favorably to the defendant on a basis not related to factual guilt or innocence. A When a trial court declares a mistrial, it all but invariably contemplates that the prosecutor will be permitted to proceed anew notwithstanding the defendant’s plea of double jeopardy. See Lee v. United States, 432 U. S. 23, 30 (1977). Such a motion may be granted upon the initiative of either party or upon the court’s own initiative. The fact that the trial judge contemplates that there will be a new trial is not conclusive on the issue of double jeopardy; in passing on the propriety of a declaration of mistrial granted at the behest of the prosecutor or on the court’s own motion, this Court has balanced “the valued right of a defendant to have his trial completed by the particular tribunal summoned to sit in judgment on him,” Downum v. United States, 372 U. S. 734, 736 (1963), against the public interest in insuring that justice is meted out to offenders. Our very first encounter with this situation came in United UNITED STATES v. SCOTT 93 82 Opinion of the Court States v. Perez, 9 Wheat. 579 (1824), in which the trial judge had on his own motion declared a mistrial because of the jury’s inability to reach a verdict. The Court said that trial judges might declare mistrials “whenever, in their opinion, taking all the circumstances into consideration, there is a manifest necessity for the act, or the ends of public justice would otherwise be defeated.” Id., at 580. In our recent decision in Arizona n. Washington, 434 U. S. 497 (1978), we reviewed this Court’s attempts to give content to the term “manifest necessity.” That case, like Downum, supra,* arose from a motion of the prosecution for a mistrial, and we noted that the trial court’s discretion must be exercised with a careful regard for the interests first described in United States v. Perez. Arizona v. Washington, supra, at 514-516. Where, on the other hand, a defendant successfully seeks to avoid his trial prior to its conclusion by a motion for mistrial, the Double Jeopardy Clause is not offended by a second prosecution. “[A] motion by the defendant for mistrial is ordinarily assumed to remove any barrier to reprosecution, even if the defendant’s motion is necessitated by a prosecutorial or judicial error.” United States v. Jorn, 400 U. S. 470, 485 (1971) (opinion of Harlan, J.). Such a motion by the defendant is deemed to be a deliberate election on his part to forgo his valued right to have his guilt or innocence determined before the first trier of fact. “The important considera- 8 Downum, in 1963, was the first case in which this Court actually reversed a subsequent conviction because of an improper declaration of a mistrial. This, too, provided greater protection for a defendant than was available at the common law. Although English precedents clearly disapproved of unnecessary mistrials, see generally Arizona v. Washington, 434 U. S., at 506-508, and nn. 21-23, the English rule at the time of the adoption of the Constitution was, as it remains today, that nothing short of a final judgment would bar further prosecution. “The fact that the jury was discharged without giving a verdict cannot be a bar to a subsequent indictment.” 11 Halsbury’s Laws of England, Criminal Law, Evidence, and Procedure If 242 (4th ed. 1976). 94 OCTOBER TERM, 1977 Opinion of the Court 437 U. S. tion, for purposes of the Double Jeopardy Clause, is that the defendant retain primary control over the course to be followed in the event of such error.” United States v. Dinitz, 424 U. S. 600, 609 (1976). But “[t]he Double Jeopardy Clause does protect a defendant against governmental actions intended to provoke mistrial requests and thereby to subject defendants to the substantial burdens imposed by multiple prosecutions.” Id., at 611. B We turn now to the relationship between the Double Jeopardy Clause and reprosecution of a defendant who has successfully obtained not a mistrial but a termination of the trial in his favor before any determination of factual guilt or innocence. Unlike the typical mistrial, the granting of a motion such as this obviously contemplates that the proceedings will terminate then and there in favor of the defendant. The prosecution, if it wishes to reinstate the proceedings in the face of such a ruling, ordinarily must seek reversal of the decision of the trial court. The Criminal Appeals Act, 18 U. S. C. § 3731 (1976 ed.), as previously noted, makes appealability of a ruling favorable to the defendant depend upon whether further proceedings upon reversal would be barred by the Double Jeopardy Clause. Jenkins, 420 U. S., at 370, held that, regardless of the character of the midtrial termination, appeal was barred if “further proceedings of some sort, devoted to the resolution of factual issues going to the elements of the offense charged, would have been required upon reversal and remand.” However, only last Term, in Lee, supra, the Government was permitted to institute a second prosecution after a midtrial dismissal of an indictment. The Court found the circumstances presented by that case “functionally indistinguishable from a declaration of mistrial.” 432 U. S., at 31. Thus, Lee demonstrated that, at least in some cases, the dismissal of an indictment may be treated on the same basis as the declaration of a mistrial. UNITED STATES v. SCOTT 95 82 Opinion of the Court In the present case, the District Court’s dismissal of the first count of the indictment was based upon a claim of preindictment delay and not on the court’s conclusion that the Government had not produced sufficient evidence to establish the guilt of the defendant. Respondent Scott points out quite correctly that he had moved to dismiss the indictment on this ground prior to trial, and that had the District Court chosen to grant it at that time the Government could have appealed the ruling under our holding in Serfass v. United States, 420 U. S. 377 (1975). He also quite correctly points out that jeopardy had undeniably “attached” at the time the District Court terminated the trial in his favor; since a successful Government appeal would require further proceedings in the District Court leading to a factual resolution of the issue of guilt or innocence, Jenkins bars the Government’s appeal. However, our growing experience with Government appeals convinces us that we must re-examine the rationale of Jenkins in light of Lee, Martin Linen, and other recent expositions of the Double Jeopardy Clause. IV Our decision in Jenkins was based upon our perceptions of the underlying purposes of the Double Jeopardy Clause, see supra, at 87: “ ‘The underlying idea, one that is deeply ingrained in at least the Anglo-American system of jurisprudence, is that the State with all its resources and power should not be allowed to make repeated attempts to convict an individual for an alleged offense, thereby subjecting him to embarrassment, expense and ordeal and compelling him to live in a continuing state of anxiety and insecurity Jenkins, supra, at 370, quoting Green, 355 U. S., at 187. Upon fuller consideration, we are now of the view that this language from Green, while entirely appropriate in the circumstances of that opinion, is not a principle which can be 96 OCTOBER TERM, 1977 Opinion of the Court 437 U. S. expanded to include situations in which the defendant is responsible for the second prosecution. It is quite true that the Government with all its resources and power should not be allowed to make repeated attempts to convict an individual for an alleged offense. This truth is expressed in the three common-law pleas of autrefois acquit, autrefois convict, and pardon, which lie at the core of the area protected by the Double Jeopardy Clause. As we have recognized in cases from United States v. Ball, 163 U. S. 662 (1896), to Sanabria v. United States, ante, p. 54, a defendant once acquitted may not be again subjected to trial without violating the Double Jeopardy Clause. But that situation is obviously a far cry from the present case, where the Government was quite willing to continue with its production of evidence to show the defendant guilty before the jury first empaneled to try him, but the defendant elected to seek termination of the trial on grounds unrelated to guilt or innocence. This is scarcely a picture of an all-powerful state relentlessly pursuing a defendant who had either been found not guilty or who had at least insisted on having the issue of guilt submitted to the first trier of fact. It is instead a picture of a defendant who chooses to avoid conviction and imprisonment, not because of his assertion that the Government has failed to make out a case against him, but because of a legal claim that the Government’s case against him must fail even though it might satisfy the trier of fact that he was guilty beyond a reasonable doubt. We have previously noted that “the trial judge’s characterization of his own action cannot control the classification of the action.” Jorn, 400 U. S., at 478 n. 7 (opinion of Harlan, J.), citing United States v. Sisson, 399 U. S. 267, 290 (1970). See also Martin Linen, 430 U. S., at 571; Wilson, 420 U. S., at 336. Despite respondent’s contentions, an appeal is not barred simply because a ruling in favor of a defendant “is based upon facts outside the face of the indictment,” id., at 348, or be- UNITED STATES v. SCOTT 97 82 Opinion of the Court cause it “is granted on the ground . . . that the defendant simply cannot be convicted of the offense charged,” Lee, 432 U. 8., at 30. Rather, a defendant is acquitted only when “the ruling of the judge, whatever its label, actually represents a resolution [in the defendant’s favor], correct or not, of some or all of the factual elements of the offense charged.” Martin Linen, supra, at 571. Where the court, before the jury returns a verdict, enters a judgment of acquittal pursuant to Fed. Rule Crim. Proc. 29, appeal will be barred only when “it is plain that the District Court . . . evaluated the Government’s evidence and determined that it was legally insufficient to sustain a conviction.” 430 U. 8., at 572.® Our opinion in Burks necessarily holds that there has been a “failure of proof,” ante, at 16, requiring an acquittal when the Government does not submit sufficient evidence to rebut a defendant’s essentially factual defense of insanity, though it may otherwise be entitled to have its case submitted to the jury. The defense of insanity, like the defense of entrapment, arises from “the notion that Congress could not have intended criminal punishment for a defendant who has committed all the elements of a proscribed offense,” United States v. Russell, 411 U. S. 423, 435 (1973), where other facts established to the satisfaction of the trier of fact provide a legally 9 In Jenkins, which was a bench trial, we had difficulty, as did the Court of Appeals in that case, in characterizing the precise import of the District Court’s order dismissing the indictment. The analysis that governed our disposition turned not on whether the defendant had been acquitted but on whether the proceeding had terminated “in the defendant’s favor,” 420 U. S., at 365 n. 7, and whether “further proceedings of some sort, devoted to the resolution of factual issues going to the elements of the offense charged, would have been required upon reversal and remand,” id., at 370. We thus had no occasion to determine whether the District Court simply had made “an erroneous interpretation of the controlling law,” id., at 365 n. 7, or whether it had “resolved [controlling] issues of fact in favor of the respondent,” id., at 367; see id., at 362 n. 3. 98 OCTOBER TERM, 1977 Opinion of the Court 437 U. S. adequate justification for otherwise criminal acts.10 Such a factual finding does “necessarily establish the criminal defendant’s lack of criminal culpability,” post, at 106 (Brennan, J., dissenting), under the existing law; the fact that “the acquittal may result from erroneous evidentiary rulings or erroneous interpretations of governing legal principles,” ibid., affects the accuracy of that determination, but it does not alter its essential character. By contrast, the dismissal of an indictment for preindictment delay represents a legal judgment that a defendant, although criminally culpable, may not be punished because of a supposed constitutional violation.11 We think that in a case such as this the defendant, by deliberately choosing to seek termination of the proceedings 10 The defense of insanity in a federal criminal prosecution was first recognized by this Court in Davis v. United States, 160 U. S. 469 (1895). Mr. Justice Harlan’s opinion for the Court construed federal law in light of the larger body of common law in other jurisdictions, and concluded: “One who takes human life cannot be said to be actuated by malice aforethought, or to have deliberately intended to take fife, or to have ‘a wicked, depraved, and malignant heart,’ or a heart ‘regardless of society duty and fatally bent on mischief’ unless at the time he had sufficient mind to comprehend the criminality or the right and wrong of such an act.” Id., at 485. While Congress has never made explicit statutory provision for this affirmative defense or any other, it has recognized the validity of the defense by regulating its use in federal prosecutions. Fed. Rule Crim. Proc. 12.2 (a). 11 While an acquittal on the merits by the trier of fact “can never represent a determination that the criminal defendant is innocent in any absolute sense,” post, at 107 (Brennan, J., dissenting), a defendant who has been released by a court for reasons required by the Constitution or laws, but which are unrelated to factual guilt or innocence, has not been determined to be innocent in any sense of that word, absolute or otherwise. In other circumstances, this Court has had no difficulty in distinguishing between those rulings which relate to “the ultimate question of guilt or innocence” and those which serve other purposes. Stone v. Powell. 428 U. S. 465, 490 (1976). We reject the contrary implication of the dissent that this Court or other courts are incapable of distinguishing between the latter and the former. UNITED STATES v. SCOTT 99 82 Opinion of the Court against him on a basis unrelated to factual guilt or innocence of the offense of which he is accused, suffers no injury cognizable under the Double Jeopardy Clause if the Government is permitted to appeal from such a ruling of the trial court in favor of the defendant. We do not thereby adopt the doctrine of “waiver” of double jeopardy rejected in Green.12 Rather, we conclude that the Double Jeopardy Clause, which guards against Government oppression, does not relieve a defendant from the consequences of his voluntary choice. In Green the question of the defendant’s factual guilt or innocence of murder in the first degree was actually submitted to the jury as a trier of fact; in the present case, respondent successfully avoided such a submission of the first count of the indictment by persuading the trial court to dismiss it on a basis which did not depend on guilt or innocence. He was thus neither acquitted nor convicted, because he himself successfully undertook to persuade the trial court not to submit the issue of guilt or innocence to the jury which had been empaneled to try him. The reason for treating a trial aborted on the initiative of the trial judge differently from a trial verdict reversed on appeal, for purposes of double jeopardy, is thus described in Jorn, 400 U. S., at 484 (opinion of Harlan, J.): “[I]n the [second] situation the defendant has not been deprived of his option to go to the first jury, and, perhaps, end the dispute then and there with an acquittal. On the other hand, where the judge, acting without the defendant’s consent, aborts the proceeding, the defendant has 12 The original jury in that case had found the defendant guilty of second-degree murder, but did not find him guilty of first-degree murder. The Court held that his appeal did not waive his objection to a second prosecution for first-degree murder, but it was careful to reaffirm the holding of United States v. Ball, 163 U. S. 662 (1896), that “a defendant can be tried a second time for an offense when his prior conviction for that same offense [has] been set aside on appeal.” 355 U. S., at 189. 100 OCTOBER TERM, 1977 Opinion of the Court 437 U. S. been deprived of his ‘valued right to have his trial completed by a particular tribunal.’ ” We think the same reasoning applies in pari passu where the defendant, instead of obtaining a reversal of his conviction on appeal, obtains the termination of the proceedings against him in the trial court without any finding by a court or jury as to his guilt or innocence. He has not been “deprived” of his valued right to go to the first jury; only the public has been deprived of its valued right to “one complete opportunity to convict those who have violated its laws.” Arizona v. Washington, 434 U. S., at 509. No interest protected by the Double Jeopardy Clause is invaded when the Government is allowed to appeal and seek reversal of such a midtrial termination of the proceedings in a manner favorable to the defendant.13 It is obvious from what we have said that we believe we pressed too far in Jenkins the concept of the “defendant’s valued right to have his trial completed by a particular tri 13 We should point out that it is entirely possible for a trial court to reconcile the public interest in the Government’s right to appeal from an erroneous conclusion of law with the defendant’s interest in avoiding a second prosecution. In United States v. Wilson, 420 U. S. 332 (1975), the court permitted the case to go to the jury, which returned a verdict of guilty, but it subsequently dismissed the indictment for preindictment delay on the basis of evidence adduced at trial. Most recently in United States v. Ceccolini, 435 U. S. 268 (1978), we described similar action with approval: “The District Court had sensibly first made its finding on the factual question of guilt or innocence, and then ruled on the motion to suppress; a reversal of these rulings would require no further proceedings in the District Court, but merely a reinstatement of the finding of guilt.” Id., at 271. Accord, United States n. Kopp, 429 U. S. 121 (1976); United States v. Rose, 429 U. S. 5 (1976); United States v. Morrison, 429 U. S. 1 (1976). We, of course, do not suggest that a midtrial dismissal of a prosecution, in response to a defense motion on grounds unrelated to guilt or innocence, is necessarily improper. Such rulings may be necessary to terminate proceedings marred by fundamental error. But where a defendant prevails on such a motion, he takes the risk that an appellate court will reverse the trial court. UNITED STATES v. SCOTT 101 82 Brennan, J., dissenting bunal.” Wade v. Hunter, 336 U. S. 684, 689 (1949). We now conclude that where the defendant himself seeks to have the trial terminated without any submission to either judge or jury as to his guilt or innocence, an appeal by the Government from his successful effort to do so is not barred by 18 U. S. C. §3731 (1976 ed.). We recognize the force of the doctrine of stare decisis, but we are conscious as well of the admonition of Mr. Justice Brandeis: “[I]n cases involving the Federal Constitution, where correction through legislative action is practically impossible, this Court has often overruled its earlier decisions. The Court bows to the lessons of experience and the force of better reasoning, recognizing that the process of trial and error, so fruitful in the physical sciences, is appropriate also in the judicial function.” Burnet v. Coronado Oil & Gas Co., 285 U. S. 393, 406-408 (1932) (dissenting opinion). Here, “the lessons of experience” indicate that Government appeals from midtrial dismissals requested by the defendant would significantly advance the public interest in assuring that each defendant shall be subject to a just judgment on the merits of his case, without “enhancing the possibility that even though innocent he may be found guilty.” Green, 355 U. S., at 188. Accordingly, the contrary holding of United States v. Jenkins is overruled. The judgment of the Court of Appeals is therefore reversed, and the cause is remanded for further proceedings. It is so ordered. Mr. Justice Brennan, with whom Mr. Justice White, Mr. Justice Marshall, and Mr. Justice Stevens join, dissenting. On the basis of his evaluation of the trial evidence, the District Judge concluded that unjustifiable preindictment de 102 OCTOBER TERM, 1977 Brennan, J., dissenting 437 U.S. lay had so prejudiced respondent’s defense as to preclude— consistently with the Due Process Clause—his conviction of the offense alleged in count one of the indictment. He therefore dismissed this count with prejudice. Under the principles of double jeopardy law that controlled until today, further prosecution of respondent under count one would unquestionably be prohibited, and appeal by the United States from the judgment of dismissal thus would not lie. See 18 U. S. C. § 3731 (1976 ed.). The dismissal would, under prior law, have been treated as an “acquittal”—i. e., “a legal determination on the basis of facts adduced at the trial relating to the general issue of the case.” United States v. Martin Linen Supply Co., 430 U. S. 564, 575 (1977) (citations omitted). Indeed, further proceedings would have been barred even if the dismissal could not have been so characterized. United States n. Jenkins, 420 U. S. 358 (1975), established that, even if a midtrial termination does not amount to an “acquittal,” an appeal by the United States from the dismissal would not lie if a reversal would, as is of course true in the present case, require “further proceedings of some sort, devoted to the resolution of factual issues going to the elements of the offense charged.” Id., at 370. This principle was reaffirmed only last Term in Lee v. United States, 432 U. S. 23, 30 (1977): “Where a midtrial dismissal is granted on the ground, correct or not, that the defendant simply cannot be convicted of the offense charged, . . . further prosecution is barred by the Double Jeopardy Clause.”1 But the Court today overrules the principle recognized in Jenkins and Lee. While reaffirming that the Government may not appeal from judgments of “acquittal” when reversals would require new trials, the Court holds that appeals by the United States will lie from all other final judgments favor- xSee also Finch n. United States, 433 U. S. 676 (1977) (applying rule of Jenkins to dismissal entered on basis of stipulated facts); United States n. Martin Linen Supply Co., 430 U. S. 564 (1977). UNITED STATES v. SCOTT 103 82 Brennan, J., dissenting able to the accused. The Court implements this new rule by fashioning a more restrictive definition of “acquittal” than heretofore followed—i. e., “a resolution, correct or not, of some or all of the factual elements of the offense”—and holds, without explanation, that, under that restrictive definition, respondent was not “acquitted” when the District Judge concluded that the facts adduced at trial established that unjustifiable and prejudicial preindictment delay gave respondent a complete defense to the charges contained in count one. I dissent. I would not overrule the rule announced in Jenkins and reaffirmed in Lee. This principle is vital to the implementation of the values protected by the Double Jeopardy Clause; indeed, it follows necessarily from the very rule the Court today reaffirms. The Court’s attempt to draw a distinction between “true acquittals” and other final judgments favorable to the accused, quite simply, is unsupportable in either logic or policy. Equally fundamental, the decision today indefensibly adopts an overly restrictive definition of “acquittal.” Its definition, moreover, in sharp contrast to the rule of Jenkins, is incapable of principled application. That is vividly evident in the Court’s own distinction between a dismissal based on a finding of preaccusation delay violative of due process, and a dismissal based upon evidence adduced at trial in support of a defense of insanity or of entrapment. Ante, at 97-98. Why should the dismissal in the latter cases raise a double jeopardy bar, but the dismissal based on preaccusation delay not also raise that bar to a retrial? The Court ventures no persuasive explanation. Because the thousands of state and federal judges who must apply today’s decision to similar “affirmative defenses” are left without meaningful guidance, only confusion can result from today’s decision. I The Court reaffirms the “most fundamental rule in the history of double jeopardy jurisprudence”: that judgments of 104 OCTOBER TERM, 1977 Brennan, J., dissenting 437 U. S. acquittal, no matter how erroneous, bar any retrial and thus that, under the proviso in 18 U. S. C. § 3731 (1976 ed.),2 appeals by the United States will not lie when reversal would require a retrial.3 The major premise for the Court’s conclusion that the Government may appeal from the final judgment entered for respondent is that there is a difference of constitutional magnitude between “acquittals” and midtrial dismissals, entered on motion of the accused, on grounds “unrelated to factual innocence.” This premise is fatally flawed. It, quite simply, misconceives the whole basis for the rule that “acquittals” bar retrials. The reason for this rule is not, as the Court suggests, primarily to safeguard determinations of innocence; rather, it is that a retrial following a final judgment for the accused necessarily threatens intolerable interference with the constitutional policy against multiple trials. Moreover, in terms of the practical operation of the adversary process, there is actually no difference between a so-called “true acquittal” and the termination in this case favorably to respondent. A While the Double Jeopardy Clause often has the effect of protecting the accused’s interest in the finality of particular favorable determinations, this is not its objective. For the Clause often permits Government appeals from final judgments favorable to the accused. See United States v. Wilson, 420 U. S. 332 (1975) (whether or not final judgment was an acquittal, Government may appeal if reversal would not ne 2 Section 3731 provides that the United States may obtain appellate review of a "dismissal” "except that no appeal shall lie where the double jeopardy clause of the United States Constitution prohibits further prosecution.” 3 The Court cites with approval Sanabria v. United States, ante, p. 54; United States v. Martin Linen Supply Co., supra; Fong Foo v. United States, 369 U. S. 141 (1962); Kepner n. United States, 195 U. S. 100 (1904); and United States v. Ball, 163 U. S. 662 (1896). UNITED STATES v. SCOTT 105 82 Brennan, J., dissenting cessitate a retrial). The purpose of the Clause, which the Court today fails sufficiently to appreciate, is to protect the accused against the agony and risks attendant upon undergoing more than one criminal trial for any single offense. See ibid. A retrial increases the financial and emotional burden that any criminal trial represents for the accused, prolongs the period of the unresolved accusation of wrongdoing, and enhances the risk that an innocent defendant may be convicted.4 See Arizona v. Washington, 434 U. S. 497, 503-504 (1978); Green v. United States, 355 U. S. 184, 187-188 (1957). Society’s “willingness to limit the Government to a single criminal proceeding to vindicate its very vital interest in enforcement of criminal laws” bespeaks society’s recognition of the gross unfairness of requiring the accused to undergo the strain and agony of more than one trial for any single offense. United States v. Jorn, 400 U. S. 470,479 (1971) (opinion of Harlan, J.). Accordingly, the policies of the Double Jeopardy Clause mandate that the Government be afforded but one complete opportunity to convict an accused and that when the first proceeding terminates in a final judgment favorable to the defendant5 any retrial be barred. The rule as to acquittals can only be understood as simply an application of this larger principle. Judgments of acquittal normally result from jury or bench 4 There are a number of reasons a retrial enhances the risk that “even though innocent, [the criminal defendant] may be found guilty.” Green v. United States, 355 U. 8. 184, 188 (1957). A retrial affords the Government the opportunity to re-examine the weaknesses of its first presentation in order to strengthen the second. And, as would any litigant, the Government has been known to take advantage of this opportunity. It is not uncommon to find that prosecution witnesses change their testimony, not always subtly, at second trials. See Arizona v. Washington, 434 U. S. 497, 504 n. 14 (1978), quoting Carsey v. United States, 129 U. S. App. D. C. 205, 208-209, 392 F. 2d 810, 813-814 (1967). 6 By “final judgment favorable to the accused,” I am, of course, referring to an order terminating all prosecution of the defendant on the ground he “simply cannot be convicted of the offense charged.” See Lee n. United States, 432 U. S. 23, 30 (1977). 106 OCTOBER TERM, 1977 Brennan, J., dissenting 437 U. S. verdicts of not guilty. In such cases, the acquittal represents the factfinder’s conclusion that, under the controlling legal principles, the evidence does not establish that the defendant can be cohvicted of the offense charged in the indictment. But the judgment does not necessarily establish the criminal defendant’s lack of criminal culpability; the acquittal may result from erroneous evidentiary rulings or erroneous interpretations of governing legal principles induced by the defense. Yet the Double Jeopardy Clause bars a second trial. In repeatedly holding that the Government may not appeal from an acquittal if a reversal would necessitate a retrial, the Court has, of course, recognized that this rule impairs to some degree the Government’s interest in enforcing its criminal laws. Yet, while we have acknowledged that permitting review of acquittals would avoid release of guilty defendants who benefited from “error, irrational behavior, or prejudice on the part of the trial judge,” United States N. Martin Linen Supply Co., 430 U. S., at 574; see United States n. Wilson, supra, at 352, we nevertheless have consistently held that the Double Jeopardy Clause bars any appellate review in such circumstances. The reason is not that the first trial established the defendant’s factual innocence, but rather that the second trial would present all the untoward consequences the Clause was designed to prevent. The Government would be allowed to seek to persuade a second trier of fact of the defendant’s guilt, to strengthen any weaknesses in its first presentation, and to subject the defendant to the expense and anxiety of a second trial. See ibid. This basic principle of double jeopardy law has heretofore applied not only to acquittals based on the verdict of the factfinder, but also to acquittals entered by the trial judge, following the presentation of evidence but before verdict, pursuant to Fed. Rule Crim. Proc. 29. See Sanabria v. United States, ante, p. 54; United States v. Martin Linen Supply Co., UNITED STATES v. SCOTT 107 82 Brennan, J., dissenting supra; Fong Foo v. United States, 369 U. S. 141 (1962). For however egregious the error of the acquittal, the termination favorable to the accused has been regarded as no different from a factfinder’s acquittal that resulted from errors of the trial judge. See also Burks v. United States, ante, p. 1. These cases teach that the Government’s means of protecting its vital interest in convicting the guilty is its participation as an adversary at the criminal trial where it has every opportunity to dissuade the trial court from committing erroneous rulings favorable to the accused. Jenkins was simply a necessary and logical extension of the rule that an acquittal bars any further trial proceedings. Jenkins recognized that an acquittal can never represent a determination that the criminal defendant is innocent in any absolute sense; the bar to a retrial following acquittal does not—and indeed could not—rest on any assumption that the finder of fact has applied the correct legal principles to all the admissible evidence and determined that the defendant was factually innocent of the offense charged. The reason further prosecution is barred following an acquittal, rather, is that the Government has been afforded one complete opportunity to prove a case of the criminal defendant’s culpability and, when it has failed for any reason to persuade the court not to enter a final judgment favorable to the accused, the constitutional policies underlying the ban against multiple trials become compelling. Thus, Jenkins and Lee recognized that it mattered not whether the final judgment constituted a formal “acquittal.” What is critical is whether the accused obtained, after jeopardy attached, a favorable termination of the charges against him. If he did, no matter how erroneous the ruling, the policies embodied in the Double Jeopardy Clause require the conclusion that “further proceedings . . . devoted to the resolution of factual issues going to the elements of the offense charged” are barred. Jenkins, 420 IT. S., at 370; see Lee, 432 U. S., at 30. 108 OCTOBER TERM, 1977 Brennan, J., dissenting 437 U. S. B The whole premise for today’s retreat from Jenkins and Lee, oi course, is the Court’s new theory that a criminal defendant who seeks to avoid conviction on a “ground unrelated to factual innocence” somehow stands on a different constitutional footing from a defendant whose participation in his criminal trial creates a situation in which a judgment of acquittal has to be entered. This premise is simply untenable. The rule prohibiting retrials following acquittals does not and could not rest on a conclusion that the accused was factually innocent in any meaningful sense. If that were the basis for the rule, the decisions that have held that even egregiously erroneous acquittals preclude retrials, see, e. g., Fong Foo v. United States, supra (acquittal entered after three of many prosecution witnesses had testified); Sanabria n. United States, ante, p. 54, were erroneous. It is manifest that the reasons that bar a retrial following an acquittal are equally applicable to a final judgment entered on a ground “unrelated to factual innocence.” The heavy personal strain of the second trial is the same in either case. So too is the risk that, though innocent, the defendant may be found guilty at a second trial. If the appeal is allowed in either situation, the Government will, following any reversal, not only obtain the benefit of the favorable appellate ruling but also be permitted to shore up any other weak points of its case and obtain all the other advantages at the second trial that the Double Jeopardy Clause was designed to forbid. Moreover, the Government’s interest in retrying a defendant simply cannot vary depending on the ground of the final termination in the accused’s favor. I reject as plainly erroneous the Court’s suggestion that final judgments not based on innocence deprive the public of “its valued right to ‘one complete opportunity to convict those who have violated its laws,’ ” ante, at 100, quoting Arizona n. Washington, 434 UNITED STATES v. SCOTT 109 82 Brennan, J., dissenting U. S., at 509,6 and therefore differ from “true acquittals.” The Government has the same “complete opportunity” in either situation by virtue of its participation as an adversary at the criminal trial.7 Equally significant, the distinction between the two is at best purely formal. Many acquittals are the consequence of rulings of law made on the accused’s motion that are not related to the question of his factual guilt or innocence: e. g., a ruling on the law respecting the scope of the offense or excluding reliable evidence. Sanabria v. United States, ante, p. 54, illustrates the point. 6 Similarly unpersuasive is the Court’s suggestion that its holding is supported by the well-recognized rules that a criminal defendant may twice be tried for the same offense if he either successfully moved for a mistrial at the first trial, see Lee, supra; United States v. Dinitz, 424 U. S. 600 (1976), or succeeded in having a conviction set aside on a ground other than the insufficiency of the evidence. See United States v. Ball, 163 U. S. 662 (1896). What distinguishes these situations, of course, is that neither involved a final judgment entered for the accused, and that in both the Government could not be said to have had a complete opportunity to convict the accused. 7 The Court’s suggestion that intervening decisions have somehow undermined Jenkins simply will not wash. Although it is quite true that the author of the Court opinion has stated that he understood Jenkins to embrace a rule that any midtrial termination that is labeled a “dismissal” erects a double jeopardy bar, see ante, at 86 n. 2, quoting Lee, 432 U. S., at 36 (Rehnquist, J., concurring), no Court opinion has adopted the position that the label attached to a trial court’s ruling could be determinative. Indeed, since Serfass v. United States, 420 U. S 377, 392 (1975), which was decided the week after Jenkins, explicitly provides that labels are not to have such talismanic significance, the unanimous Court in Jenkins could scarcely have contemplated that it had announced such a mechanical formula. Thus, the Court’s suggestion, see ante, at 94, that Lee, which held that a termination that was labeled a “dismissal” did not erect a double jeopardy bar, could have undermined Jenkins is unpersuasive on its face. In Lee, we treated the dismissal as the equivalent of a mistrial because both the trial judge and the parties had so regarded it. See 432 U. S., at 29. 110 OCTOBER TERM, 1977 Brennan, J., dissenting 437 U. S. In Sanabria, the District Court, acting on the defendant’s motions, made a series of erroneous legal rulings which began with an erroneous construction of the indictment and culminated in the exclusion of most of the evidence of defendant’s guilt. The trial court then granted defendant’s motion for a judgment of acquittal on the ground that the remaining evidence was insufficient. Sanabria held that the midtrial termination of the prosecution erected an absolute bar to any further proceedings against the defendant, and we reached that result even though the rulings which led to the acquittal were purely legal determinations, unrelated to any question of defendant’s factual guilt, and had been precipitated entirely by the defendant’s “voluntary choice” to seek a narrow construction of his indictment. Here the legal ruling that the Court characterizes as unrelated to the defendant’s factual guilt itself terminated the prosecution with prejudice. In Sanabria, after the District Court rendered the two erroneous rulings that excluded most of the relevant evidence of defendant’s guilt, it remained for the trial court to take the pro forma step of granting the defendant’s motion for a judgment of acquittal. Surely, this difference between the cases should not possess constitutional significance. By holding that it does, the Court suggests that the present case would have been decided differently if the trial court had remedied the due process violation by excluding all the Government’s evidence on count one and then entering an acquittal pursuant to Rule 29. Sanabria simply confirms that the distinction the Court today draws is wholly arbitrary, bearing no conceivable relationship to the policies protected by the Double Jeopardy Clause. II The Court’s definition of “acquittal” compounds the damage that repudiation of Jenkins and Lee has done to the fabric of double jeopardy law. Not only is this definition unduly UNITED STATES v. SCOTT 111 82 Brennan, J., dissenting restrictive, it is literally incapable of principled application. The Court’s application of its definition to the facts of this case proves the point. The doctrine of preindictment delay, like a host of other principles and policies of the law—e. g., entrapment, insanity, right to speedy trial, statute of limitations—operates to preclude the imposition of criminal liability on defendants, notwithstanding a showing that they committed criminal acts. Like these other doctrines, the question whether preindictment delay violates due process of law cannot ordinarily be considered apart from the factual development at trial since normally only the “ ‘[e] vents of the trial [can demonstrate] actual prejudice.’ ” United States n. Lovasco, 431 U. S. 783, 789 (1977), quoting United States v. Marion, 404 U. S. 307, 326 (1971); see United States v. MacDonald, 435 U. S. 850, 858, 858-859 (1978). Here, therefore, the District Court, quite properly, deferred consideration of the respondent’s pretrial motion to dismiss for preaccusation delay until trial. At the close of the evidence, respondent renewed his motion. The District Court recognized that there was sufficient evidence of guilt to permit submission of count one to the jury, but granted the motion as to this count because, evaluating the facts adduced at trial, the court found that the delay between the offense alleged and respondent’s indictment had been unjustifiable and had so prejudiced respondent’s ability to present his defense as to constitute a denial of due process of law. A critical feature of today’s holding appears to be the Court’s definition of acquittal as “ ‘a resolution [in the defendant’s favor], correct or not, of some or all of the factual elements of the offense charged,’ ” ante, at 97, quoting United States v. Martin Linen Supply Co., 430 U. S., at 571. But this definition, which is narrower than the traditional one, enjoys no significant support in our prior decisions. The language quoted from Martin Linen Supply Co. was tied to the par 112 OCTOBER TERM, 1977 Brennan, J., dissenting 437U.S. ticular issue in that case and was never intended to serve as an all-encompassing definition of acquittal for all purposes. Rather, Martin Linen Supply referred generally to “acquittal” as “a legal determination on the basis of facts adduced at the trial relating to the general issue of the case,” id., at 575 (citations omitted), and this is the accepted definition. See Serfass v. United States, 420 U. S. 377, 393 (1975), quoting United States v. Sisson, 399 U. S. 267, 290 n. 19 (1970). This definition, moreover, clearly encompasses rulings pertaining to all “affirmative defenses” that depend on the factual development at trial. The traditional definition of “acquittal” obviously is responsive to the values protected by the Double Jeopardy Clause. While it perhaps might not be objectionable to permit retrial of a defendant whose first trial was terminated on the basis of a midtrial ruling on a motion that could— because it did not depend upon the facts adduced at trial—have been raised before jeopardy attached, see Serfass v. United States, supra, at 394,8 it would be intolerable to permit the retrial of a defendant whose first prosecution ended on the basis of a ruling—like the one in the present case—which could only be made after the factual development at trial. Notably, the Court neither explains why it chooses to reject the more traditional definition of “acquittal” nor attempts to justify its more restrictive definition in terms of the constitutional policy against multiple trials. But I will not dwell further on this point. As the Court opinion itself demonstrates, what is perhaps as important as the actual definition is how it is applied. The pertinent question, thus, is one the Court never addresses: Why, for pur 8 In Serfass, we reserved decision on the question whether a defendant who was afforded an opportunity to obtain a determination of a legal defense prior to trial but who nevertheless knowingly allowed himself to be placed in jeopardy before raising the defense could claim the protections of the Double Jeopardy Clause. 420 U. S., at 394. UNITED STATES v. SCOTT 113 82 Brennan, J., dissenting poses of its new definition of “acquittal,” is not the fact vel non of preindictment delay one of the “factual elements of the offense charged”? The Court plainly cannot answer that preindictment delay is not referred to in the statutory definition of the offense charged in count one, cf. Patterson n. New York, 432 U. S. 197 (1977), for it states that dismissals based on the defenses of insanity9 and entrapment—neither of which is bound up with the statutory definition of federal crimes—will constitute “acquittals.” Ante, at 97-98. How can decisions based on the trial evidence that a defendant is “not guilty by reason of insanity” or “not guilty by reason of entrapment” erect a double jeopardy bar, and a decision—equally based on evaluation of the trial evidence— that the defendant is “not guilty by reason of preaccusation delay” not also prohibit further prosecution? None of these defenses is bound up in the definition of a crime, and the availability of each depends on the factual development at trial. More fundamentally, to permit a retrial following an appellate court’s reversal of a judgment entered on any of these grounds presents all the evils the Double Jeopardy Clause was designed to prevent. The Court offers no satisfactory explanation for the difference in treatment. The suggestion that determinations concerning insanity and entrapment are “factual” whereas dismissals of indictments for preindictment delay represent “legal judgments,” see ante, 9 A contrary position would not only be inconsistent with Burks v. United States, ante, p. 1, but would also have untoward consequences for criminal defendants. The premise of such a ruling would necessarily be that a criminal defendant has no legitimate interest in protecting the finality of a verdict of not guilty by reason of insanity. It would then follow that there could be appellate review not only of all directed verdicts of not guilty by reason of insanity, but also of all jury verdicts that had been preceded by a prior finding of guilt of the statutory offense. The implications of such a holding would be particularly significant in jurisdictions providing for bifurcated determinations of guilt and sanity. 114 OCTOBER TERM, 1977 Brennan, J., dissenting 437 U. S. at 98, is simply untenable. Consideration of all three defenses requires the application of legal standards to the evidence adduced at trial, and the most likely ground for reversal and reprosecution following the entry of a final judgment favorable to the accused on such grounds would be an appellate court’s conclusion that the trial court applied an erroneous legal test. The question the Court fails to address, therefore, is why an egregiously erroneous dismissal on entrapment grounds—e. g., a ruling in a federal trial that a defendant has been entrapped as a matter of law because it had been shown that the Government had supplied the contraband the defendant had been charged with selling, cf. Hampton v. United States, 425 U. S. 484 (1976)—should erect a double jeopardy bar but not a possibly erroneous dismissal on the ground of preaccusation delay. The Court’s observation that factual defenses of insanity and entrapment provide “legal justifications for otherwise criminal acts”—and is unlike the doctrine of preindictment delay, which is intended to protect the integrity of the trial process—reflects common legal parlance but in no wise explains why the two classes of dismissals should have different double jeopardy consequences. Whether or not the Court’s ipse dixit concerning the consequences of a ruling of unlawful preaccusation delay is defensible, the enormous practical problems that today’s decision portends are very clear. A particularly appealing virtue of the Jenkins and Lee principle—in addition, of course, to its protection of constitutional values—was its simplicity. Any midtrial order contemplating an end to all prosecution of the accused would automatically erect a double jeopardy bar to a retrial. Under today’s decision, the thousands of state and federal courts will be required to decide, with only minimal guidance from this Court, the question of the double jeopardy consequences of all favorable terminations of criminal proceedings on the basis of affirmative defenses. The only guidance the Court offers is its suggestion that defenses which UNITED STATES v. SCOTT 115 82 Brennan, J., dissenting provide legal justifications for otherwise criminal acts will erect double jeopardy bars whereas those defenses that arise from unlawful or unconstitutional Government acts will not. Consideration of the defense of entrapment illustrates how difficult the Court’s decision will be to apply. To the extent the defense applies when there has been a showing the defendant was not “predisposed” to commit a criminal act, it perhaps does provide a “legal justification.” But the defense of entrapment, in many jurisdictions, see Park, The Entrapment Controversy, 60 Minn. L. Rev. 163, 171-176 (1976), is a device to deter police officials from engaging in reprehensible law enforcement techniques. Is the entrapment defense to erect a double jeopardy bar in such jurisdictions? Are the double jeopardy consequences to depend upon the appellate court’s characterization of the operation of the defense in the particular case before it? And what of other traditional factual defenses, which are routinely submitted to the jury and which could be the basis for Rule 29 motions: e. g., the statute of limitations?10 Ironically, it seems likely that, when all is said and done, there will be few instances indeed in which defenses can be deemed unrelated to factual innocence. If so, today’s decision may be limited to disfavored doctrines like preaccusation delay. See generally United States v. Lovasco, 431 U. S. 783 (1977). It is regrettable that the Court should introduce such confusion in an area of the law that, until today, had been crystal clear. Its introduction might be tolerable if necessary to advance some important policy or to serve values protected by 10 In any case in which the date upon which the defendant committed the crime is disputed and may have been outside the statute of limitations provided by law, a trial judge could, and probably would, submit this question to the jury along with the general issue. Similarly, in any case in which the evidence adduced at trial revealed that the defendant had committed the criminal act outside the limitation period, the defendant would move for a “directed verdict.” 116 OCTOBER TERM, 1977 Brennan, J., dissenting 437 U. S. the Double Jeopardy Clause, but that manifestly is not the case. Rather, today’s decision fashions an entirely arbitrary distinction that creates precisely the evils that the Double Jeopardy Clause was designed to prevent. I would affirm the judgment of the Court of Appeals. EXXON CORP. v. GOVERNOR OF MARYLAND 117 Syllabus EXXON CORP, et al. v. GOVERNOR OF MARYLAND ET AL. APPEAL FROM THE COURT OF APPEALS OF MARYLAND No. 77-10. Argued February 28, 1978—Decided June 14, 1978* Responding to evidence that during the 1973 petroleum shortage oil producers or refiners were favoring company-operated gasoline stations, Maryland enacted a statute prohibiting producers or refiners from operating retail service stations within the State, and requiring them to extend all “voluntary allowances” (temporary price reductions granted to independent dealers injured by local competitive price reductions) uniformly to all stations they supply. In actions by several oil companies challenging the validity of the statute on various grounds, the Maryland trial court held the statute invalid primarily on substantive due process grounds, but the Maryland Court of Appeals reversed, upholding the validity of the statute against contentions, inter alia, that it violated the Commerce and Due Process Clauses and conflicted with § 2 (b) of the Clayton Act, as amended by the Robinson-Patman Act, which prohibits price discrimination, with the proviso that a seller can defend a price discrimination charge by showing that he charged a lower price in good faith to meet a competitor’s equally low price. Held: 1. The Maryland statute does not violate the Due Process Clause, since, regardless of the ultimate efficacy of the statute, it bears a reasonable relation to the State’s legitimate purpose in controlling the gasoline retail market. Pp. 124-125. 2. The divestiture provisions of the statute do not violate the Commerce Clause. Pp. 125-129. (a) That the burden of such provisions falls solely on interstate companies does not, by itself, establish a claim of discrimination against interstate commerce. The statute creates no barrier against interstate independent dealers, nor does it prohibit the flow of interstate goods, place added costs upon them, or distinguish between in-state and out-of-state companies in the retail market. Hunt v. Washington Apple *Together with No. 77-11, Shell Oil Co. v. Governor of Maryland et al.; No. 77-12, Continental Oil Co. et al. n. Governor of Maryland et al.; No. 77-47, Gulf Oil Corp. v. Governor of Maryland et al.; and No. 77-64, Ashland Oil, Inc., et al. v. Governor of Maryland et al., also on appeal from the same court. 118 OCTOBER TERM, 1977 Syllabus 437 U. S. Advertising Common, 432 U. S. 333; and Dean Milk Co. v. Madison, 340 U. S. 349, distinguished. Pp. 125-126. (b) Nor does the fact that the burden of state regulation falls on interstate companies show that the statute impermissibly burdens interstate commerce, even if some refiners were to stop selling in the State because of the divestiture requirement and even if the elimination of company-operated stations were to deprive consumers of certain special services. Interstate commerce is not subjected to an impermissible burden simply because an otherwise valid regulation causes some business to shift from one interstate supplier to another. The Commerce Clause protects the interstate market, not particular interstate firms, from prohibitive or burdensome regulations. Pp. 127-128. (c) The Commerce Clause does not, by its own force, pre-empt the field of retail gasoline marketing, but, absent a relevant congressional declaration of policy, or a showing of a specific discrimination against, or burdening of, interstate commerce, the States have the power to regulate in this area. Pp. 128-129. 3. The “voluntary allowances” requirement of the Maryland statute is not pre-empted by § 2 (b) of the Clayton Act, as amended by the Robinson-Patman Act, or the Sherman Act. Pp. 129-134. (a) Any hypothetical “conflict” arising from the possibility that the Maryland statute may require uniformity in some situations in which the Robinson-Patman Act wohld permit localized price discrimination is not sufficient to warrant pre-emption. Pp. 130-131. (b) Neither § 2 (b) nor the federal policy favoring competition establishes a federal right to engage in discriminatory pricing in certain situations. Section 2 (b)’s proviso is merely an exception to that statute’s broad prohibition against discriminatory pricing and does not create any new federal right, but rather defines a specific, limited defense. Pp. 131-133. (c) While in the sense that the Maryland statute might have an anticompetitive effect there is a conflict between that statute and the Sherman Act’s central policy of “economic liberty,” nevertheless this sort of conflict cannot by itself constitute a sufficient reason for invalidating the Maryland statute, for if an adverse effect on competition were, in and of itself, enough to invalidate a state statute, the States’ power to engage in economic regulation would be effectively destroyed. Pp. 133-134. 279 Md. 410, 370 A. 2d 1102 and 372 A. 2d 237, affirmed. Stevens, J., delivered the opinion of the Court, in which Burger, C. J., and Brennan, Stewart, White, Marshall, and Rehnquist, JJ., EXXON CORP. V. GOVERNOR OF MARYLAND 119 117 Opinion of the Court joined. Blackmun, J., filed an opinion concurring in part and dissenting in part, post, p. 134. Powell, J., took no part in the consideration or decision of the cases. William Simon argued the cause for appellants in all cases. With him on the briefs for appellants in Nos. 77-10, 77-11, and 77-47 were William L. Marbury, Lewis A. Noonberg, David F. Tufaro, Robert L. Stem, J. Edward Davis, Damiel T. Doherty, Jr., Robert G. Abrams, Lawrence S. Greenwald, Bernard J. Caillouet, Richard P. Delaney, Lauric J. Cusack, Jerry Miller, and A. M. Minotti. Wilbur D. Preston, Jr., Stanley B. Rohd, Andrew K. McColpin, and Richard R. Linn filed a brief for appellants in No. 77-12. David Ginsburg, Fred W. Drogula, and James E. Wesner filed briefs for appellants in No. 77-64. Francis B. Burch, Attorney General of Maryland, and Thomas M. Wilson III, Assistant Attorney General, argued the cause for respondents in all cases. With them on the brief were John F. Oster, Deputy Attorney General, and John A. Woodstock and Steven P. Resnick, Assistant Attorneys General.! Mr. Justice Stevens delivered the opinion of the Court. A Maryland statute provides that a producer or refiner of petroleum products (1) may not operate any retail service station within the State, and (2) must extend all “voluntary fBriefs of amici curiae urging reversal were filed by Eugene Gressman for Charter Oil Co. et al.; and by John S. McDaniel, Jr., and William J. Rubin for Crown Petroleum Corp. Jerry S. Cohen filed a brief for the National Congress of Petroleum Retailers as amicus curiae urging affirmance. Briefs of amici curiae were filed by Evelle J. Younger, Attorney General, Sanford N. Gruskin, Chief Assistant Attorney General, Warren J. Abbott, Assistant Attorney General, and Michael I. Spiegel and Linda L. Tedeschi, Deputy Attorneys General, for the State of California; by Erwin N. Griswold for Champlin Petroleum Co. et al.; and by George W. Liebmann, Robert B. Levin, and Robert G. Levy for Day Enterprises, Inc., et al. 120 OCTOBER TERM, 1977 Opinion of the Court 437 U. S. allowances” uniformly to all service stations it supplies.1 The questions presented are whether the statute violates either the Commerce or the Due Process Clause of the Constitution of the United States, or is directly or indirectly pre-empted by the congressional expression of policy favoring vigorous competition found in § 2 (b) of the Clayton Act, 38 Stat. 730, as amended by the Robinson-Patman Act, 49 Stat. 1526.2 The Court of Appeals of Maryland answered these questions in xThe pertinent provisions of the statute are as follows: “(b) After July 1, 1974, no producer or refiner of petroleum products shall open a major brand, secondary brand or unbranded retail service station in the State of Maryland, and operate it with company personnel, a subsidiary company, commissioned agent, or under a contract with any person, firm, or corporation, managing a service station on a fee arrangement with the producer or refiner. The station must be operated by a retail service station dealer. “(c) After July 1, 1975, no producer or refiner of petroleum products shall operate a major brand, secondary brand, or unbranded retail service station in the State of Maryland, with company personnel, a subsidiary company, commissioned agent, or under a contract with any person, firm, or corporation managing a service station on a fee arrangement with the producer or refiner. The station must be operated by a retail service station dealer. “(d) Every producer, refiner, or wholesaler of petroleum products supplying gasoline and special fuels to retail service station dealers shall extend all voluntary allowances uniformly to all retail service station dealers supplied.” Md. Code Ann., Art. 56, § 157E (Supp. 1977). 2 “Upon proof being made, at any hearing on a complaint under this section, that there has been discrimination in price or services or facilities furnished, the burden of rebutting the prima-facie case thus made by showing justification shall be upon the person charged with a violation of this section, and unless justification shall be affirmatively shown, the Commission is authorized to issue an order terminating the discrimination: Provided, however, That nothing herein contained shall prevent a seller rebutting the prima-facie case thus made by showing that his lower price or the furnishing of services or facilities to any purchaser or purchasers was made in good faith to meet an equally low price of a competitor, or the services or facilities furnished by a competitor.” 15 U. S. C. § 13 (b) (1976 ed.). EXXON CORP. V. GOVERNOR OF MARYLAND 121 117 Opinion of the Court favor of the validity of the statute. 279 Md. 410, 370 A. 2d 1102 and 372 A. 2d 237 (1977). We affirm. I The Maryland statute is an outgrowth of the 1973 shortage of petroleum. In response to complaints about inequitable distribution of gasoline among retail stations, the Governor of Maryland directed the State Comptroller to conduct a market survey. The results of that survey indicated that gasoline stations operated by producers or refiners had received preferential treatment during the period of short supply. The Comptroller therefore proposed legislation which, according to the Court of Appeals, was “designed to correct the inequities in the distribution and pricing of gasoline reflected by the survey.” Id., at 421, 370 A. 2d, at 1109. After legislative hearings and a “special veto hearing” before the Governor, the bill was enacted and signed into law. Shortly before the effective date of the Act, Exxon Corp, filed a declaratory judgment action challenging the statute in the Circuit Court of Anne Arundel County, Md. The essential facts alleged in the complaint are not in dispute. All of the gasoline sold by Exxon in Maryland is transported into the State from refineries located elsewhere. Although Exxon sells the bulk of this gas to wholesalers and independent retailers, it also sells directly to the consuming public through 36 company-operated stations.3 Exxon uses these stations to test innovative marketing concepts or products.4 Focusing primarily on the Act’s requirement that it discontinue its operation of these 36 retail stations, Exxon’s complaint challenged the 3 As used by the Court of Appeals and in this opinion, “company-operated station” refers to a retail service station operated directly by employees of a refiner or producer of petroleum products (or a subsidiary). 279 Md., at 419 n. 2,370 A. 2d, at 1108 n. 2. 4 For instance, Exxon has used its company-operated stations to introduce such marketing ideas as partial self-service, in-bay car-wash units, and motor-oil vending machines. App. 205-209. 122 OCTOBER TERM, 1977 Opinion of the Court 437 U. S. validity of the statute on both constitutional and federal statutory grounds.5 During the ensuing nine months, six other oil companies instituted comparable actions. Three of these plaintiffs, or their subsidiaries, sell their gasoline in Maryland exclusively through company-operated stations/5 These refiners, using trade names such as “Red Head” and “Scot,” concentrate largely on high-volume sales with prices consistently lower than those offered by independent dealer-operated major brand stations. Testimony presented by these refiners indicated that company ownership is essential to their method of private brand, low-priced competition. They therefore joined Exxon in its attack on the divestiture provisions of the Maryland statute. The three other plaintiffs, like Exxon, sell major brands primarily through dealer-operated stations, although they also operate at least one retail station each.7 They, too, challenged the statute’s divestiture provisions, but, in addition, they specially challenged the requirement that “voluntary allowances” be extended uniformly to all retail service stations supplied in the State. Although not defined in the statute, the term “voluntary allowances” refers to temporary price reductions granted by the oil companies to independent dealers who 5 Exxon presented nine arguments, both constitutional and statutory. It contended that the statute was arbitrary and irrational under the Due Process Clause; constituted an unconstitutional taking of property without just compensation; denied it, in two distinct ways, the equal protection of the laws; constituted an unlawful delegation of legislative authority; was unconstitutionally vague; discriminated against and burdened interstate commerce; and was pre-empted by the Robinson-Patman Act and the Federal Emergency Petroleum Allocation Act of 1973. Id., at 14-16. 6 These plaintiffs are Continental Oil Co. (and its subsidiary Kayo Oil Co.), Commonwealth Oil Refining Co. (and its subsidiary Petroleum Marketing Corp.), and Ashland Oil Co. 7 These plaintiffs are Phillips Petroleum Co., Shell Oil Co., and Gulf Oil Corp. EXXON CORP. V. GOVERNOR OE MARYLAND 123 117 Opinion of the Court are injured by local competitive price reductions of competing retailers.8 The oil companies regard these temporary allowances as legitimate price reductions protected by § 2 (b). In advance of trial, Exxon, Shell, and Gulf moved for a partial summary judgment declaring this portion of the Act invalid as in conflict with § 2 (b). The Circuit Court granted the motion, and the trial then focused on the validity of the divestiture provisions. As brought out during the trial, the salient characteristics of the Maryland retail gasoline market are as follows: Approximately 3,800 retail service stations in Maryland sell over 20 different brands of gasoline. However, no petroleum products are produced or refined in Maryland, and the number of stations actually operated by a refiner or an affiliate is relatively small, representing about 5% of the total number of Maryland retailers. The refiners introduced evidence indicating that their ownership of retail service stations has produced significant benefits for the consuming public.9 Moreover, the three refiners that now market solely through company-operated stations may elect to withdraw from the Maryland market altogether if the statute is enforced. There was, however, no evidence that the total quantity of petroleum products shipped into Maryland would be affected by the statute.10 After trial, the Circuit Court held the entire statute invalid, primarily on substantive due process grounds. The Maryland Court of Appeals reversed, rejecting all of the refiners’ attacks against both the divestiture provisions and 8 See 279 Md., at 445-446, 370 A. 2d, at 1121-1122. 9 Id., at 418-420, 370 A. 2d, at 1107-1108. 10 The Court of Appeals stated that the statute “would not in any way restrict the free flow of petroleum products into or out of the state.” Id., at 431, 370 A. 2d, at 1114. While the evidence in the record does not directly support this assertion, it is certainly a permissible inference to be drawn from the evidence, or lack thereof, presented by the appellants. See Reply Brief for Appellants in No. 77-64, p. 7. 124 OCTOBER TERM, 1977 Opinion of the Court 437 U. S. the voluntary-allowance provision. Most of those attacks are not pursued here;11 instead, appellants have focused their appeals on the claims that the Maryland statute violates the Due Process and Commerce Clauses and that it is in conflict with the Robinson-Patman Act. II Appellants’ substantive due process argument requires little discussion.12 The evidence presented by the refiners may cast soipe doubt on the wisdom of the statute, but it is, by now, absolutely clear that the Due Process Clause does not empower the. judiciary “to sit as a ‘superlegislature to weigh the wisdom of legislation’ . . . .” Ferguson v. Skrupa, 372 U. S. 726, 731 (citation omitted). Responding to evidence that producers and refiners were favoring company-operated stations in the allocation of gasoline and that this would eventually decrease the competitiveness of the retail market, the State enacted a law prohibiting producers and refiners from operating their own stations. Appellants argue that this response is irrational and that it will frustrate rather than further the State’s desired goal of enhancing competition. But, as the Court of Appeals observed, this argument rests simply on an evaluation of the economic wisdom of the statute, 279 Md., at 428, 370 A. 2d, at 1112, and cannot override the State’s authority “to legislate against what are found to be injurious practices in their internal commercial and business affairs ....” Lincoln Federal Labor Union n. Northwestern Iron & Metal Co., 335 U. S. 525, 536.13 Regardless of the ultimate economic 11 See n. 5, supra. 12 Indeed, although the Circuit Court’s decision rested primarily on the substantive due process claim, only appellants Continental Oil and its subsidiary, Kayo Oil, press that claim here. 13 It is worth noting that divestiture is by no means a novel method of economic regulation, and is found in both federal and state statutes. To date, the courts have had little difficulty sustaining suoh statutes against a substantive due process attack. See, e. g., Paramount Pictures, EXXON CORP. V. GOVERNOR OF MARYLAND 125 117 Opinion of the Court efficacy of the statute, we have no hesitancy in concluding that it bears a reasonable relation to the State’s legitimate purpose in controlling the gasoline retail market, and we therefore reject appellants’ due process claim. Ill Appellants argue that the divestiture provisions of the Maryland statute violate the Commerce Clause in three ways: (1) by discriminating against interstate commerce; (2) by unduly burdening interstate commerce; and (3) by imposing controls on a commercial activity of such an essentially interstate character that it is not amenable to state regulation. Plainly, the Maryland statute does not discriminate against interstate goods, nor does it favor local producers and refiners. Since Maryland’s entire gasoline supply flows in interstate commerce and since there are no local producers or refiners, such claims of disparate treatment between interstate and local commerce would be meritless. Appellants, however, focus on the retail market, arguing that the effect of the statute is to protect in-state independent dealers from out-of-state competition. They contend that the divestiture provisions “create a protected enclave for Maryland independent dealers .. . .”14 As support for this proposition, they rely on the fact that the burden of the divestiture requirements falls solely on interstate companies. But this fact does not lead, either logically or as a practical matter, to a conclusion that the State is discriminating against interstate commerce at the retail level. As the record shows, there are several major interstate marketers of petroleum that own and operate their own retail Inc. v. Langer, 23 F. Supp. 890 (ND 1938), dismissed as moot, 306 U. S. 619; see generally Comment, Gasoline Marketing Practices and “Meeting Competition” under the Robinson-Patman Act, 37 Md. L. Rev. 323, 329 n. 44 (1977). 14 Brief for Appellants in No. 77-10, p. 27. 126 OCTOBER TERM, 1977 Opinion of the Court 437U.S. gasoline stations.15 These interstate dealers, who compete directly with the Maryland independent dealers, are not affected by the Act because they do not refine or produce gasoline. In fact, the Act creates no barriers whatsoever against interstate independent dealers; it does not prohibit the flow of interstate goods, place added costs upon them, or distinguish between in-state and out-of-state companies in the retail market. The absence of any of these factors fully distinguishes this case from those in which a State has been found to have discriminated against interstate commerce. See, e. g., Hunt v. Washington Apple Advertising Common, 432 U. S. 333; Dean Milk Co. v. Madison, 340 U. S. 349. For instance, the Court in Hunt noted that the challenged state statute raised the cost of doing business for out-of-state dealers, and, in various other ways, favored the in-state dealer in the local market. 432 U. S., at 351-352. No comparable claim can be made here. While the refiners will no longer enjoy their same status in the Maryland market, in-state independent dealers will have no competitive advantage over out-of-state dealers. The fact that the burden of a state regulation falls on some interstate companies does not, by itself, establish a claim of discrimination against interstate commerce.16 15 For instance, as of July 1, 1974, such interstate, nonrefining or nonproducing, companies as Sears, Roebuck & Co., Hudson Oil Co., and Pantry Pride operated retail gas stations in Maryland. App. 190-191. Hudson has, however, recently acquired a refinery. See Brief for Appellants in No. 77-10, p. 33 n. 17. 16 If the effect of a state regulation is to cause local goods to constitute a larger share, and goods with an out-of-state source to constitute a smaller share, of the total sales in the market—as in Hunt, 432 U. S., at 347, and Dean Milk, 340 U. S., at 354—the regulation may have a discriminatory effect on interstate commerce. But the Maryland statute has no impact on the relative proportions of local and out-of-state goods sold in Maryland and, indeed, no demonstrable effect whatsoever on the interstate flow of goods. The sales by independent retailers are just as much a part of EXXON CORP. V. GOVERNOR OF MARYLAND 127 117 Opinion of the Court Appellants argue, however, that this fact does show that the Maryland statute impermissibly burdens interstate commerce. They point to evidence in the record which indicates that, because of the divestiture requirements, at least three refiners will stop selling in Maryland, and which also supports their claim that the elimination of company-operated stations will deprive the consumer of certain special services. Even if we assume the truth of both assertions, neither warrants a finding that the statute impermissibly burdens interstate commerce. Some refiners may choose to withdraw entirely from the Maryland market, but there is no reason to assume that their share of the entire supply will not be promptly replaced by other interstate refiners. The source of the consumers’ supply may switch from company-operated stations to independent dealers, but interstate commerce is not subjected to an impermissible burden simply because an otherwise valid regulation causes some business to shift from one interstate supplier to another. The crux of appellants’ claim is that, regardless of whether the State has interfered with the movement of goods in interstate commerce, it has interfered “with the natural functioning of the interstate market either through prohibition or through burdensome regulation.” Hughes v. Alexandria Scrap Corp., 426 U. S. 794, 806. Appellants then claim that the statute “will surely change the market structure by weakening the independent refiners . . . .”17 We cannot, however, accept appellants’ underlying notion that the Commerce Clause protects the particular structure or methods of operation in a retail market. See Breard v. Alexandria, 341 U. S. 622. As indicated by the Court in Hughes, the Clause protects the interstate market, not particular interstate firms, from prohib the flow of interstate commerce as the sales made by the refiner-operated stations. 17 Reply Brief for Appellants in No. 77-64, p. 7. 128 OCTOBER TERM, 1977 Opinion of the Court 437 U. S. itive or burdensome regulations. It may be true that the consuming public will be injured by the loss of the high-volume, low-priced stations operated by the independent refiners, but again that argument relates to the wisdom of the statute, not to its burden on commerce. Finally, we cannot adopt appellants’ novel suggestion that because the economic market for petroleum products is nationwide, no State has the power to regulate the retail marketing of gas. Appellants point out that many state legislatures have either enacted or considered proposals similar to Maryland’s,18 and that the cumulative effect of this sort of legislation may have serious implications for their national marketing operations. While this concern is a significant one, we do not find that the Commerce Clause, by its own force, pre-empts the field of retail gas marketing. To be sure, “the Commerce Clause acts as a limitation upon state power even without congressional implementation.” Hunt v. Washington Apple Advertising Comm’n, supra, at 350. But this Court has only rarely held that the Commerce Clause itself pre-empts an entire field from state regulation, and then only when a lack of national uniformity would impede the flow of interstate goods. See Wabash, St. L. Ac P. R. Co. v. Illinois, 118 U. S. 557; see also Cooley v. Board of Wardens, 12 How. 299, 319. The evil that appellants perceive in this litigation is not that the several States will enact differing regulations, but rather that they will all conclude that divestiture provisions are warranted. The problem thus is not one of national uniformity. In the absence of a relevant congressional declaration of policy, or a showing of a specific discrimination against, or burdening 18 California, Delaware, the District of Columbia, and Florida have adopted laws restricting refiners’ operation of service stations. Similar proposals have been before the legislatures of 32 other jurisdictions. See Brief for Appellants in No. 77-10, p. 45 nn. 21 and 22; Brief for the State of California as Amicus Curiae. EXXON CORP. v. GOVERNOR OF MARYLAND 129 117 Opinion of the Court of, interstate commerce, we cannot conclude that the States are without power to regulate in this area. IV Exxon, Phillips, Shell, and Gulf contend that the requirement that voluntary allowances be extended to all retail service stations is either in direct conflict with § 2 (b) of the Clayton Act, as amended by the Robinson-Patman Act, or, more generally, in conflict with the basic federal policy in favor of competition, which is reflected in the Sherman Act as well as § 2 (b). In rejecting these contentions, the Maryland Court of Appeals noted that the Maryland statute covered two different competitive situations.19 In the first situation a competing retailer lowers its price on its own, and the oil company gives its own retailer a price reduction to enable it to meet that lower price. In the second situation, the competing retailer’s lower price is subsidized by its supplier, and the oil company gives its own retailer a price reduction to meet the competition. The good-faith defense of § 2 (b) is clearly not available to the oil company in the first situation because the voluntary allowance would not be a response to competition from another oil company. See FTC v. Sun Oil Co., 371 U .8. 505. In the second situation the law is unsettled,20 but the 19 The Court of Appeals also noted that there is a third competitive situation—a discriminatory price reduction made to meet an equally low price offered to the same buyer by a competing seller. In the lower court’s view, this situation clearly fell within the § 2 (b) defense, but was not encompassed by the term “voluntary allowances.” 279 Md., at 452, 370 A. 2d, at 1125. 20 The Court left the question open in Sun OU, 371 U. 8., at 512 n. 7, and the lower courts have reached conflicting results. Compare Enterprise Industries v. Texas Co., 136 F. Supp. 420 (Conn. 1955), rev’d on other grounds, 240 F. 2d 457 (CA2 1957), cert, denied, 353 U. S. 965, with Bargain Car Wash, Inc. n. Standard OU Co. (Indiana), 466 F. 2d 1163 (CA7 1972). 130 OCTOBER TERM, 1977 Opinion of the Court 437 U. S. Court of Appeals concluded that the defense would also be unavailable. The court therefore reasoned that there was no conflict between the Maryland statute and § 2 (b), since the statute did not apply to any allowance protected by federal law. In our opinion, it is not necessary to decide whether the § 2 (b) defense would apply in the second situation, for even assuming that it does, there is no conflict between the Maryland statute and the Robinson-Patman Act sufficient to require pre-emption. Appellants’ first argument is that compliance with the Maryland statute may cause them to violate the Robinson-Patman Act. They stress the possibility that the requirement that a price reduction be made on a statewide basis may result in discrimination between customers who would otherwise receive the same price, and they describe various hypothetical situations to illustrate this point.21 But, “[i]n this as in other areas of coincident federal and state regulation, the ‘teaching of this Court’s decisions . . . enjoin [s] seeking out conflicts between state and federal regulation where none clearly exists.’ Huron Cement Co. v. Detroit, 362 U. S. 440,446.” Seagram & Sons, Inc. v. Hostetter, 384 U. S. 35, 45. See also State v. Texaco, Inc., 14 Wis. 2d 625, 111 N. W. 2d 918 (1961). The Court in Seagram & Sons went on to say that “[a]lthough it is possible to envision circumstances under which price dis- 21 Appellants argue that compliance with the “voluntary allowance” provision may expose them to both primary-line and secondary-line liability under § 2 (a) of the Clayton Act, as amended by the Robinson-Patman Act. With respect to primary-line liability, they pose the hypothesis of a seller who responds to a competitor’s lower price in Baltimore. Under the statute, he must lower his prices throughout the State, even though the competitive market justifying that price is confined to Baltimore. Appellants then argue that a competitor operating only in Salisbury, Md., may be injured by this price reduction. But an injury flowing from a uniform price reduction is not actionable under the Robinson-Patman Act, which only prohibits price discrimination. See F. Rowe, Price Discrimination Under the Robinson-Patman Act 93 (1962). EXXON CORP. v. GOVERNOR OF MARYLAND 131 117 Opinion of the Court criminations proscribed by the Robinson-Patman Act might be compelled by [the state statute], the existence of such potential conflicts is entirely too speculative in the present posture of this case” to warrant pre-emption. 384 U. S., at 46. That counsel of restraint applies with even greater force here. For even if we were to delve into the hypothetical situations posed by appellants, we would not be presented with a state statute that requires a violation of the Robinson-Patman Act. Instead, the alleged “conflict” here is in the possibility that the Maryland statute may require uniformity in some situations in which the Robinson-Patman Act would permit localized discrimination.22 This sort of hypothetical conflict is not sufficient to warrant pre-emption. Appellants, however, also claim that the Robinson-Patman Act does not simply permit localized discrimination, but actually establishes a federal right to engage in discriminatory pricing in certain situations. They argue that this federal right may be found directly in § 2 (b), or, more generally, in our Nation’s basic policy favoring competition as reflected in the Sherman Act as well as § 2 (b). We find neither argument persuasive. The proviso in § 2 (b) of the Clayton Act, as amended by 22 Thus, appellants’ claim that the statute will create secondary-line Eability is premised on the possibility that price differentials may arise between stations located in Maryland and those in neighboring States. With respect to this claim, it is sufficient to note that, although the Maryland statute may affect the business decision of whether or not to reduce prices, it does not create any irreconcilable conflict with the Robinson-Patman Act. The statute may require that a voluntary allowance that could legally have been confined to the Baltimore area be extended to Salisbury. We may then assume, arguendo, that the Robinson-Patman Act could require a further extension of the allowance into the neighboring State. The possible scope of the voluntary allowance may, therefore, have an impact on the company’s decision on whether or not to meet the competition in Baltimore, but the state statute does not in any way require discriminatory prices. See also n. 20, supra. 132 OCTOBER TERM, 1977 Opinion of the Court 437 U. S. the Robinson-Patman Act, is merely an exception to that statute’s broad prohibition against discriminatory pricing. It created no new federal right; quite the contrary, it defined a specific, limited defense, and even narrowed the good-faith defense that had previously existed.23 To be sure, the defense is an important one, and the interpretation of its contours has been informed by the underlying national policy favoring competition which it reflects.24 But it is illogical to infer that by excluding certain competitive behavior from the general ban against discriminatory pricing, Congress intended to preempt the States’ power to prohibit any conduct within that exclusion. This Court is generally reluctant to infer preemption, see, e. g., De Canas v. Bica, 424 U. S. 351, 357-358, n. 5; Merrill Lynch, Pierce, Fenner & Smith v. Ware, 414 U. S. 117, 127, and it would be particularly inappropriate to do so in this case because the basic purposes of the state statute and the Robinson-Patman Act are similar. Both reflect a policy choice favoring the interest in equal treatment of all customers 23 Section 2 of the original Clayton Act, 38 Stat. 730, established an absolute defense for a seller’s reductions in price made “in good faith to meet competition . . . .” The legislative history of the Robinson-Patman Act shows that § 2 (b) was intended to limit that broad defense. See Standard Oil Co. v. FTC, 340 U. S. 231, 247-249, n. 14. 24 In holding that § 2 (b) created a substantive, rather than merely a procedural, defense, the Court explained: “The heart of our national economic policy long has been faith in the value of competition. In the Sherman and Clayton Acts, as well as in the Robinson-Patman Act, ‘Congress was dealing with competition, which it sought to protect, and monopoly, which it sought to prevent.’ Staley Mfg. Co. v. Federal Trade Comm’n, 135 F. 2d 453, 455. We need not now reconcile, in its entirety, the economic theory which underlies the Robinson-Patman Act with that of the Sherman and Clayton Acts. It is enough to say that Congress did not seek by the Robinson-Patman Act either to abolish competition or so radically to curtail it that a seller would have no substantial right of self-defense against a price raid by a competitor.” Standard Oil Co., supra, at 248-249 (footnote omitted). EXXON CORP. v. GOVERNOR OF MARYLAND 133 117 Opinion of the Court over the interest in allowing sellers freedom to make selective competitive decisions.25 Appellants point out that the Robinson-Patman Act itself may be characterized as an exception to, or a qualification of, the more basic national policy favoring free competition,26 and argue that the Maryland statute “undermin [es]” the competitive balance that Congress struck between the Robinson-Patman and Sherman Acts.27 This is merely another way of stating that the Maryland statute will have an anticompetitive effect. In this sense, there is a conflict between the statute and the central policy of the Sherman Act—our “charter of economic liberty.” Northern Pacific R. Co. v. United States, 356 U. S. 1, 4. Nevertheless, this sort of conflict cannot itself constitute a sufficient reason for invalidating the Maryland statute. For if an adverse effect on competition were, in and of itself, enough to render a state statute invalid, the States’ power to engage in economic regulation would be effectively destroyed.28 We are, therefore, satisfied that neither the broad implications of the Sherman Act nor the Robinson-Patman Act can fairly 25 Just as the political and economic stimulus for the Robinson-Patman Act was the perceived need to protect independent retail stores from “chain stores,” see U. S. Department of Justice, Report on the Robinson-Patman Act 114—124 (1977), so too the Maryland statute was prompted by the perceived need to protect independent retail service station dealers from the vertically integrated oil companies. 279 Md., at 422, 370 A. 2d, at 1109. 26 Indeed, many have argued that the Robinson-Patman Act is fundamentally anticompetitive and undermines the purposes of the Sherman Act. See generally U. S. Department of Justice Report, supra. 27 Brief for Appellants in No. 77-10, p. 80. 28 Appellants argue that Maryland has actually regulated beyond its boundaries, pointing to the possibility that they may have to extend voluntary allowances into neighboring States in order to avoid liability under the Robinson-Patman Act. See nn. 21 and 22, supra. But this alleged extra-territorial effect arises from the Robinson-Patman Act, not the Maryland statute. 134 437 U.S. OCTOBER TERM, 1977 Opinion of Blackmun, J. be construed as a congressional decision to pre-empt the power of the Maryland Legislature to enact this law. The judgment is affirmed. So ordered. Mr. Justice Powell took no part in the consideration or decision of these cases. Mr. Justice Blackmun, concurring in part and dissenting in part. Although I agree that the Maryland Motor Fuel Inspection Law1 does not offend substantive due process or federal anti- 1 The presently challenged portions of the law were enacted four years ago and amended once since then. 1974 Md. Laws, ch. 854; 1975 Md. Laws, ch. 608. The statute is now codified as Md. Code Ann., Art. 56, § 157E (Supp. 1977), and reads: “(a) For the purpose of this law all gasoline and special fuels sold or offered or exposed for sale shall be subject to inspection and analysis as hereinafter provided. . . . “(b) After July 1, 1974, no producer or refiner of petroleum products shall open a major brand, secondary brand or unbranded retail service station in the State of Maryland, and operate it with company personnel, a subsidiary company, commissioned agent, or under a contract with any person, firm, or corporation, managing a service station on a fee arrangement with the producer or refiner. The station must be operated by a retail service station dealer. “(c) After July 1, 1975, no producer or refiner of petroleum products shall operate a major brand, secondary brand, or unbranded retail service station in the State of Maryland, with company personnel, a subsidiary company, commissioned agent, or under a contract with any person, firm, or corporation managing a service station on a fee arrangement with the producer or refiner. The station must be operated by a retail service station dealer. “(d) Every producer, refiner, or wholesaler of petroleum products supplying gasoline and special fuels to retail service station dealers shall extend all voluntary allowances uniformly to all retail service station dealers supplied. “(e) Every producer, refiner, or wholesaler of petroleum products supplying gasoline and special fuels to retail service station dealers shall EXXON CORP. V. GOVERNOR OF MARYLAND 135 117 Opinion of Blackmun, J. trust policy, I dissent from Part III of the Court’s opinion because it fails to condemn impermissible discrimination against interstate commerce in retail gasoline marketing. The divestiture provisions, Md. Code Ann., Art. 56, §§ 157E (b) and (c) (Supp. 1977) (hereinafter referred to as §§ (b) and (c)), preclude out-of-state competitors from retailing gasoline within Maryland. The effect is to protect in-state retail service station dealers from the competition of the out-of-state businesses. This protectionist discrimination is not justified by any legitimate state interest that cannot be vindicated by more evenhanded regulation. Sections (b) and (c), therefore, violate the Commerce Clause.2 I In Maryland the retail marketing of gasoline is interstate commerce, for all petroleum products come from outside the State. Retailers serve interstate travelers. To the extent that particular retailers succeed or fail in their businesses, the interstate wholesale market for petroleum products is affected. Cf. Dean Milk Co. v. Madison, 340 U. S. 349 (1951).3 The apply all equipment rentals uniformly to all retail service station dealers supplied. “(f) Every producer, refiner or wholesaler of petroleum products shall apportion uniformly all gasoline and special fuels to all retail service station dealers during periods of shortages on an equitable basis, and shall not discriminate among the dealers in their allotments.” 2 U. S. Const., Art. I, § 8, cl. 3: “The Congress shall have Power . . . “To regulate Commerce with foreign Nations, and among the several States, and with the Indian Tribes.” 3 The inherent effect of local regulation of retail sales on interstate commerce is well illustrated by Dean Milk. The city of Madison forbade the sale of pasteurized milk unless pasteurization occurred at a plant located within five miles of the center of the city. General Ordinances of the City of Madison §7.21 (1949). Even though only local sale was prohibited, the Court considered the ordinance to be a regulation of interstate commerce. 136 OCTOBER TERM, 1977 Opinion of Blackmun, J. 437 U. S. regulation of retail gasoline sales is therefore within the scope of the Commerce Clause. See ibid.; Minnesota v. Barber, 136 U. S. 313 (1890).4 A The Commerce Clause forbids discrimination against interstate commerce, which repeatedly has been held to mean that States and localities may not discriminate against the transactions of out-of-state actors in interstate markets. E. g., Hunt v. Washington Apple Advertising Comm’n, 432 U. S. 333, 350-352 (1977); Halliburton Oil Well Co. v. Reily, 373 U. S. 64, 69-73 (1963); Dean Milk Co. v. Madison, 340 U. S., at 354; Best & Co. v. Maxwell, 311 U. S. 454, 455-456 (1940). The discrimination need not appear on the face of the state or local regulation. “The commerce clause forbids discrimination, whether forthright or ingenious. In each case it is our duty to determine whether the statute under attack, whatever its name may be, will in its practical operation work discrimination against interstate commerce.” Ibid, (footnote omitted). The state or local authority need not intend to discriminate in order to offend the policy of maintaining a free-flowing national economy. As demonstrated in Hunt, a statute that on its face restricts both intrastate and interstate transactions may violate the Clause by having the “practical effect” of discriminating in its operation. 432 U. S., at 350-352. If discrimination results from a statute, the burden falls upon the state or local government to demonstrate legitimate local benefits justifying the inequality and to show that less discriminatory alternatives cannot protect the local interests. 4 Cf. Best & Co. v. Maxwell, 311 U. S. 454 (1940) (holding that taxation of local retailing was within the reach of the Commerce Clause); United States v. Frankfort Distilleries, Inc., 324 U. S. 293 (1945) (holding that retailing was interstate commerce within the scope of the Sherman Act). See generally Note, Gasoline Marketing Divestiture Statutes: A Preliminary Constitutional and Economic Assessment, 28 Vand. L. Rev. 1277, 1303 (1975). EXXON CORP. V. GOVERNOR OF MARYLAND 137 117 Opinion of Blackmun, J. Id., at 353; Dean Milk Co. v. Madison, 340 U. S., at 354. This Court does not merely accept without analysis purported local interests. Instead, it independently identifies the character of the interests and judges for itself whether alternatives will be adequate. For example, in Dean Milk the city attempted to justify a milk pasteurization ordinance by claiming it to be a necessary health measure. The city’s assertion was not conclusive, however: “A different view, that the ordinance is valid simply because it professes to be a health measure, would mean that the Commerce Clause of itself imposes no limitations on state action other than those laid down by the Due Process Clause, save for the rare instance where a state artlessly discloses an avowed purpose to discriminate against interstate goods.” Ibid. In an independent assessment of the asserted purpose, the Court determined exactly how the ordinance protected public health and then concluded that other measures could accomplish the same ends. Id., at 354-356. The city’s public health purpose therefore did not justify the discrimination, and the ordinance violated the Commerce Clause. B With this background, the unconstitutional discrimination in the Maryland statute becomes apparent. No facial inequality exists; §§ (b) and (c) preclude all refiners and producers from marketing gasoline at the retail level. But given the structure of the retail gasoline market in Maryland, the effect of §§ (b) and (c) is to exclude a class of predominantly out-of-state gasoline retailers while providing protection from competition to a class of nonintegrated retailers that is overwhelmingly composed of local businessmen. In 1974, of the 3,780 gasoline service stations in the State, 3,547 were operated by nonintegrated local retail dealers. App. 191, 569, 755. Of the 233 company-operated stations, 197 belonged to out-of 138 OCTOBER TERM, 1977 Opinion of Blackmun, J. 437 U. S. state integrated producers or refiners. Id., at 190-191. Thirty-four were operated by nonintegrated companies that would not have been affected immediately by the Maryland statute.5 Ibid. The only in-state integrated petroleum firm, Crown Central Petroleum, Inc., operated just two service stations. Id., at 189. Of the class of stations statutorily insulated from the competition of the out-of-state integrated firms, then, more than 99% were operated by local business interests. Of the class of enterprises excluded entirely from participation in the retail gasoline market, 95% were out-of-state firms, operating 98% of the stations in the class. Ibid. The discrimination suffered by the out-of-state integrated producers and refiners is significant. Five of the excluded enterprises, Ashland Oil, Inc., BP Oil, Inc., Kayo Oil Co., Petroleum Marketing Corp., and Southern States Cooperative, Inc., market nonbranded gasoline through price competition rather than through brand recognition. Of the 98 stations marketing gasoline in this manner, all but 6 are company operated. The company operations result from the dominant fact of price competition marketing. According to repeated testimony from petroleum economics experts and officers of price marketers—testimony that the trial court did not discredit—such nonbranded stations can compete successfully only if they have day-to-day control of the retail price of their products, the hours of operation of their stations, and related business details. App. 320, 357, 370-371, 449-451, 503-504, 5 In 1974 Fisca Oil Co., Giant Food, Inc., Hi-Way Oil, Inc., Homes Oil Co., Hudson Oil Co., Midway Petroleum, National Oil Co., Pantry Pride, Savon Gas Stations, and Sears, Roebuck & Co. operated gasoline stations in Maryland. Because none of these organizations produced or refined petroleum at that time, the statute would not have restricted their operations. It should be noted, however, that the statute will reach any of these firms deciding to integrate backwards from retailing to refining or producing. After this suit was filed, Hudson Oil Co. acquired a refinery and thus became another out-of-state business subject to the ban of §§ (b) and (c). App. 518-519. EXXON CORP. v. GOVERNOR OF MARYLAND 139 117 Opinion of Blackmun, J. 517, 529-530; Joint App. to Jurisdictional Statements 102a et seq. Only with such control can sufficient sales volume be achieved to produce satisfactory profits at prices two to three cents a gallon below those of the major branded stations. Dealer operation of stations precludes such control because of the illegality of vertical price fixing. See, e. g., 15 U. S. C. § 1 (1976 ed.); White Motor Co. v. United States, 372 U. S. 253 (1963). Therefore, because §§ (b) and (c) forbid company operations, these out-of-state competitors will have to abandon the Maryland retail market altogether. App. 100, 357-358, 455, 519; Joint App. to Jurisdictional Statements 103a et seq? For the same reason 32 other out-of-state national nonbranded integrated marketers, who operate their own stations without dealers, will be precluded from entering the Maryland retail gasoline market. The record also contains testimony that the discrimination will burden the operations of major branded companies, such as appellants Exxon, Phillips, Shell, and Gulf, all of which are out-of-state firms. Most importantly, §§ (b) and (c) will preclude these companies, as well as those mentioned in the previous paragraph, from competing directly for the profits of retail marketing. According to Richard T. Harvin, retail sales manager for Exxon’s eastern marketing region, Exxon’s company-operated stations in Maryland annually return 15% of the company’s investment—a profit of $700,000 in 1974. App. 316. Sections (b) and (c) will force this return to be shared with the local dealers. In addition, the ban of the sections will preclude the majors from enhancing brand recognition and consumer acceptance through retail outlets with company-controlled standards. Id., at 316, 320, 647, 668-669. Their ability directly to monitor consumer preferences and 6 The sections will force Ashland to divest 17 stations in which it has invested $2,381,385. Id., at 257, 258-259. Petroleum Marketing has 21 stations valued at $2,043,710. Id., at 656. 140 OCTOBER TERM, 1977 Opinion of Blackmon, J. 437U.S. reactions will be diminished. Id., at 315, 649, 669. And their opportunity for experimentation with retail marketing techniques will be curtailed. Id., at 316-317, 647-649, 669. In short, the divestiture provisions, which will require the appellant majors to cease operation of property valued at more than $10 million, will inflict significant economic hardship on Maryland’s major brand companies, all of which are out-of-state firms. Similar hardship is not imposed upon the local service station dealers by the divestiture provisions. Indeed, rather than restricting their ability to compete, the Maryland Act effectively and perhaps intentionally improves their competitive position by insulating them from competition by out-of-state integrated producers and refiners. In its answers to the various complaints in this case, the State repeatedly conceded that the Act was intended to protect “the retail dealer as an independent businessman [by] reducing the control and dominance of the vertically integrated petroleum producer and refiner in the retail market.” Id., at 33; see id., at 51, 54, 104, 128, 132, 145, 147. At trial the State’s expert said that the legislation would have the effect of protecting the local dealers against the out-of-state competition. Id., at 613. In short, the foundation of the discrimination in this case is that the local dealers may continue to enter retail transactions and to compete for retail profits while the statute will deny similar opportunities to the class composed almost entirely of out-of-state businesses.T 7 Another indication of the discrimination against out-of-state business was the amendment of the original legislative proposal to exempt wholesalers of gasoline from the divestiture requirements. The author of the proposal intended to ban retailing by wholesalers and “not to discriminate against one class as to another.” Id., at 568. On cross-examination he was asked why the exemption was enacted. He replied: “It was up to the General Assembly to make that decision. Apparently the wholesalers were represented at the testimony in the hearings. ... I did hear at a later date that they wanted to be exempt from it because EXXON CORP. V. GOVERNOR OF MARYLAND 141 117 Opinion of Blackmun, J. With discrimination proved against interstate commerce, the burden falls upon the State to justify the distinction with legitimate state interests that cannot be vindicated with more evenhanded regulation. On the record before the Court, the State fails to carry its burden. It asserts only in general terms a desire to maintain competition in gasoline retailing. Although this is a laudable goal, it cannot be accepted without further analysis, just as the Court could not accept the mere assertion of a public health justification in Dean Milk. Here, the State ignores the second half of its responsibility; it does not even attempt to demonstrate why competition cannot be preserved without banning the out-of-state interests from the retail market. The State’s showing may be so meager because any legitimate interest in competition can be vindicated with more evenhanded regulation. First, to the extent that the State’s interest in competition is nothing more than a desire to protect particular competitors—less efficient local businessmen—from the legal competition of more efficient out-of-state firms, the interest is illegitimate under the Commerce Clause. A national economy would hardly flourish if each State could effectively insist that local nonintegrated dealers handle product retailing to the exclusion of out-of-state integrated firms that would not have sufficient local political clout to challenge the influence of local businessmen with their local government leaders.8 Each State would be encouraged to “legislate accord some of the wholesalers being local jobbers had no investment or financial activity or engagement with the producer-refiner so they wanted to plea upon the mercy of the committee so to speak .... “Q. You have no information then as to why the Legislature of Maryland chose to make that discrimination? A. Not other than hearsay as to the general data that these men were local businessmen, had no definite tie in with the refinery . . . .” Id., at 568-569. 8 There is support in the record for the inference that the Maryland Legislature passed the divestiture provisions in response to the pleas of local 142 OCTOBER TERM, 1977 Opinion of Blackmun, J. 437 U. S. ing to its estimate of its own interests, the importance of its own products, and the local advantages or disadvantages of its position in a political or commercial view.” J. Story, Commentaries on the Constitution of the United States § 259 (4th ed. 1873), quoted in H. P. Hood de Sons v. Du Mond, 336 U. S. 525, 533 (1949). See also, e. g., The Federalist, Nos. 7, 11, 12 (Hamilton), No. 42 (Madison). The Commerce Clause simply does not countenance such parochialism. Second, a legitimate concern of the State could be to limit the economic power of vertical integration. But nothing in the record suggests that the vertical integration that has gasoline dealers for protection against the competition of both the price marketers and the major oil companies. For example, the executive director of the Greater Washington/Maryland Service Station Association, which represents almost 700 local Maryland dealers, testified before the Economic Matters Committee of the Maryland Senate: “I would like to begin by telling you gentlemen that these are desperate days for service station dealers. . . . “Now beset by the critical gasoline supply situation, the squeeze by his landlord-supplier and the shrinking service and tire, battery and accessory market, the dealer is now faced with an even more serious problem. “That is the sinister threat of the major oil companies to complete their takeover of the retail-marketing of gasoline, not just to be in competition with their own branded dealers, but to squeeze them out and convert their stations to company operation. “Our oil industry has grown beyond the borders of our country to where its American character has been replaced by a multinational one. “Are the legislators of Maryland now about to let this octopus loose and unrestricted in the state of Maryland, among our small businessmen to devour them? We sincerely hope not. “The men that you see here today are the back-bone of American small business. . . . “We are here today asking you, our own legislators to protect us from an economic giant who would take away our very livelihood and our children’s future in its greed for greater profits. Please give us the protection we need to save our stations.” Id., at 755, 756,761. EXXON CORP. V. GOVERNOR OF MARYLAND 143 117 Opinion of Blackmun, J. already occurred in the Maryland petroleum market has inhibited competition. Indeed, the trial court found that the retail market, dominated by 3,547 dealer outlets constituting more than 90% of the State’s service stations, is highly competitive.9 Therefore, the State has shown no need for the divestiture of existing company-owned stations required by § (c). The legitimacy of any concern about future integration, which could support the discrimination of § (b), is suspect because of the exemption granted wholesalers, which, not surprisingly, are local businesses able to influence the state legislature.10 See n. 7, supra. 9 From the facts stipulated by the parties, the trial court found: “Retail petroleum marketing in the State of Maryland is and has been a highly competitive industry. This is a result of the number and location of available facilities, the comparatively small capital costs for entering the business, the mobility of the purchaser at the time of purchasing the products, the relative interchangeability of one competitor’s products with another in the mind of the consumer, the visibility of price information, and the many choices the consumer has in terms of prices, brands, and services offered.” Joint App. to Jurisdictional Statements 99a. The continuing competitive nature of the Maryland gasoline market provided one basis for the trial court’s holding that the State had not “demonstrated a real and substantial relation to the object sought to be attained by the means selected [;] the evidence presented before it indicates that the statute is inversely related to the public welfare.” Id., at 131a-132a. The trial court therefore considered the statute unconstitutional. 10 The trial court entered several findings about the integration of the oil companies and the need for divestiture: “Apart from restraining free competition, it was shown that divestiture would be harmful to competition in the industry, and would primarily serve to protect the independent dealers rather than the public at large. There was no proven detrimental effect upon the retail market caused by company-owned-and-operated stations which could not be curbed by federal and state anti-trust laws. “The court also finds from the preponderance of the evidence that the law will preclude all of some thirty-two producer-refiners not now in the State from ever entering the competitive market in Maryland, and vertical 144 OCTOBER TERM, 1977 Opinion of Blackmun, J. 437 U. S. Third, the State appears to be concerned about unfair competitive behavior such as predatory pricing or inequitable allocation of petroleum products by the integrated firms. These are the only examples of specific misconduct asserted in the State’s answers. App. 33-34, 54—55, 81-83, 109-111, 133-134, 148-149. But none of the concerns support the discrimination in §§ (b) and (c). There is no proof in the record that any significant portion of the class of out-of-state firms burdened by the divestiture sections has engaged in such misconduct. Furthermore, predatory pricing and unfair allocation already have been prohibited by both state and federal law. See, e. g., Emergency Petroleum Allocation Act of 1973, 87 Stat. 628, 15 U. S. C. §751 et seq. (1976 ed.); Energy Policy and Conservation Act, § 461, 89 Stat. 955, 15 U. S. C. § 760g (1976 ed.); Maryland Motor Fuel Inspection Law, Md. Code Ann., Art. 56, § 157E (f) (Supp. 1977); Maryland Antitrust Act, Md. Com. Law Code Ann. § 11-201 et seq. (1975); Maryland Unfair Sales Act, Md. Com. Law Code Ann. § 11-401 et seq. (1975). Less discriminatory legislation, which would regulate the leasing of all service stations, not just those owned by the out-of-state integrated producers and refiners, could prevent whatever evils arise from short- integration will be prohibited. Neither effect is in the public interest since competition is essentially for consumer benefit. “Noteworthy also is the fact that the original draft of the law included wholesalers in the prohibition against retail selling. The final draft of the law eliminated wholesalers, for the sole reason, according to Mr. Coleman, that the wholesalers requested their elimination from the act. There is no evidence whatsoever relative to why wholesalers should have been included initially, nor how the general public benefited from their exemption. “In all the more than one hundred eighty-five pounds of pleadings, motions, briefs, exhibits and depositions before this court, there is no concrete evidence that the act was justified as to the classes of operators singled out to be affected in order to promote the general welfare of the citizens of the State. Rather, it is apparent that the entire bill is designed to benefit one class of merchants to the detriment of another.” Id., at 130a-131a (emphasis supplied). EXXON CORP. V. GOVERNOR OF MARYLAND 145 117 Opinion of Blackmun, J. term leases. Cf. Maryland Gasoline Products Marketing Act, Md. Com. Law Code Ann. § 11-304 (g) (Supp, 1977).11 In sum, the State has asserted before this Court only a vague interest in preserving competition in its retail gasoline market. It has not shown why its interest cannot be vindicated by legislation less discriminatory toward out-of-state retailers. It therefore has not met its burden to justify the discrimination inherent in §§ (b) and (c), and they violate the Commerce Clause. II The arguments of the Court’s opinion, the Maryland Court of Appeals decision,12 and appellees do not remove the unconstitutional taint from the discrimination inherent in §§ (b) and (c). A The Court offers essentially three responses to the discrimination in the retail gasoline market imposed by the divestiture provisions.13 First, the Court says that the discrimination 11 This statute states: “(g) Distributor may not unreasonably withhold certain consents . . . The distributor may not unreasonably withhold his consent to any assignment, transfer, sale, or renewal of a marketing agreement. . . .” 12 279 Md. 410, 370 A. 2d 1102 and 372 A. 2d 237 (1977). The trial court, the Circuit Court for Anne Arundel County, Md., did not address the question whether §§ (b) and (c) unconstitutionally discriminated against interstate commerce. It held that the statute offended substantive due process, in violation of the Maryland Declaration of Rights, Art. 23. 13 The Court also notes that §§ (b) and (c) do not discriminate against interstate goods and do not favor local producers and refiners. While true, the observation is irrelevant because it does not address the discrimination inflicted upon retail marketing in the State. Cf. Part II-B, infra. Footnote 16 of the Court’s opinion, ante, at 126-127, suggests that unconstitutional discrimination does not exist unless there is an effect on the quantity of out-of-state goods entering a State. This is too narrow a view of the Commerce Clause. First, interstate commerce consists of far more than mere production of goods. It also consists of transactions—of repeated buying and selling of both goods and services. By focusing exclu- 146 OCTOBER TERM, 1977 Opinion of Blackmun, J. 437 U. S. against the class of out-of-state producers and refiners does not violate the Commerce Clause because the State has not imposed similar discrimination against other out-of-state retailers. Ante, at 125-126. This is said to distinguish the present case from Hunt v. Washington Apple Advertising Comm’n. In fact, however, the unconstitutional discrimination in Hunt was not against all out-of-state interests. North Carolina had enacted a statute requiring that apples marketed in closed containers within the State bear “ ‘no grade other than the applicable U. S. grade or standard.’ ” 432 U. S., at 335. The Commission contended that the provision discriminated against interstate commerce because it prohibited the display of superior Washington State apple grading marks. The Court did not strike down the provision because it discriminated against the marketing techniques of all out-of-state growers. The provision imposed no discrimination on growers from States that employed only the United States Department of Agriculture grading system.14 Despite this sively on the quantity of goods, the Court limits the protection of the Clause to producers and handlers of goods before they enter a discriminating State. In our complex national economy, commercial transactions continue after the goods enter a State. The Court today permits a State to impose protectionist discrimination upon these later transactions to the detriment of out-of-state participants. Second, the Court cites no case in which this Court has held that a burden on the flow of goods is a prerequisite to establishing a case of unconstitutional discrimination against interstate commerce. Neither Hunt nor Dean Milk contains such a holding. In both of those cases the Court upheld the claims of discrimination; in neither did it say that a burden on the wholesale flow of goods was a necessary part of its holding. Regarding Hunt, the Court cites to 432 U. 8., at 347, which discusses only whether the appellants had met the $10,000 amount-in-controversy requirement of 28 U. 8. C. § 1331. As explained in Part II-B, infra, this case presents a threat to the flow of gasoline in Maryland identical to the threat to the flow of milk in Dean Milk. 14 Growers from 13 States marketed apples in North Carolina. Six of the States did not have state grading systems apart from the USDA regulations. 432 U. S., at 349. EXXON CORP. v. GOVERNOR OF MARYLAND 147 117 Opinion of Blackmun, J. lack of universal discrimination, the Court declared the provision unconstitutional because it discriminated against a single segment of out-of-state marketers of apples, namely, the Washington State growers who employed the superior grading system. In this regard, the Maryland divestiture provisions are identical to, not distinguishable from, the North Carolina statute in Hunt. Here, the discrimination has been imposed against a segment of the out-of-state retailers of gasoline, namely, those who also refine or produce petroleum. To accept the argument of the Court, that is, that discrimination must be universal to offend the Commerce Clause, naively will foster protectionist discrimination against interstate commerce. In the future, States will be able to insulate in-state interests from competition by identifying the most potent segments of out-of-state business, banning them, and permitting less effective out-of-state actors to remain. The record shows that the Court permits Maryland to effect just such discrimination in this case. The State bans the most powerful out-of-state firms from retailing gasoline within its boundaries. It then insulates the forced divestiture of 199 service stations from constitutional attack by permitting out-of-state firms such as Pantry Pride, Fisca, Hi-Way, and Midway to continue to operate 34 gasoline stations. Effective out-of-state competition is thereby emasculated—no doubt, an ingenious discrimination. But as stated at the outset, “the commerce clause forbids discrimination, whether forthright or ingenious.” Best ■& Co. v. Maxwell, 311 U. S., at 455. Second, the Court contends, as a subpart of its primary argument, that the discrimination in Hunt “raised the cost of doing business for out-of-state dealers, and, in various other ways, favored the in-state dealer in the local market. 432 U. S., at 351-352. No comparable claim can be made here.” Ante, at 126. Once it is seen that the discrimination in Hunt raised the cost of doing business for only one group of the out-of-state marketers of apples, the fallacy of the Court’s 148 OCTOBER TERM, 1977 Opinion of Blackmun, J. 437 U. S. argument appears. In fact, here the burden imposed upon the class of out-of-state retailers subject to the discrimination of §§ (b) and (c) far exceeds the burdens in Hunt. In Hunt the statute merely increased costs and deprived the Washington growers of the competitive advantages of the use of their grading system. Here, the statute bans the refiners and producers from the retail market altogether—a burden that lacks comparability with the effects in Hunt only because it is more severe. Third, the Court asserts without citation: “The fact that the burden of a state regulation falls on some interstate companies does not, by itself, establish a claim of discrimination against interstate commerce.” Ante, at 126. This proposition is correct only to the extent that it is incomplete; it does not apply to the facts present here. It is true that merely demonstrating a burden on some out-of-state actors does not prove unconstitutional discrimination. But when the burden is significant, when it falls on the most numerous and effective group of out-of-state competitors, when a similar burden does not fall on the class of protected in-state businessmen, and when the State cannot justify the resulting disparity by showing that its legislative interests cannot be vindicated by more evenhanded regulation, unconstitutional discrimination exists. The facts of this litigation demonstrate such discrimination, and the Court does not argue persuasively to the contrary. B The contentions of the Maryland Court of Appeals, which also found no violation of the Commerce Clause, are no more convincing than the arguments of the Court’s opinion. First, the Court of Appeals reasoned that §§ (b) and (c) did not discriminate against the class of out-of-state refiners and producers because the wholesale flow of petroleum products into the State was not restricted. 279 Md. 410, 431, 370 A. 2d 1102, 1114 (1977). This supposedly distinguished the present EXXON CORP. V. GOVERNOR OF MARYLAND 149 117 Opinion of Blackmun, J. facts from those of Dean Milk Co. v. Madison, which involved unconstitutional discrimination against interstate commerce. To begin with, however, the distinction drawn by the Court of Appeals is basically irrelevant. The Maryland statute has not effected discrimination with regard to the wholesaling or interstate transport of petroleum. The discrimination exists with regard to retailing. The fact that gasoline will continue to flow into the State does not permit the State to deny out-of-state firms the opportunity to retail it once it arrives. Furthermore, Dean Milk cannot be distinguished on the ground asserted by the Court of Appeals. There, this Court invalidated § 7.21 of the General Ordinances of the city of Madison (1949), which outlawed the local sale of milk not pasteurized within five miles of the city. The section did not legally or effectively block the flow of out-of-state milk into Madison to any greater extent than the restrictions on sales of gasoline by out-of-state companies block the flow of gasoline here. In Dean Milk out-of-state producers could bring their milk to Madison, have it pasteurized in Madison, and sell it in Madison without violating § 7.21. If the flow of milk were at all restricted, it was merely because the out-of-state producers chose not to deal with the Madison pasteurizers. Similarly, the flow of gasoline into Maryland may be restricted if the out-of-state producers and refiners choose not to supply the dealers who replace the company-owned operations.15 Second, the Court of Appeals said the Maryland legislation did not offend the Commerce Clause because the legislature intended to preserve competition, not to discriminate against interstate commerce. 279 Md., at 431, 370 A. 2d, at 1114. 15 In fact, the disruption of the flow of gasoline in this case could be greater than the disruption of the flow of milk in Madison. The record supports the proposition that the ban on company operations may so unsettle the wholesale and refining enterprises of the independent price marketers that they will not be able profitably to supply gasoline to the stations of nonintegrated retailers in Maryland. App. 504-505, 509, 531. 150 OCTOBER TERM, 1977 Opinion of Blackmun, J. 437 U. S. With this argument, the court fell into the same trap that confines the State’s proffered justifications for the discrimination of §§ (b) and (c). To begin with, the fact that no discrimination was intended is irrelevant where, as here, discriminatory effects result from the statutory scheme. Furthermore, the fact that the legislature might have had a laudable intent when it passed the law cannot by itself justify the divestiture provisions. The State must also show that its interests cannot be vindicated by less discriminatory alternatives. The Court of Appeals erroneously failed to require such a showing from the appellees. Third, the Court of Appeals resurrected the outdated notion that retailing is merely local activity not subject to the strictures of the Commerce Clause. 279 Md., at 432, 370 A. 2d, at 1114-1115, citing Crescent Oil Co. v. Mississippi, 257 U. S. 129 (1921). In Crescent Oil the Court said that the operation of cotton gins was local manufacturing rather than interstate commerce. As explained at the beginning of Part I of this opinion, however, the interstate character of the retail gasoline market and 57 years of intervening constitutional and economic development prevent the application of Crescent Oil to the facts of this litigation. See nn. 3 and 4, and accompanying text, supra. C Finally, nothing in the argument of the appellees saves the distinctions in §§ (b) and (c) from the taint of unconstitutionality. First, the State argues that discrimination against interstate commerce has not occurred because “[n]o nexus between interstate as opposed to local interests inheres in the production or refining of petroleum.” Brief for Appellees 23. Although this statement might be correct in the abstract, it is incorrect in reality, given the structure of the Maryland petroleum market. Due to geological formation as so far known, no petroleum is produced in Maryland; due to the economics of production and refining, as well as to the geology, EXXON CORP. V. GOVERNOR OF MARYLAND 151 117 Opinion of Blackmun, J. no petroleum is refined in Maryland. As a matter of actual fact, then, an inherent nexus does exist between the out-of-state status of producers and refiners and the distribution and retailing of gasoline in Maryland. The Commerce Clause does not forbid only legislation that discriminates under all factual circumstances. It forbids discrimination in effect against interstate commerce on the specific facts of each case. If production or refining of gasoline occurred in Maryland, §§ (b) and (c) might not be unconstitutional. Under those different circumstances, however, the producers and refiners would have a fair opportunity to influence their local legislators and thereby to prevent the enactment of economically disruptive legislation. Under those circumstances, the economic disruption would be felt directly in Maryland, which would tend to make the local political processes responsive to the problems thereby created. Under those circumstances, §§ (b) and (c) might never have been passed. In this case, however, the economic disruption of the sections is visited upon out-of-state economic interests and not upon in-state businesses. One of the basic assumptions of the Commerce Clause is that local political systems will tend to be unresponsive to problems not felt by local constituents; instead, local political units are expected to act in their constituents’ in-terests.16 One of the basic purposes of the Clause, therefore, is to prevent the vindication of such self-interest from unfairly burdening out-of-state concerns and thereby disrupting the national economy. 16 Given the Nation’s experience under the Articles of Confederation, the assumption is not an unreasonable one. At that time authority to regulate commerce rested with the States rather than with Congress. The pursuit by each State of the particular interests of its economy and constituents nearly wrecked the national economy. “The almost catastrophic results from this sort of situation were harmful commercial wars and reprisals at home among the States . . . .” P. Hartman, State Taxation of Interstate Commerce 2 (1953), citing, e. g., The Federalist, Nos. 7, 11, 22 (Hamilton), No. 42 (Madison). 152 OCTOBER TERM, 1977 Opinion of Blackmun, J. 437 U. S. Second, appellees argue, as did the Court of Appeals, that §§ (b) and (c) do not discriminate impermissibly because the Maryland Legislature passed them with the intent to preserve competition. As explained above, however, the mere assertion of a laudable purpose does not carry the State’s burden to justify the discriminatory effects of the statute. See Parts I-B and II-B, supra. Third, appellees rely upon the Court of Appeals’ contention that unconstitutional discrimination against interstate commerce can be found only where the flow of interstate goods is curtailed. Appellees’ assertion fares no better than did the court’s because the appellees fail to show how the effect on the flow of interstate goods varies in kind between this case and Dean Milk. See Part II-B, supra. Ill The Court’s decision brings to mind the well-known words of Mr. Justice Cardozo: “To give entrance to [protectionism] would be to invite a speedy end of our national solidarity. The Constitution was framed under the dominion of a political philosophy less parochial in range. It was framed upon the theory that the peoples of the several states must sink or swim together, and that in the long run prosperity and salvation are in union and not division.” Baldwin v. G. A. F. Seelig, Inc., 294 U. S. 511, 523 (1935). Today, the Court fails to heed the Justice’s admonition. The parochial political philosophy of the Maryland Legislature thereby prevails. I would reverse the judgment of the Maryland Court of Appeals. TVA v. HILL 153 Syllabus TENNESSEE VALLEY AUTHORITY v. HILL et al. CERTIORARI TO THE UNITED STATES COURT OF APPEALS FOR THE SIXTH CIRCUIT No. 76-1701. Argued April 18, 1978—Decided June 15, 1978 The Endangered Species Act of 1973 (Act) authorizes the Secretary of the Interior (Secretary) in § 4 to declare a species of life “endangered.” Section 7 specifies that all “Federal departments and agencies shall, . . . with the assistance of the Secretary, utilize their authorities in furtherance of the purposes of [the] Act by carrying out programs for the conservation of endangered species . . . and by taking such action necessary to insure that actions authorized, funded, or carried out by them do not jeopardize the continued existence of such endangered species and threatened species or result in the destruction or modification of habitat of such species which is determined by the Secretary . . . to be critical.” Shortly after the Act’s passage the Secretary was petitioned to list a small fish popularly known as the snail darter as an endangered species under the Act. Thereafter the Secretary made the designation. Having determined that the snail darter apparently lives only in that portion of the Little Tennessee River that would be completely inundated by the impoundment of the reservoir created as a consequence of the completion of the Tellico Dam, he declared that area as the snail darter’s “critical habitat.” Notwithstanding the near completion of the multimillion-dollar dam, the Secretary issued a regulation in which it was declared that, pursuant to § 7, “all Federal agencies must take such action as is necessary to ensure that actions authorized, funded, or carried out by them do not result in the destruction or modification of this critical habitat area.” Respondents brought this suit to enjoin completion of the dam and impoundment of the reservoir, claiming that those actions would violate the Act by causing the snail darter’s extinction. The District Court after trial denied relief and dismissed the complaint. Though finding that the impoundment of the reservoir would probably jeopardize the snail darter’s continued existence, the court noted that Congress, though fully aware of the snail darter problem, had continued Tellico’s appropriations, and concluded that “[a]t some point in time a federal project becomes so near completion and so incapable of modification that a court of equity should not apply a statute enacted long after inception of the project to produce an unreasonable result. . . .” The Court of Appeals reversed and 154 OCTOBER TERM, 1977 Syllabus 437 U. S. ordered the District Court permanently to enjoin completion of the project “until Congress, by appropriate legislation, exempts Tellico from compliance with the Act or the snail darter has been deleted from the list of endangered species or its critical habitat materially redefined.” The court held that the record revealed a prima facie violation of § 7 in that the Tennessee Valley Authority had failed to take necessary action to avoid jeopardizing the snail darter’s critical habitat by its “actions.” The court thus rejected the contention that the word “actions” as used in § 7 was not intended by Congress to encompass the terminal phases of ongoing projects. At various times before, during, and after the foregoing judicial proceedings, TVA represented to congressional Appropriations Committees that the Act did not prohibit completion of the Tellico Project and described its efforts to transplant the snail darter. The Committees consistently recommended appropriations for the dam, sometimes stating their views that the Act did not prohibit completion of the dam at its advanced stage, and Congress each time approved TVA’s general budget, which contained funds for the dam’s continued construction. Held: 1. The Endangered Species Act prohibits impoundment of the Little Tennessee River by the Tellico Dam. Pp. 172-193. (a) The language of § 7 is plain and makes no exception such as that urged by petitioner whereby the Act would not apply to a project like Tellico that was well under way when Congress passed the Act. Pp. 172-174. (b) It is clear from the Act’s legislative history that Congress intended to halt and reverse the trend toward species extinction— whatever the cost. The pointed omission of the type of qualified language previously included in endangered species legislation reveals a conscious congressional design to give endangered species priority over the “primary missions” of federal agencies. Congress, moreover, foresaw that § 7 would on occasion require agencies to alter ongoing projects in order to fulfill the Act’s goals. Pp. 174^187. (c) None of the limited “hardship exemptions” provided in the Act would even remotely apply to the Tellico Project. P. 188. (d) Though statements in Appropriations Committee Reports reflected the view of the Committees either that the Act did not apply to Tellico or that the dam should be completed regardless of the Act’s provisions, nothing in the TVA appropriations measures passed by Congress stated that the Tellico Project was to be completed regardless of the Act’s requirements. To find a repeal under these circumstances, as petitioner has urged, would violate the “ ‘cardinal rule . . . that repeals by implication are not favored.’ ” Morton v. Mancari, 417 U. S. 535, 549. The TVA v. HILL 155 153 Syllabus doctrine disfavoring repeals by implication applies with full vigor when the subsequent legislation is an appropriations measure. When voting on appropriations measures, legislators are entitled to assume that the funds will be devoted to purposes that are lawful and not for any purpose forbidden. A contrary policy would violate the express rules of both Houses of Congress, which provide that appropriations measures may not change existing substantive law. An appropriations committee’s expression does not operate to repeal or modify substantive legislation. Pp. 189-193. 2. The Court of Appeals did not err in ordering that completion of the Tellico Dam, which would have violated the Act, be enjoined. Congress has spoken in the plainest words, making it clear that endangered species are to be accorded the highest priorities. Since that legislative power has been exercised, it is up to the Executive Branch to administer the law and for the Judiciary to enforce it when, as here, enforcement has been sought. Pp. 193-194. 549 F. 2d 1064, affirmed. Burger, C. J., delivered the opinion of the Court, in which Brennan, Stewart, White, Marshall, and Stevens, JJ., joined. Powell, J., filed a dissenting opinion, in which Blackmun, J., joined, post, p. 195. Rehnquist, J., filed a dissenting opinion, post, p. 211. Attorney General Bell argued the cause for petitioner. On the briefs were Acting Solicitor General Friedman, Deputy Solicitor General Barnett, Herbert S. Sanger, Jr., Richard A. Allen, Charles A. Wagner III, Thomas A. Pedersen, and Nicholas A. Della Volpe. Zygmunt J. B. Plater argued the cause for respondents. With him on the brief was W. P. Boone Dougherty.* *Briefs of amici curiae urging reversal were filed by Robert J. Pennington for Monroe County et al.; and by Ronald A. Zumbrun, Raymond M. Momboisse, Robert K. Best, Albert Ferri, Jr., Donald C. Simpson, and W. Hugh O’ Riordan for the Pacific Legal Foundation. Briefs of amici curiae urging affirmance were filed by Ben Oshel Bridgers for the Eastern Band of Cherokee Indians; by William A. Butler for the Environmental Defense Fund et al.; and by Howell H. Sherrod, Jr., for the East Tennessee Valley Landowners’ Assn. Ben B. Blackburn and Wayne T. Elliott filed a brief for the Southeastern Legal Foundation as amicus curiae. 156 OCTOBER TERM, 1977 Opinion of the Court 437 U. S. Mr. Chief Justice Burger delivered the opinion of the Court. The questions presented in this case are (a) whether the Endangered Species Act of 1973 requires a court to enjoin the operation of a virtually completed federal dam—which had been authorized prior to 1973—when, pursuant to authority vested in him by Congress, the Secretary of the Interior has determined that operation of the dam would eradicate an endangered species; and (b) whether continued congressional appropriations for the dam after 1973 constituted an implied repeal of the Endangered Species Act, at least as to the particular dam. I The Little Tennessee River originates in the mountains of northern Georgia and flows through the national forest lands of North Carolina into Tennessee, where it converges with the Big Tennessee River near Knoxville. The lower 33 miles of the Little Tennessee takes the river’s clear, free-flowing waters through an area of great natural beauty. Among other environmental amenities, this stretch of river is said to contain abundant trout. Considerable historical importance attaches to the areas immediately adjacent to this portion of the Little Tennessee’s banks. To the south of the river’s edge lies Fort Loudon, established in 1756 as England’s southwestern outpost in the French and Indian War. Nearby are also the ancient sites of several native American villages, the archeological stores of which are to a large extent unexplored.1 These include the Cherokee towns of Echota and Tennase, the former 1 This description is taken from the opinion of the District Judge in the first litigation involving the Tellico Dam and Reservoir Project. Environmental Defense Fund n. TV A, 339 F. Supp. 806, 808 (ED Tenn. 1972). In his opinion, “all of these benefits of the present Little Tennessee River Valley will be destroyed by impoundment of the river . . . .” Ibid. The District Judge noted that “[t]he free-flowing river is the likely habitat of one or more of seven rare or endangered fish species.” Ibid. TVA v. HILL 157 153 Opinion of the Court being the sacred capital of the Cherokee Nation as early as the 16th century and the latter providing the linguistic basis from which the State of Tennessee derives its name.2 In this area of the Little Tennessee River the Tennessee Valley Authority, a wholly owned public corporation of the United States, began constructing the Tellico Dam and Reservoir Project in 1967, shortly after Congress appropriated initial funds for its development.3 Tellico is a multipurpose regional development project designed principally to stimulate shoreline development, generate sufficient electric current to heat 20,000 homes,4 and provide flatwater recreation and flood control, as well as improve economic conditions in “an area characterized by underutilization of human resources and outmigration of young people.” Hearings on Public Works for Power and Energy Research Appropriation Bill, 1977, before a Subcommittee of the House Committee on Appropriations, 94th Cong., 2d Sess., pt. 5, p. 261 (1976). Of particular relevance to this case is one aspect of the project, a dam which TVA determined to place on the Little Tennessee, a short distance from where the river’s waters meet with the Big Tennessee. When fully operational, the dam would impound water covering some 16,500 acres—much of which represents valuable and productive farmland—thereby converting the river’s shallow, fast-flowing waters into a deep reservoir over 30 miles in length. The Tellico Dam has never opened, however, despite the fact that construction has been virtually completed and the 2 See Brief for the Eastern Band of Cherokee Indians as Amicus Curiae 2. See also Mooney, Myths of the Cherokee, 19 Bureau of American Ethnology Ann. Rep. 11 (1900); H. Timberlake, Memoirs, 1756-1765 (Watauga Press 1927); A. Brewer & C. Brewer, Valley So Wild: A Folk History (East Tenn. Historical Soc. 1975). 3 Public Works Appropriation Act, 1967, 80 Stat. 1002, 1014. 4 Tellico Dam itself will contain no electric generators; however, an interreservoir canal connecting Tellico Reservoir with a nearby hydroelectric plant will augment the latter’s capacity. 158 OCTOBER TERM, 1977 Opinion of the Court 437 U. S. dam is essentially ready for operation. Although Congress has appropriated monies for Tellico every year since 1967, progress was delayed, and ultimately stopped, by a tangle of lawsuits and administrative proceedings. After unsuccessfully urging TVA to consider alternatives to damming the Little Tennessee, local citizens and national conservation groups brought suit in the District Court, claiming that the project did not conform to the requirements of the National Environmental Policy Act of 1969 (NEPA), 83 Stat. 852, 42 U. S. C. §4321 et seq. After finding TVA to be in violation of NEPA, the District Court enjoined the dam’s completion pending the filing of an appropriate environmental impact statement. Environmental Defense Fund v. TVA, 339 F. Supp. 806 (ED Tenn.), aff’d, 468 F. 2d 1164 (CA6 1972). The injunction remained in effect until late 1973, when the District Court concluded that TVA’s final environmental impact statement for Tellico was in compliance with the law. Environmental Defense Fund v. TVA, 371 F. Supp. 1004 (ED Tenn. 1973), aff’d, 492 F. 2d 466 (CA6 1974).5 A few months prior to the District Court’s decision dissolving the NEPA injunction, a discovery was made in the waters of the Little Tennessee which would profoundly affect the Tellico Project. Exploring the area around Coy tee Springs, which is about seven miles from the mouth of the river, a University of Tennessee ichthyologist, Dr. David A. Etnier, found a previously unknown species of perch, the snail darter, or Percina (Imostoma) tanasi.6 This three-inch, tannish-colored fish, 5 The NEPA injunction was in effect some 21 months; when it was entered TVA had spent some $29 million on the project. Most of these funds have gone to purchase land, construct the concrete portions of the dam, and build a four-lane steel-span bridge to carry a state highway over the proposed reservoir. 339 F. Supp., at 808. 6 The snail darter was scientifically described by Dr. Etnier in the Proceedings of the Biological Society of Washington, Vol. 88, No. 44, pp. 469-488 (Jan. 22, 1976). The scientific merit and content of Dr. Etnier’s TVA v. HILL 159 153 Opinion of the Court whose numbers are estimated to be in the range of 10,000 to 15,000, would soon engage the attention of environmentalists, the TVA, the Department of the Interior, the Congress of the United States, and ultimately the federal courts, as a new and additional basis to halt construction of the dam. Until recently the finding of a new species of animal life would hardly generate a cause célèbre. This is particularly so in the case of darters, of which there are approximately 130 known species, 8 to 10 of these having been identified only in the last five years.7 The moving force behind the snail darter’s sudden fame came some four months after its discovery, when the Congress passed the Endangered Species Act of 1973 (Act), 87 Stat. 884, 16 U. S. C. § 1531 et seq. (1976 ed.). This legislation, among other things, authorizes the Secretary of the Interior to declare species of animal life “endangered” 8 and to paper on the snail darter were checked by a panel from the Smithsonian Institution prior to publication. See App. 111. 7 In Tennessee alone there are 85 to 90 species of darters, id., at 131, of which upward to 45 live in the Tennessee River system. Id., at 130. New species of darters are being constantly discovered and classified—at the rate of about one per year. Id., at 131. This is a difficult task for even trained ichthyologists since species of darters are often hard to differentiate from one another. Ibid. 8 An “endangered species” is defined by the Act to mean “any species which is in danger of extinction throughout all or a significant portion of its range other than a species of the Class Insecta determined by the Secretary to constitute a pest whose protection under the provisions of this chapter would present an overwhelming and overriding risk to man.” 16 U. S. C. § 1532 (4) (1976 ed.). “ ‘The act covers every animal and plant species, subspecies, and population in the world needing protection. There are approximately 1.4 million full species of animals and 600,000 full species of plants in the world. Various authorities calculate as many as 10% of them—some 200,000—may need to be listed as Endangered or Threatened. When one counts in subspecies, not to mention individual populations, the total could increase to three to five times that number.’ ” Keith Shreiner, Associate Director and Endangered Species Program Manager of the U. S. Fish and Wildlife Service, quoted in a letter from A. J. Wagner, Chairman, TVA, to 160 OCTOBER TERM, 1977 Opinion of the Court 437 U. S. identify the “critical habitat”9 of these creatures. When a species or its habitat is so listed, the following portion of the Act—relevant here—becomes effective: “The Secretary [of the Interior] shall review other programs administered by him and utilize such programs in furtherance of the purposes of this chapter. All other Federal departments and agencies shall, in consultation with and with the assistance of the Secretary, utilize their authorities in furtherance of the purposes of this chapter by carrying out programs for the conservation of endangered species and threatened species listed pursuant to section 1533 of this title and by taking such action necessary to insure that actions authorized, funded, or carried out by them do not jeopardize the continued existence of such endangered species and threatened species or result in the destruction or modification of habitat of such species which is determined by the Secretary, after consultation as appropriate with the affected States, to be critical.” 16 U. S. C. § 1536 (1976 ed.) (emphasis added). Chairman, House Committee on Merchant Marine and Fisheries, dated Apr. 25, 1977, quoted in Wood, On Protecting an Endangered Statute: The Endangered Species Act of 1973, 37 Federal B. J. 25, 27 (1978). 9 The Act does not define “critical habitat,” but the Secretary of the Interior has administratively construed the term: “ ‘Critical habitat’ means any air, land, or water area (exclusive of those existing man-made structures or settlements which are not necessary to the survival and recovery of a listed species) and constituent elements thereof, the loss of which would appreciably decrease the likelihood of the survival and recovery of a listed species or a distinct segment of its population. The constituent elements of critical habitat include, but are not limited to: physical structures and topography, biota, climate, human activity, and the quality and chemical content of land, water, and air. Critical habitat may represent any portion of the present habitat of a listed species and may include additional areas for reasonable population expansion.” 43 Fed. Reg. 874 (1978) (to be codified as 50 CFR § 402.02). TVA v. HILL 161 153 Opinion of the Court In January 1975, the respondents in this case10 and others petitioned the Secretary of the Interior11 to list the snail darter as an endangered species. After receiving comments from various interested parties, including TVA and the State of Tennessee, the Secretary formally listed the snail darter as an endangered species on October 8, 1975. 40 Fed. Reg. 47505-47506; see 50 CFR § 17.11 (i) (1976). In so acting, it was noted that “the snail darter is a living entity which is genetically distinct and reproductively isolated from other fishes.” 40 Fed. Reg. 47505. More important for the purposes of this case, the Secretary determined that the snail darter apparently lives only in that portion of the Little Tennessee River which would be completely inundated by the reservoir created as a consequence of the Tellico Dam’s completion. Id., at 47506.12 10 Respondents are a regional association of biological scientists, a Tennessee conservation group, and individuals who are citizens or users of the Little Tennessee Valley area which would be affected by the Tellico Project. 11 The Act authorizes “interested person [s]” to petition the Secretary of the Interior to list a species as endangered. 16 U. S. C. § 1533 (c) (2) (1976 ed.); see 5 U. S. C. § 553 (e) (1976 ed.). 12 Searches by TVA in more than 60 watercourses have failed to find other populations of snail darters. App. 36, 410-412. The Secretary has noted that “more than 1,000 collections in recent years and additional earlier collections from central and east Tennessee have not revealed the presence of the snail darter outside the Little Tennessee River.” 40 Fed. Reg. 47505 (1975). It is estimated, however, that the snail darter’s range once extended throughout the upper main Tennessee River and the lower portions of its major tributaries above Chattanooga—all of which are now the sites of dam impoundments. See Hearings on Public Works for Water and Power Development and Energy Research Appropriation Bill, 1978, before a Subcommittee of the House Committee on Appropriations, 95th Cong., 1st Sess., pt. 4, pp. 240-241 (1977) (statement of witness for TVA); Hearings on Endangered Species Act Oversight, before the Subcommittee on Resource Protection of the Senate Committee on Environment and Public Works, 95th Cong., 1st Sess., 291 (1977); App. 139. 162 OCTOBER TERM, 1977 Opinion of the Court 437 U. S. The Secretary went on to explain the significance of the dam to the habitat of the snail darter: u[T]he snail darter occurs only in the swifter portions of shoals over clean gravel substrate in cool, low-turbidity water. Food of the snail darter is almost exclusively snails which require a clean gravel substrate for their survival. The proposed impoundment of water behind the proposed Tellico Dam would result in total destruction of the snail darter’s habitat.” Ibid, (emphasis added). Subsequent to this determination, the Secretary declared the area of the Little Tennessee which would be affected by the Tellico Dam to be the “critical habitat” of the snail darter. 41 Fed. Reg. 13926-13928 (1976) (to be codified as 50 CFR § 17.81). Using these determinations as a predicate, and notwithstanding the near completion of the dam, the Secretary declared that pursuant to § 7 of the Act, “all Federal agencies must take such action as is necessary to insure that actions authorized, funded, or carried out by them do not result in the destruction or modification of this critical habitat area.” 41 Fed. Reg. 13928 (1976) (to be codified as 50 CFR § 17.81 (b)). This notice, of course, was pointedly directed at TVA and clearly aimed at halting completion or operation of the dam. During the pendency of these administrative actions, other developments of relevance to the snail darter issue were transpiring. Communication was occurring between the Department of the Interior’s Fish and Wildlife Service and TVA with a view toward settling the issue informally. These negotiations were to no avail, however, since TVA consistently took the position that the only available alternative was to attempt relocating the snail darter population to another suitable location. To this end, TVA conducted a search of alternative sites which might sustain the fish, culminating in the experimental transplantation of a number of snail darters to the nearby Hiwassee River. However, the Secretary of the Interior was TVA v. HILL 163 153 Opinion of the Court not satisfied with the results of these efforts, finding that TVA had presented “little evidence that they have carefully studied the Hiwassee to determine whether or not” there were “biological and other factors in this river that [would] negate a successful transplant.”13 40 Fed. Reg. 47506 (1975). Meanwhile, Congress had also become involved in the fate of the snail darter. Appearing before a Subcommittee of the House Committee on Appropriations in April 1975—some seven months before the snail darter was listed as endangered—TVA representatives described the discovery of the fish and the relevance of the Endangered Species Act to the Tellico Project. Hearings on Public Works for Water and Power Development and Energy Research Appropriation Bill, 1976, before a Subcommittee of the House Committee on Appropriations, 94th Cong., 1st Sess., pt. 7, pp. 466-467 (1975); Hearings on H. R. 8122, Public Works for Water and Power Development and Energy Research Appropriations for Fiscal Year 1976, before a Subcommittee of the Senate Committee on Appropriations, 94th Cong., 1st Sess., pt. 4, pp. 3775-3777 (1975). At that time TVA presented a position which it would advance in successive forums thereafter, namely, that the Act did not prohibit the completion of a project authorized, funded, and substantially constructed before the Act was passed. TVA also described its efforts to transplant the snail darter, but contended that the dam should be finished regardless of the 13 The Fish and Wildlife Service and Dr. Etnier have stated that it may take from 5 to 15 years for scientists to determine whether the snail darter can successfully survive and reproduce in this new environment. See General Accounting Office, The Tennessee Valley Authority’s Tellico Dam Project—Costs, Alternatives, and Benefits 4 (Oct. 14, 1977). In expressing doubt over the long-term future of the Hiwassee transplant, the Secretary noted: “That the snail darter does not already inhabit the Hiwassee River, despite the fact that the fish has had access to it in the past, is a strong indication that there may be biological and other factors in this river that negate a successful transplant.” 40 Fed. Reg. 47506 (1975). 164 OCTOBER TERM, 1977 Opinion of the Court 437 U. S. experiment’s success. Thereafter, the House Committee on Appropriations, in its June 20, 1975, Report, stated the following in the course of recommending that an additional $29 million be appropriated for Tellico: “The Committee directs that the project, for which an environmental impact statement has been completed and provided the Committee, should be completed as promptly as possible . . . .” H. R. Rep. No. 94-319, p. 76 (1975). (Emphasis added.) Congress then approved the TVA general budget, which contained funds for continued construction of the Tellico Project.14 In December 1975, one month after the snail darter was declared an endangered species, the President signed the bill into law. Public Works for Water and Power Development and Energy Research Appropriation Act, 1976, 89 Stat. 1035, 1047. In February 1976, pursuant to § 11 (g) of the Endangered Species Act, 87 Stat. 900, 16 U. S. C. § 1540 (g) (1976 ed.),15 respondents filed the case now under review, seeking to enjoin completion of the dam and impoundment of the reservoir on the ground that those actions would violate the Act by directly causing the extinction of the species Percina (Imostoma) tanasi. The District Court denied respondents’ request for a preliminary injunction and set the matter for trial. Shortly thereafter the House and Senate held appropriations hearings which would include discussions of the Tellico budget. 14 TVA projects generally are authorized by the Authority itself and are funded—without the need for specific congressional authorization—from lump-sum appropriations provided in yearly budget grants. See 16 U. S. C. §§ 831c (j) and 831z (1976 ed.). 15 Section 11 (g) allows “any person” to commence a civil action in a United States District Court to, inter alia, “enjoin any person, including the United States and any other governmental instrumentality or agency (to the extent permitted by the eleventh amendment to the Constitution), who is alleged to be in violation of any provision” of the Act “or regulation issued under the authority thereof . . . .” TVA v. HILL 165 153 Opinion of the Court At these hearings, TVA Chairman Wagner reiterated the agency’s position that the Act did not apply to a project which was over 50% finished by the time the Act became effective and some 70% to 80% complete when the snail darter was officially listed as endangered. It also notified the Committees of the recently filed lawsuit’s status and reported that TVA’s efforts to transplant the snail darter had “been very encouraging.” Hearings on Public Works for Water and Power Development and Energy Research Appropriation Bill, 1977, before a Subcommittee of the House Committee on Appropriations, 94th Cong., 2d Sess., pt. 5, pp. 261-262 (1976); Hearings on Public Works for Water and Power Development and Energy Research Appropriations for Fiscal Year 1977, before a Subcommittee of the Senate Committee on Appropriations, 94th Cong., 2d Sess., pt. 4, pp. 3096-3099 (1976). Trial was held in the District Court on April 29 and 30,1976, and on May 25, 1976, the court entered its memorandum opinion and order denying respondents their requested relief and dismissing the complaint. The District Court found that closure of the dam and the consequent impoundment of the reservoir would “result in the adverse modification, if not complete destruction, of the snail darter’s critical habitat,” 16 16 The District Court made the following findings with respect to the dam’s effect on the ecology of the snail darter: “The evidence introduced at trial showed that the snail darter requires for its survival a clear, gravel substrate, in a large-to-medium, flowing river. The snail darter has a fairly high requirement for oxygen and since it tends to exist in the bottom of the river, the flowing water provides the necessary oxygen at greater depths. Reservoirs, unlike flowing rivers, tend to have a low oxygen content at greater depths. “Reservoirs also tend to have more silt on the bottom than flowing rivers, and this factor, combined with the lower oxygen content, would make it highly probable that snail darter eggs would smother in such an environment. Furthermore, the adult snail darters would probably find this type of reservoir environment unsuitable for spawning. “Another factor that would tend to make a reservoir habitat unsuitable for snail darters is that their primary source of food, snails, probably 166 OCTOBER TERM, 1977 Opinion of the Court 437 U. S. making it “highly probable” that “the continued existence of the snail darter” would be “jeopardize [d].” 419 F. Supp. 753, 757 (ED Tenn.). Despite these findings, the District Court declined to embrace the plaintiffs’ position on the merits: that once a federal project was shown to jeopardize an endangered species, a court of equity is compelled to issue an injunction restraining violation of the Endangered Species Act. In reaching this result, the District Court stressed that the entire project was then about 80% complete and, based on available evidence, “there [were] no alternatives to impoundment of the reservoir, short of scrapping the entire project.” Id., at 758. The District Court also found that if the Tellico Project was permanently enjoined, “some $53 million would be lost in nonrecoverable obligations,” id., at 759, meaning that a large portion of the $78 million already expended would be wasted. The court also noted that the Endangered Species Act of 1973 was passed some seven years after construction on the dam commenced and that Congress had continued appropriations for Tellico, with full awareness of the snail darter problem. Assessing these various factors, the District Court concluded: “At some point in time a federal project becomes so near completion and so incapable of modification that a court of equity should not apply a statute enacted long after inception of the project to produce an unreasonable result. . . . Where there has been an irreversible and irretrievable commitment of resources by Congress to a project over a span of almost a decade, the Court should proceed with a great deal of circumspection.” Id., at 760. To accept the plaintiffs’ position, the District Court argued, would inexorably lead to what it characterized as the absurd result of requiring “a court to halt impoundment of water would not survive, in such an environment.” 419 F. Supp. 753, 756 (ED Tenn. 1976). TVA v. HILL 167 153 Opinion of the Court behind a fully completed dam if an endangered species were discovered in the river on the day before such impoundment was scheduled to take place. We cannot conceive that Congress intended such a result.” Id., at 763. Less than a month after the District Court decision, the Senate and House Appropriations Committees recommended the full budget request of $9 million fqr continued work on Tellico. See S. Rep. No. 94-960, p. 96 (1976); H. R. Rep. No. 94-1223, p. 83 (1976). In its Report accompanying the appropriations bill, the Senate Committee stated: “During subcommittee hearings, TVA was questioned about the relationship between the Tellico project’s completion and the November 1975 listing of the snail darter (a small 3-inch fish which was discovered in 1973) as an endangered species under the Endangered Species Act. TVA informed the Committee that it was continuing its efforts to preserve the darter, while working towards the scheduled 1977 completion date. TVA repeated its view that the Endangered Species Act did not prevent the completion of the Tellico project, which has been under construction for nearly a decade. The subcommittee brought this matter, as well as the recent U. S. District Court’s decision upholding TVA’s decision to complete the project, to the attention of the full Committee. The Committee does not view the Endangered Species Act as prohibiting the completion of the Tellico project at its advanced stage and directs that this project be completed as promptly as possible in the public interest.” S. Rep. No. 94-960, supra, at 96. (Emphasis added.) On June 29, 1976, both Houses of Congress passed TVA’s general budget, which included funds for Tellico; the President signed the bill on July 12, 1976. Public Works for Water and Power Development and Energy Research Appropriation Act, 1977, 90 Stat. 889, 899. 168 OCTOBER TERM, 1977 Opinion of the Court 437 U. S. Thereafter, in the Court of Appeals, respondents argued that the District Court had abused its discretion by not issuing an injunction in the face of “a blatant statutory violation.” 549 F. 2d 1064, 1069 (CA6 1977). The Court of Appeals agreed, and on January 31, 1977, it reversed, remanding “with instructions that a permanent injunction issue halting all activities incident to the Tellico Project which may destroy or modify the critical habitat of the snail darter.” Id., at 1075. The Court of Appeals directed that the injunction “remain in effect until Congress, by appropriate legislation, exempts Tellico from compliance with the Act or the snail darter has been deleted from the list of endangered species or its critical habitat materially redefined.” Ibid. The Court of Appeals accepted the District Court’s finding that closure of the dam would result in the known population of snail darters being “significantly reduced if not completely extirpated.” Id., at 1069. TVA, in fact, had conceded as much in the Court of Appeals, but argued that “closure of the Tellico Dam, as the last stage of a ten-year project, falls outside the legitimate purview of the Act if it is rationally construed.” Id., at 1070. Disagreeing, the Court of Appeals held that the record revealed a prima facie violation of § 7 of the Act, namely that TVA had failed to take “such action . . . necessary to insure” that its “actions” did not jeopardize the snail darter or its critical habitat. The reviewing court thus rejected TVA’s contention that the word “actions” in § 7 of the Act was not intended by Congress to encompass the terminal phases of ongoing projects. Not only could the court find no “positive reinforcement” for TVA’s argument in the Act’s legislative history, but also such an interpretation was seen as being “inimical to ... its objectives.” 549 F. 2d, at 1070. By way of illustration, that court pointed out that “the detrimental impact of a project upon an endangered species may not always be clearly perceived before construction is well underway.” Id., at 1071. Given such a TVA v. HILL 169 153 Opinion of the Court likelihood, the Court of Appeals was of the opinion that TVA’s position would require the District Court, sitting as a chancellor, to balance the worth of an endangered species against the value of an ongoing public works measure, a result which the appellate court was not willing to accept. Emphasizing the limits on judicial power in this setting, the court stated: “Current project status cannot be translated into a workable standard of judicial review. Whether a dam is 50% or 90% completed is irrelevant in calculating the social and scientific costs attributable to the disappearance of a unique form of life. Courts are ill-equipped to calculate how many dollars must be invested before the value of a dam exceeds that of the endangered species. Our responsibility under § 1540 (g)(1)(A) is merely to preserve the status quo where endangered species are threatened, thereby guaranteeing the legislative or executive branches sufficient opportunity to grapple with the alternatives.” Ibid. As far as the Court of Appeals was concerned, it made no difference that Congress had repeatedly approved appropriations for Tellico, referring to such legislative approval as an “advisory opinio [n]” concerning the proper application of an existing statute. In that court’s view, the only relevant legislation was the Act'itself, “[t]he meaning and spirit” of which was “clear on its face.” Id., at 1072. Turning to the question of an appropriate remedy, the Court of Appeals ruled that the District Court had erred by not issuing an injunction. While recognizing the irretrievable loss of millions of dollars of public funds which would accompany injunctive relief, the court nonetheless decided that the Act explicitly commanded precisely that result: “It is conceivable that the welfare of an endangered species may weigh more heavily upon the public conscience, as expressed by the final will of Congress, than the writeoff of those millions of dollars already expended 170 OCTOBER TERM, 1977 Opinion of the Court 437 U. S. for Tellico in excess of its present salvageable value.” Id., at 1074. Following the issuance of the permanent injunction, members of TVA’s Board of Directors appeared before Subcommittees of the House and Senate Appropriations Committees to testify in support of continued appropriations for Tellico. The Subcommittees were apprised of all aspects of Tellico’s status, including the Court of Appeals’ decision. TVA reported that the dam stood “ready for the gates to be closed and the reservoir filled,” Hearings on Public Works for Water and Power Development and Energy Research Appropriation Bill, 1978, before a Subcommittee of the House Cömmittee on Appropriations, 95th Cong., 1st Sess., pt. 4, p. 234 (1977), and requested funds for completion of certain ancillary parts of the project, such as public use areas, roads, and bridges. As to the snail darter itself, TVA commented optimistically on its transplantation efforts, expressing the opinion that the relocated fish were “doing well and ha[d] reproduced.” Id., at 235, 261-262. Both Appropriations Committees subsequently recommended the full amount requested for completion of the Tellico Project. In its June 2, 1977, Report, the House Appropriations Committee stated: “It is the Committee’s view that the Endangered Species Act was not intended to halt projects such as these in their advanced stage of completion, and [the Committee] strongly recommends that these projects not be stopped because of misuse of the Act.” H. R. Rep. No. 95-379, p. 104. (Emphasis added.) As a solution to the problem, the House Committee advised that TVA should cooperate with the Department of the Interior “to relocate the endangered species to another suitable habitat so as to permit the project to proceed as rapidly as possible.” Id., at 11. Toward this end, the Committee recom TVA v. HILL 171 153 Opinion of the Court mended a special appropriation of $2 million to facilitate relocation of the snail darter and other endangered species which threatened to delay or stop TVA projects. Much the same occurred on the Senate side, with its Appropriations Committee recommending both the amount requested to complete Tellico and the special appropriation for transplantation of endangered species. Reporting to the Senate on these measures, the Appropriations Committee took a particularly strong stand on the snail darter issue: “This committee has not viewed the Endangered Species Act as preventing the completion and use of these projects which were well under way at the time the affected species were listed as endangered. If the act has such an effect, which is contrary to the Committee’s understanding of the intent of Congress in enacting the Endangered Species Act, funds should be appropriated to allow these projects to be completed and their benefits realized in the public interest, the Endangered Species Act notwithstanding.” S. Rep. No. 95-301, p. 99 (1977). (Emphasis added.) TVA’s budget, including funds for completion of Tellico and relocation of the snail darter, passed both Houses of Congress and was signed into law on August 7, 1977. Public Works for Water and Power Development and Energy Research Appropriation Act, 1978, 91 Stat. 797. We granted certiorari, 434 U. S. 954 (1977), to review the judgment of the Court of Appeals. II We begin with the premise that operation of the Tellico Dam will either eradicate the known population of snail darters or destroy their critical habitat. Petitioner does not now seriously dispute this fact.17 In any event, under § 4 (a)(1) 17 The District Court findings are to the same effect and are unchallenged here. 172 OCTOBER TERM, 1977 Opinion of the Court 437 U. S. of the Act, 87 Stat. 886, 16 U. S. C. § 1533 (a)(1) (1976 ed.), the Secretary of the Interior is vested with exclusive authority to determine whether a species such as the snail darter is “endangered” or “threatened” and to ascertain the factors which have led to such a precarious existence. By § 4 (d) Congress has authorized—indeed commanded—the Secretary to “issue such regulations as he deems necessary and advisable to provide for the conservation of such species.” 16 U. S. C. § 1533 (d) (1976 ed.). As we have seen, the Secretary promulgated regulations which declared the snail darter an endangered species whose critical habitat would be destroyed by creation of the Tellico Reservoir. Doubtless petitioner would prefer not to have these regulations on the books, but there is no suggestion that the Secretary exceeded his authority or abused his discretion in issuing the regulations. Indeed, no judicial review of the Secretary’s determinations has ever been sought and hence the validity of his actions are not open to review in this Court. Starting from the above premise, two questions are presented: (a) would TVA be in violation of the Act if it completed and operated the Tellico Dam as planned? (b) if TVA’s actions would offend the Act, is an injunction the appropriate remedy for the violation? For the reasons stated hereinafter, we hold that both questions must be answered in the affirmative. (A) It may seem curious to some that the survival of a relatively small number of three-inch fish among all the countless millions of species extant would require the permanent halting of a virtually completed dam for which Congress has expended more than $100 million. The paradox is not minimized by the fact that Congress continued to appropriate large sums of public money for the project, even after congressional Appropriations Committees were apprised of its apparent impact upon the survival of the snail darter. We conclude, TVA v. HILL 173 153 Opinion of the Court however, that the explicit provisions of the Endangered Species Act require precisely that result. One would be hard pressed to find a statutory provision whose terms were any plainer than those in § 7 of the Endangered Species Act. Its very words affirmatively command all federal agencies “to insure that actions authorized, junded, or carried out by them do not jeopardize the continued existence” of an endangered species or “result in the destruction or modification of habitat of such species . . . .” 16 U. S. C. § 1536 (1976 ed.). (Emphasis added.) This language admits of no exception. Nonetheless, petitioner urges, as do the dissenters, that the Act cannot reasonably be interpreted as applying to a federal project which was well under way when Congress passed the Endangered Species Act of 1973. To sustain that position, however, we would be forced to ignore the ordinary meaning of plain language. It has not been shown, for example, how TVA can close the gates of the Tellico Dam without “carrying out” an action that has been “authorized” and “funded” by a federal agency. Nor can we understand how such action will “insure” that the snail darter’s habitat is not disrupted.18 Accepting the Secretary’s determinations, as 18 In dissent, Mr. Justice Powell argues that the meaning of “actions” in § 7 is “far from ‘plain,’ ” and that “it seems evident that the ‘actions’ referred to are not all actions that an agency can ever take, but rather actions that the agency is deciding whether to authorize, to fund, or to carry out.” Post, at 205. Aside from this bare assertion, however, no explanation is given to support the proffered interpretation. This recalls Lewis Carroll’s classic advice on the construction of language: “ ‘When I use a word,’ Humpty Dumpty said, in rather a scornful tone, ‘it means just what I choose it to mean—neither more nor less.’ ” Through the Looking Glass, in The Complete Works of Lewis Carroll 196 (1939). Aside from being unexplicated, the dissent’s reading of § 7 is flawed on several counts. First, under its view, the words “or carry out” in § 7 would be superfluous since all prospective actions of an agency remain to be “authorized” or “funded.” Second, the dissent’s position logically means that an agency would be obligated to comply with § 7 only when a project is in the planning stage. But if Congress had meant to so limit the Act, it 174 OCTOBER TERM, 1977 Opinion of the Court 437 U. S. we must, it is clear that TVA’s proposed operation of the dam will have precisely the opposite effect, namely the eradication of an endangered species. Concededly, this view of the Act will produce results requiring the sacrifice of the anticipated benefits of the project and of many millions of dollars in public funds.19 But examination of the language, history, and structure of the legislation under review here indicates beyond doubt that Congress intended endangered species to be afforded the highest of priorities. When Congress passed the Act in 1973, it was not legislating on a clean slate. The first major congressional concern for the preservation of the endangered species had come with passage of the Endangered Species Act of 1966, 80 Stat. 926, repealed, 87 Stat. 9O3.20 In that legislation Congress gave the surely would have used words to that effect, as it did in the National Environmental Policy Act, 42 U. S. C. §§ 4332 (2) (A), (C). 19 The District Court determined that failure to complete the Tellico Dam would result in the loss of some $53 million in nonrecoverable obligations; see supra, at 166. Respondents dispute this figure, and point to a recent study by the General Accounting Office, which suggests that the figure could be considerably less. See GAO Study, n. 13, supra, at 5-14; see also Cook, Cook, & Gove, The Snail Darter & the Dam, 51 National Parks & Conservation Magazine 10 (1977); Conservation Foundation Letter 1-2 (Apr. 1978). The GAO study also concludes that TVA and Congress should explore alternatives to impoundment of the reservoir, such as the creation of a regional development program based on a free-flowing river. None of these considerations are relevant to our decision, however; they are properly addressed to the Executive and Congress. 20 Prior federal involvement with endangered species had been quite limited. For example, the Lacey Act of 1900, 31 Stat. 187, partially codified in 16 U. S. C. §§ 667e and 701 (1976 ed.), and the Black Bass Act of 1926, 44 Stat. 576, as amended, 16 U. S. C. §851 et seq. (1976 ed.), prohibited the transportation in interstate commerce of fish or wildlife taken in violation of national, state, or foreign law. The effect of both of these statutes was constrained, however, by the fact that prior to passage of the Endangered Species Act of 1973, there were few laws regulating these TVA v. HILL 175 153 Opinion of the Court Secretary power to identify “the names of the species of native fish and wildlife found to be threatened with extinction,” § 1 (c), 80 Stat. 926, as well as authorization to purchase land for the conservation, protection, restoration, and propagation of “selected species” of “native fish and wildlife” threatened with extinction. §§ 2 (a)-(c), 80 Stat. 926-927. Declaring the preservation of endangered species a national policy, the 1966 Act directed all federal agencies both to protect these species and “insofar as is practicable and consistent with the[ir] primary purposes,” § 1 (b), 80 Stat. 926, “preserve the habitats of such threatened species on lands under their jurisdiction.” Ibid. (Emphasis added.) The 1966 statute was not a sweeping prohibition on the taking of endangered species, however, except on federal lands, § 4 (c), 80 Stat. 928, and even in those federal areas the Secretary was authorized to allow the hunting and fishing of endangered species. § 4 (d)(1), 80 Stat. 928. In 1969 Congress enacted the Endangered Species Conservation Act, 83 Stat. 275, repealed, 87 Stat. 903, which continued the provisions of the 1966 Act while at the same time broadening federal involvement in the preservation of endangered species. Under the 1969 legislation, the Secretary was empowered to list species “threatened with worldwide extinction,” § 3 (a), 83 Stat. 275; in addition, the importation of any species so recognized into the United States was prohibited. § 2, 83 Stat. 275. An indirect approach to the taking of creatures. See Coggins, Conserving Wildlife Resources: An Overview of the Endangered Species Act of 1973, 51 N. D. L. Rev. 315, 317-318 (1975). The Migratory Bird Treaty Act, passed in 1918, 40 Stat. 755, as amended, 16 U. S. C. § 703 et seq. (1976 ed.), was more extensive, giving the Secretary of the Interior power to adopt regulations for the protection of migratory birds. Other measures concentrated on establishing refuges for wildlife. See, e. g., Land and Water Conservation Fund Act of 1965, 78 Stat. 897, 16 U. S. C. § 460Z-4 et seq. (1976 ed.). See generally Environmental Law Institute, The Evolution of National Wildlife Law (1977). 176 OCTOBER TERM, 1977 Opinion of the Court 437U.S. endangered species was also adopted in the Conservation Act by way of a ban on the transportation and sale of wildlife taken in violation of any federal, state, or foreign law. §§ 7 (a)-(b), 83 Stat. 279.21 Despite the fact that the 1966 and 1969 legislation represented “the most comprehensive of its type to be enacted by any nation” 22 up to that time, Congress was soon persuaded that a more expansive approach was needed if the newly declared national policy of preserving endangered species was to be realized. By 1973, when Congress held hearings on what would later become the Endangered Species Act of 1973, it was informed that species were still being lost at the rate of about one per year, 1973 House Hearings 306 (statement of Stephen R. Seater, for Defenders of Wildlife), and “the pace of disappearance of species” appeared to be “accelerating.” H. R. Rep. No. 93-412, p. 4 (1973). Moreover, Congress was also told that the primary cause of this trend was something other than the normal process of natural selection: “[M]an and his technology has [sic] continued at an ever-increasing rate to disrupt the natural ecosystem. This has resulted in a dramatic rise in the number and severity of the threats faced by the world’s wildlife. The truth in this is apparent when one realizes that half of the recorded extinctions of mammals over the past 2,000 years have occurred in the most recent 50-year period.” 1973 House Hearings 202 (statement of Assistant Secretary of the Interior). 21 This approach to the problem of taking, of course, contained the same inherent limitations as the Lacey and Black Bass Acts, discussed, n. 20, supra. 22 Hearings on Endangered Species before the Subcommittee of the House Committee on Merchant Marine and Fisheries, 93d Cong., 1st Sess., 202 (1973) (statement of Assistant Secretary of the Interior) (hereinafter cited as 1973 House Hearings). TVA v. HILL 177 153 Opinion of the Court That Congress did not view these developments lightly was stressed by one commentator: “The dominant theme pervading all Congressional discussion of the proposed [Endangered Species Act of 1973] was the overriding need to devote whatever effort and resources were necessary to avoid further diminution of national and worldwide wildlife resources. Much of the testimony at the hearings and much debate was devoted to the biological problem of extinction. Senators and Congressmen uniformly deplored the irreplaceable loss to aesthetics, science, ecology, and the national heritage should more species disappear.” Coggins, Conserving Wildlife Resources: An Overview of the Endangered Species Act of 1973, 51 N. D. L. Rev. 315, 321 (1975). (Emphasis added.) The legislative proceedings in 1973 are, in fact, replete with expressions of concern over the risk that might lie in the loss of any endangered species.23 Typifying these sentiments is the Report of the House Committee on Merchant Marine and 23 See, e. g., 1973 House Hearings 280 (statement of Rep. Roe); id., at 281 (statement of Rep. Whitehurst); id., at 301 (statement of Friends of the Earth); id., at 306-307 (statement of Defenders of Wildlife). One statement, made by the Assistant Secretary of the Interior, particularly deserves notice: “I have watched in my lifetime a vast array of mollusks in southern streams totally disappear as a result of damming, channelization, and pollution. It is often asked of me, 'what is the importance of the mollusks for example in Alabama.’ I do not know, and I do not know whether any of us will ever have the insight to know exactly why these mollusks evolved over millions of years or what their importance is in the total ecosystem. However, I have great trouble being party to their destruction without ever having gained such knowledge.” Id., at 207. One member of the mollusk family existing in these southern rivers is the snail, see 12 Encyclopedia Britannica 326 (15th ed. 1974), which ironically enough provides the principal food for snail darters. See supra, at 162, 165-166, n. 16. 178 OCTOBER TERM, 1977 Opinion of the Court 437U.S. Fisheries on H. R. 37, a bill which contained the essential features of the subsequently enacted Act of 1973; in explaining the need for the legislation, the Report stated: “As we homogenize the habitats in which these plants and animals evolved, and as we increase the pressure for products that they are in a position to supply (usually unwillingly) we threaten their—and our own—genetic heritage. “The value of this genetic heritage is, quite literally, incalculable. “From the most narrow possible point of view, it is in the best interests of mankind to minimize the losses of genetic variations. The reason is simple: they are potential resources. They are keys to puzzles which we cannot solve, and may provide answers to questions which we have not yet learned to ask. “To take a homely, but apt, example: one of the critical chemicals in the regulation of ovulations in humans was found in a common plant. Once discovered, and analyzed, humans could duplicate it synthetically, but had it never existed—or had it been driven out of existence before we knew its potentialities—we would never have tried to synthesize it in the first place. “Who knows, or can say, what potential cures for cancer or other scourges, present or future, may lie locked up in the structures of plants which may yet be undiscovered, much less analyzed? . . . Sheer self-interest impels us to be cautious. “The institutionalization of that caution lies at the heart of H. R. 37 . . . ” H. R. Rep. No. 93-412, pp. 4-5 (1973). (Emphasis added.) As the examples cited here demonstrate, Congress was concerned about the unknown uses that endangered species might TVA v. HILL 179 153 Opinion of the Court have and about the unforeseeable place such creatures may have in the chain of life on this planet. In shaping legislation to deal with the problem thus presented, Congress started from the finding that “[t]he two major causes of extinction are hunting and destruction of natural habitat.” S. Rep. No. 93-307, p. 2 (1973). Of these twin threats, Congress was informed that the greatest was destruction of natural habitats; see 1973 House Hearings 236 (statement of Associate Deputy Chief for National Forest System, Dept, of Agriculture) ; id., at 241 (statement of Director of Mich. Dept, of Natural Resources) ; id., at 306 (statement of Stephen R. Seater, Defenders of Wildlife) ; Lachenmeier, The Endangered Species Act of 1973: Preservation or Pandemonium?, 5 Environ. Law 29, 31 (1974). Witnesses recommended, among other things, that Congress require all land-managing agencies “to avoid damaging critical habitat for endangered species and to take positive steps to improve such habitat.” 1973 House Hearings 241 (statement of Director of Mich. Dept, of Natural Resources). Virtually every bill introduced in Congress during the 1973 session responded to this concern by incorporating language similar, if not identical, to that found in the present § 7 of the Act.24 These provisions were designed, in the words of an administration witness, “for the first time [to] prohibit [a] federal agency from taking action which does jeopardize the status of endangered species,” Hearings on S. 1592 and S. 1983 before the Subcommittee on Environment of the Senate Committee on Commerce, 93d Cong., 1st Sess., 68 (1973) (statement of 24 For provisions in the House bills, see § 5 (d) of H. R. 37, 470, 471, 1511, 2669, 3696, and 3795; §3 (d) of H. R. 1461 and 4755; § 5 (d) of H. R. 2735; §3 (d) of H. R. 4758. For provisions in the Senate bills, see § 3 (d) of S. 1592; § 5 (d) of S. 1983. The House bills are collected in 1973 House Hearings 87-185; the Senate bills are found in the Hearings on S. 1592 and S. 1983 before the Subcommittee on Environment of the Senate Committee on Commerce, 93d Cong., 1st Sess., 3-49 (1973). 180 OCTOBER TERM, 1977 Opinion of the Court 437 U. S. Deputy Assistant Secretary of the Interior) (emphasis added); furthermore, the proposed bills would “direc[t\ all . . . Federal agencies to utilize their authorities for carrying out programs for the protection of endangered animals.” 1973 House Hearings 205 (statement of Assistant Secretary of the Interior), (Emphasis added.) As it was finally passed, the Endangered Species Act of 1973 represented the most comprehensive legislation for the preservation of endangered species ever enacted by any nation. Its stated purposes were “to provide a means whereby the ecosystems upon which endangered species and threatened species depend may be conserved,” and “to provide a program for the conservation of such . . . species . . . .” 16 U. S. C. § 1531 (b) (1976 ed.). In furtherance of these goals, Congress expressly stated in § 2 (c) that “all Federal departments and agencies shall seek to conserve endangered species and threatened species . . . .” 16 U. S. C. § 1531 (c) (1976 ed.). (Emphasis added.) Lest there be any ambiguity as to the meaning of this statutory directive, the Act specifically defined “conserve” as meaning “to use and the use of all methods and procedures which are necessary to bring any endangered species or threatened species to the point at which the measures provided pursuant to this chapter are no longer necessary.” § 1532 (2). (Emphasis added.) Aside from § 7, other provisions indicated the seriousness with which Congress viewed this issue: Virtually all dealings with endangered species, including taking, possession, transportation, and sale, were prohibited, 16 U. S. C. § 1538 (1976 ed.), except in extremely narrow circumstances, see § 1539 (b). The Secretary was also given extensive power to develop regulations and programs for the preservation of endangered and threatened species.25 § 1533 (d). Citizen 25 A further indication of the comprehensive scope of the 1973 Act lies in Congress’ inclusion of “threatened species” as a class deserving federal protection. Threatened species are defined as those which are “likely to become an endangered species within the foreseeable future throughout all TVA v. HILL 181 153 Opinion of the Court involvement was encouraged by the Act, with provisions allowing interested persons to petition the Secretary to list a species as endangered or threatened, § 1533 (c)(2), see n. 11, supra, and bring civil suits in United States district courts to force compliance with any provision of the Act, §§ 1540 (c) and (g). Section 7 of the Act, which of course is relied upon by respondents in this case, provides a particularly good gauge of congressional intent. As we have seen, this provision had its genesis in the Endangered Species Act of 1966, but that legislation qualified the obligation of federal agencies by stating that they should seek to preserve endangered species only “insofar as is practicable and consistent with the[ir] primary purposes . . . Likewise, every bill introduced in 1973 contained a qualification similar to that found in the earlier statutes.26 Exemplary of these was the administration bill, H. R. 4758, which in § 2 (b) would direct federal agencies to use their authorities to further the ends of the Act “insofar as is practicable and consistent with the[ir] primary purposes ... y (Emphasis added.) Explaining the idea behind this language, an administration spokesman told Congress that it “would further signal to all . . . agencies of the Government that this is the first priority, consistent with their primary objectives.” 1973 House Hearings 213 (statement of Deputy Assistant Secretary of the Interior). (Emphasis added.) This type of language did not go unnoticed by those advocating strong endangered species legislation. A representative of the or a significant portion of [their] range.” 16 U. S. C. § 1532 (15) (1976 ed.). 26 For provisions in the House bills, see §§ 2 (c) and 5 (d) of H. R. 37, 470, 471, 1511, 2669, 3310, 3696, and 3795; § 3 (d) of H. R. 1461 and 4755; § 5 (d) of H. R. 2735; § 2 (b) of H. R. 4758; one other House bill, H. R. 2169, imposed no requirements on federal agencies. For provisions in the Senate bills, see § 2 (b) of S. 1592 ; §§ 2 (b), and 5 (d) of S. 1983. 182 OCTOBER TERM, 1977 Opinion of the Court 437U.S. Sierra Club, for example, attacked the use of the phrase “consistent with the primary purpose” in proposed H. R. 4758, cautioning that the qualification “could be construed to be a declaration of congressional policy that other agency purposes are necessarily more important than protection of endangered species and would always prevail if conflict were to occur.” 1973 House Hearings 335 (statement of the chairman of the Sierra Club’s National Wildlife Committee); see id., at 251 (statement for the National Audubon Society). What is very significant in this sequence is that the final version of the 1973 Act carefully omitted all of the reservations described above. In the bill which the Senate initially approved (S. 1983), however, the version of the current § 7 merely required federal agencies to “carry out such programs as are practicable for the protection of species listed . . . 27 S. 1983, § 7 (a). (Emphasis added.) By way of contrast, the bill that originally passed the House, H. R. 37, contained a provision which was essentially a mirror image of the subsequently passed § 7—indeed all phrases which might have qualified an agency’s responsibilities had been omitted from the bill.28 In explaining the expected impact of this provision in H. R. 37 on federal agencies, the House Committee’s Report states: “This subsection requires the Secretary and the heads of all other Federal departments and agencies to use their authorities in order to carry out programs for the pro 27 We note, however, that in the version of S. 1983 which was sent to the floor of the Senate by the Senate Committee on Commerce, the qualifying language “wherever practicable” had been omitted from one part of the bill, that being §2 (b). See 119 Cong. Rec. 25663 (1973). Section 2 (b) was the portion of S. 1983 that stated the “purposes and policy” of Congress. But the Committee’s version of S. 1983—which was reported to the full Senate—retained the limitation on § 7 that we note here. 119 Cong. Rec. 25664 (1973). 28 See id., at 30157-30162. TVA v. HILL 183 153 Opinion of the Court tection of endangered species, and it further requires that those agencies take the necessary action that will not jeopardize the continuing existence of endangered species or result in the destruction of critical habitat of those species.” H. R. Rep. No. 93-412, p. 14 (1973). (Emphasis added.) Resolution of this difference in statutory language, as well as other variations between the House and Senate bills, was the task of a Conference Committee. See 119 Cong. Rec. 30174-30175, 31183 (1973). The Conference Report, H. R. Conf. Rep. No. 93-740 (1973), basically adopted the Senate bill, S. 1983; but the conferees rejected the Senate version of § 7 and adopted the stringent, mandatory language in H. R. 37. While the Conference Report made no specific reference to this choice of provisions, the House manager of the bill, Representative Dingell, provided an interpretation of what the Conference bill would require, making it clear that the mandatory provisions of § 7 were not casually or inadvertently included: “[Section 7] substantially amplifie[s] the obligation of [federal agencies] to take steps within their power to carry out the purposes of this act. A recent article . . . illustrates the problem which might occur absent this new language in the bill. It appears that the whooping cranes of this country, perhaps the best known of our endangered species, are being threatened by Air Force bombing activities along the gulf coast of Texas. Under existing law, the Secretary of Defense has some discretion as to whether or not he will take the necessary action to see that this threat disappears .... [O]nce the bill is enacted, [the Secretary of Defense] would be required to take the proper steps. . . . “Another example . . . [has] to do with the continental population of grizzly bears which may or may not be endangered, but which is surely threatened. . . . Once this 184 OCTOBER TERM, 1977 Opinion of the Court 437 U. S. bill is enacted, the appropriate Secretary, whether of Interior, Agriculture or whatever, will have to take action to see that this situation is not permitted to worsen, and that these bears are not driven to extinction. The purposes of the bill included the conservation of the species and of the ecosystems upon which they depend, and every agency of government is committed to see that those purposes are carried out. . . . [T]he agencies of Government can no longer plead that they can do nothing about it. They can, and they must. The law is clear.” 119 Cong. Rec. 42913 (1973). (Emphasis added.) It is against this legislative background29 that we must measure TVA’s claim that the Act was not intended to stop operation of a project which, like Tellico Dam, was near completion when an endangered species was discovered in its path. While there is no discussion in the legislative history of precisely this problem, the totality of congressional action makes it abundantly clear that the result we reach today is wholly in accord with both the words of the statute and the intent of Congress. The plain intent of Congress in enacting this statute was to halt and reverse the trend toward species extinction, whatever the cost. This is reflected not only in the stated policies of the Act, but in literally every section of the statute. All persons, including federal agencies, are specifically instructed not to “take” endangered species, meaning that no one is “to harass, harm,[30] pursue, hunt, shoot, 29 When confronted with a statute which is plain and unambiguous on its face, we ordinarily do not look to legislative history as a guide to its meaning. Ex parte Collett, 337 U. S. 55, 61 (1949), and cases, cited therein. Here it is not necessary to look beyond the words of the statute. We have undertaken such an analysis only to meet Mr. Justice Powell’s suggestion that the “absurd” result reached in this case, post, at 196, is not in accord with congressional intent. 30 We do not understand how TVA intends to operate Tellico Dam without “harming” the snail darter. The Secretary of the Interior has defined the term “harm” to mean “an act or omission which actually TVA v. HILL 185 153 Opinion of the Court wound, kill, trap, capture, or collect” such life forms. 16 U. S. C. §§ 1532 (14), 1538 (a)(1)(B) (1976 ed.). Agencies in particular are directed by §§ 2 (c) and 3 (2) of the Act to “use ... all methods and procedures which are necessary” to preserve endangered species. 16 U. S. C. §§ 1531 (c), 1532 (2) (1976 ed.) (emphasis added). In addition, the legislative history undergirding § 7 reveals an explicit congressional decision to require agencies to afford first priority to the declared national policy of saving endangered species. The pointed omission of the type of qualifying language previously included in endangered species legislation reveals a conscious decision by Congress to give endangered species priority over the “primary missions” of federal agencies. It is not for us to speculate, much less act, on whether Congress would have altered its stance had the specific events of this case been anticipated. In any event, we discern no hint in the deliberations of Congress relating to the 1973 Act that would compel a different result than we reach here.31 injures or kills wildlife, including acts which annoy it to such an extent as to significantly disrupt essential behavioral patterns, which include, but are not limited to, breeding, feeding or sheltering; significant environmental modification or degradation which has such effects is included within the meaning of ‘harm.’ ” 50 CFR § 17.3 (1976) (emphasis added); see S. Rep. No. 93-307, p. 7 (1973). 31 The only portion of the legislative history which petitioner cites as being favorable to its position consists of certain statements made by Senator Tunney on the floor of the Senate during debates on S. 1983; see 119 Cong. Rec. 25691-25692 (1973). Senator Tunney was asked whether the proposed bill would affect the Army Corps of Engineers’ decision to build a road through a particular area of Kentucky. Responding to this question, Senator Tunney opined that § 7 of S. 1983 would require consultation among the agencies involved, but that the Corps of Engineers “would not be prohibited from building such a road if they deemed it necessary to do so.” 119 Cong. Rec. 25689 (1973). Petitioner interprets these remarks to mean that an agency, after balancing the respective interests involved, could decide to take action which would extirpate an endangered species. If that is what Senator Tunney meant, his views are in distinct 186 OCTOBER TERM, 1977 Opinion of the Court 437 U. S. Indeed, the repeated expressions of congressional concern over what it saw as the potentially enormous danger presented by the eradication of any endangered species suggest how the balance would have been struck had the issue been presented to Congress in 1973. Furthermore, it is clear Congress foresaw that § 7 would, on occasion, require agencies to alter ongoing projects in order to fulfill the goals of the Act.32 Congressman Dingell’s discussion of Air Force practice bombing, for instance, obviously pinpoints a particular activity—intimately related to contrast to every other expression in the legislative history as to the meaning of § 7. For example, when the Kentucky example was brought up in the Senate hearings, an administration spokesman interpreted an analogous provision in S. 1592 as “prohibit[ing] [a] federal agency from taking action which does jeopardize the status of endangered species.” Supra, at 179. Moreover, we note that the version of S. 1983 being discussed by Senator Tunney contained the “as practicable” limitation in § 7 (a) which we have previously mentioned. See supra, at 182. Senator Tunney’s remarks perhaps explain why the Conference Committee subsequently deleted all such qualifying expressions. We construe the Senator’s remarks as simply meaning that under the 1973 Act the agency responsible for the project would have the “final decision,” 119 Cong. Rec. 25690 (1973), as to whether the action should proceed, notwithstanding contrary advice from the Secretary of the Interior. The Secretary’s recourse would be to either appeal to higher authority in the administration, or proceed to federal court under the relevant provisions of the Act; citizens may likewise seek enforcement under 16 U. S. C. § 1540 (g) (1976 ed.), as has been done in this case. 32 Mr. Justice Powell characterizes the result reached here as giving “retroactive” effect to the Endangered Species Act of 1973. We cannot accept that contention. Our holding merely gives effect to the plain words of the statute, namely, that § 7 affects all projects which remain to be authorized, funded, or carried out. Indeed, under the Act there could be no “retroactive” application since, by definition, any prior action of a federal agency which would have come under the scope of the Act must have already resulted in the destruction of an endangered species or its critical habitat. In that circumstance the species would have already been extirpated or its habitat destroyed; the Act would then have no subject matter to which it might apply. TVA v. HILL 187 153 Opinion of the Court the national defense—which a major federal department would be obliged to alter in deference to the strictures of § 7. A similar example is provided by the House Committee Report: “Under the authority of [§ 7], the Director of the Park Service would be required to conform the practices of his agency to the need for protecting the rapidly dwindling stock of grizzly bears within Yellowstone Park. These bears, which may be endangered, and are undeniably threatened, should at least be protected by supplying them with carcasses from excess elk within the park, by curtailing the destruction of habitat by clearcutting National Forests surrounding the Park, and by preventing hunting until their numbers have recovered sufficiently to withstand these pressures.” H. R. Rep. No. 93-412, p. 14 (1973). (Emphasis added.) One might dispute the applicability of these examples to the Tellico Dam by saying that in this case the burden on the public through the loss of millions of unrecoverable dollars would greatly outweigh the loss of the snail darter.33 But neither the Endangered Species Act nor Art. Ill of the Constitution provides federal courts with authority to make such fine utilitarian calculations. On the contrary, the plain language of the Act, buttressed by its legislative history, shows clearly that Congress viewed the value of endangered species as “incalculable.” Quite obviously, it would be difficult for 33 Mr. Justice Powell’s dissent places great reliance on Church of the Holy Trinity v. United States, 143 U. S. 457, 459 (1892), post, at 204, to support his view of the 1973 Act’s legislative history. This Court, however, later explained Holy Trinity as applying only in “rare and exceptional circumstances. . . . And there must be something to make plain the intent of Congress that the letter of the statute is not to prevail.” Crooks v. Harrelson, 282 U. S. 55, 60 (1930). As we have seen from our explication of the structure and history of the 1973 Act, there is nothing to support the assertion that the literal meaning of § 7 should not apply in this case. 188 OCTOBER TERM, 1977 Opinion of the Court 437U.S. a court to balance the loss of a sum certain—even $100 million—against a congressionally declared “incalculable” value, even assuming we had the power to engage in such a weighing process, which we emphatically do not. In passing the Endangered Species Act of 1973, Congress was also aware of certain instances in which exceptions to the statute’s broad sweep would be necessary. Thus, § 10, 16 U. S. C. § 1539 (1976 ed.), creates a number of limited “hardship exemptions,” none of which would even remotely apply to the Tellico Project. In fact, there are no exemptions in the Endangered Species Act for federal agencies, meaning that under the maxim expressio unius est exclusio alterius, we must presume that these were the only “hardship cases” Congress intended to exempt. Cf. National Railroad Passenger Corp. v. National Assn, of Railroad Passengers, 414 U. S. 453, 458 (1974).34 34 Mr. Justice Powell’s dissent relies on cases decided under the National Environmental Policy Act to support its position that the 1973 Act should only apply to prospective actions of an agency. Post, at 205-206. The NEPA decisions, however, are completely inapposite. First, the two statutes serve different purposes. NEPA essentially imposes a procedural requirement on agencies, requiring them to engage in an extensive inquiry as to the effect of federal actions on the environment; by way of contrast, the 1973 Act is substantive in effect, designed to prevent the loss of any endangered species, regardless of the cost. Thus, it would make sense to hold NEPA inapplicable at some point in the life of a project, because the agency would no longer have a meaningful opportunity to weigh the benefits of the project versus the detrimental effects on the environment. Section 7, on the other hand, compels agencies not only to consider the effect of their projects on endangered species, but to take such actions as are necessary to insure that species are not extirpated as a result of federal activities. Second, even the NEPA cases have generally required agencies to file environmental impact statements when the remaining governmental action would be environmentally “significant.” See, e. g., Environmental Defense Fund n. TVA, 468 F. 2d 1164, 1177 (CA6 1972). Under §7, the loss of any endangered species has been determined by Congress to be environmentally “significant.” See supra, at 177-179. TVA v. HILL 189 153 Opinion of the Court Notwithstanding Congress’ expression of intent in 1973, we are urged to find that the continuing appropriations for Tellico Dam constitute an implied repeal of the 1973 Act, at least insofar as it applies to the Tellico Project. In support of this view, TVA points to the statements found in various House and Senate Appropriations Committees’ Reports; as described in Part I, supra, those Reports generally reflected the attitude of the Committees either that the Act did not apply to Tellico or that the dam should be completed regardless of the provisions of the Act. Since we are unwilling to assume that these latter Committee statements constituted advice to ignore the provisions of a duly enacted law, we assume that these Committees believed that the Act simply was not applicable in this situation. But even under this interpretation of the Committees’ actions, we are unable to conclude that the Act has been in any respect amended or repealed. There is nothing in the appropriations measures, as passed, which states that the Tellico Project was to be completed irrespective of the requirements of the Endangered Species Act. These appropriations, in fact, represented relatively minor components of the lump-sum amounts for the entire TVA budget.35 To find a repeal of the Endangered Species Act under these circumstances would surely do violence to the “ ‘cardinal rule . . . that repeals by implication are not favored.’” Morton v. Mancari, 417 U. S. 535, 549 (1974), quoting Posadas N. National City Bank, 296 U. S. 497, 503 (1936). In Posadas this Court held, in no uncertain terms, that “the intention of the legislature to repeal must be clear and manifest.” Ibid. See Georgia v. Pennsylvania R. Co., 35 The Appropriations Acts did not themselves identify the projects for which the sums had been appropriated; identification of these projects requires reference to the legislative history. See n. 14, supra. Thus, unless a Member scrutinized in detail the Committee proceedings concerning the appropriations, he would have no knowledge of the possible conflict between the continued funding and the Endangered Species Act. 190 OCTOBER TERM, 1977 Opinion of the Court 437 U. S. 324 U. S. 439, 456-457 (1945) (“Only a clear repugnancy between the old . . . and the new [law] results in the former giving way . . .”); United States v. Borden Co., 308 U. S. 188, 198-199 (1939) (“[I]ntention of the legislature to repeal ‘must be clear and manifest’. . . . ‘[A] positive repugnancy [between the old and the new laws]’”); Wood v. United States, 16 Pet. 342, 363 (1842) (“[T]here must be a positive repugnancy ...”). In practical terms, this “cardinal rule” means that “[i]n the absence of some affirmative showing of an intention to repeal, the only permissible justification for a repeal by implication is when the earlier and later statutes are irreconcilable.” Mancari, supra, at 550. The doctrine disfavoring repeals by implication “applies with full vigor when . . . the subsequent legislation is an appropriations measure.” Committee for Nuclear Responsibility v. Seaborg, 149 U. S. App. D. C. 380, 382, 463 F. 2d 783, 785 (1971) (emphasis added); Environmental Defense Fund v. Froehlke, 473 F. 2d 346, 355 (CA8 1972). This is perhaps an understatement since it would be more accurate to say that the policy applies with even greater force when the claimed repeal rests solely on an Appropriations Act. We recognize that both substantive enactments and appropriations measures are “Acts of Congress,” but the latter have the limited and specific purpose of providing funds for authorized programs. When voting on appropriations measures, legislators are entitled to operate under the assumption that the funds will be devoted to purposes which are lawful and not for any purpose forbidden. Without such an assurance, every appropriations measure would be pregnant with prospects of altering substantive legislation, repealing by implication any prior statute which might prohibit the expenditure. Not only would this lead to the absurd result of requiring Members to review exhaustively the background of every authorization before voting on an appropriation, but it would flout the very rules the Congress carefully adopted to avoid TVA v. HILL 191 153 Opinion of the Court this need. House Rule XXI (2), for instance, specifically provides: “No appropriation shall be reported in any general appropriation bill, or be in order as an amendment thereto, for any expenditure not previously authorized by law, unless in continuation of appropriations for such public works as are already in progress. Nor shall any provision in any such bill or amendment thereto changing existing law be in order.” (Emphasis added.) See also Standing Rules of the Senate, Rule 16.4. Thus, to sustain petitioner’s position, we would be obliged to assume that Congress meant to repeal pro tanto § 7 of the Act by means of a procedure expressly prohibited under the rules of Congress. Perhaps mindful of the fact that it is “swimming upstream” against a strong current of well-established precedent, TVA argues for an exception to the rule against implied repealers in a circumstance where, as here, Appropriations Committees have expressly stated their “understanding” that the earlier legislation would not prohibit the proposed expenditure. We cannot accept such a proposition. Expressions of committees dealing with requests for appropriations cannot be equated with statutes enacted by Congress, particularly not in the circumstances presented by this case. First, the Appropriations Committees had no jurisdiction over the subject of endangered species, much less did they conduct the type of extensive hearings which preceded passage of the earlier Endangered Species Acts, especially the 1973 Act. We venture to suggest that the House Committee on Merchant Marine and Fisheries and the Senate Committee on Commerce would be somewhat surprised to learn that their careful work on the substantive legislation had been undone by the simple—and brief— insertion of some inconsistent language in Appropriations Committees’ Reports. 192 OCTOBER TERM, 1977 Opinion of the Court 437 U. S. Second, there is no indication that Congress as a whole was aware of TVA’s position, although the Appropriations Committees apparently agreed with petitioner’s views. Only recently, in SEC v. Sloan, 436 U. S. 103 (1978), we declined to presume general congressional acquiescence in a 34-year-old practice of the Securities and Exchange Commission, despite the fact that the Senate Committee having jurisdiction over the Commission’s activities had long expressed approval of the practice. Mr. Justice Rehnquist, speaking for the Court, observed that we should be “extremely hesitant to presume general congressional awareness of the Commission’s construction based only upon a few isolated statements in the thousands of pages of legislative documents.” Id., at 121. A fortiori, we should not assume that petitioner’s views—and the Appropriations Committees’ acceptance of them—were any better known, especially when the TVA is not the agency with primary responsibility for administering the Endangered Species Act. Quite apart from the foregoing factors, we would still be unable to find that in this case “the earlier and later statutes are irreconcilable,” Mancari, 417 U. 8., at 550; here it is entirely possible “to regard each as effective.” Id., at 551. The starting point in this analysis must be the legislative proceedings leading to the 1977 appropriations since the earlier funding of the dam occurred prior to the listing of the snail darter as an endangered species. In all successive years, TVA confidently reported to the Appropriations Committees that efforts to transplant the snail darter appeared to be successful; this surely gave those Committees some basis for the impression that there was no direct conflict between the Tellico Project and the Endangered Species Act. Indeed, the special appropriation for 1978 of $2 million for transplantation of endangered species supports the view that the Committees saw such relocation as the means whereby collision between Tellico and the Endangered Species Act could be avoided. It should also TVA v. HILL 193 153 Opinion of the Court be noted that the Reports issued by the Senate and House Appropriations Committees in 1976 came within a month of the District Court’s decision in this case, which hardly could have given the Members cause for concern over the possible applicability of the Act. This leaves only the 1978 appropriations, the Reports for which issued after the Court of Appeals’ decision now before us. At that point very little remained to be accomplished on the project; the Committees understandably advised TVA to cooperate with the Department of the Interior “to relocate the endangered species to another suitable habitat so as to permit the project to proceed as rapidly as possible.” H. R. Rep. No. 95-379, p. 11 (1977). It is true that the Committees repeated their earlier expressed “view” that the Act did not prevent completion of the Tellico Project. Considering these statements in context, however, it is evident that they “ ‘represent only the personal views of these legislators,’ ” and “however explicit, [they] cannot serve to change the legislative intent of Congress expressed before the Act’s passage.” Regional Rail Reorganization Act Cases, 419 U. S. 102, 132 (1974). (B) Having determined that there is an irreconcilable conflict between operation of the Tellico Dam and the explicit provisions of § 7 of the Endangered Species Act, we must now consider what remedy, if any, is appropriate. It is correct, of course, that a federal judge sitting as a chancellor is not mechanically obligated to grant an injunction for every violation of law. This Court made plain in Hecht Co. v. Bowles, 321 U. S. 321, 329 (1944), that “[a] grant of jurisdiction to issue compliance orders hardly suggests an absolute duty to do so under any and all circumstances.” As a general matter it may be said that “[s]ince all or almost all equitable remedies are discretionary, the balancing of equities and hardships is appropriate in almost any case as a guide to the chancellor’s discretion.” D. Dobbs, Remedies 52 (1973). Thus, in Hecht 194 OCTOBER TERM, 1977 Opinion of the Court 437 U. S. Co. the Court refused to grant an injunction when it appeared from the District Court findings that “the issuance of an injunction would have ‘no effect by way of insuring better compliance in the future’ and would [have been] ‘unjust’ to [the] petitioner and not ‘in the public interest.’ ” 321 U. S., at 326. But these principles take a court only so far. Our system of government is, after all, a tripartite one, with each branch having certain defined functions delegated to it by the Constitution. While “[i]t is emphatically the province and duty of the judicial department to say what the law is,” Marbury v. Madison, 1 Cranch 137,177 (1803), it is equally—and emphatically—the exclusive province of the Congress not only to formulate legislative policies and mandate programs and projects, but also to establish their relative priority for the Nation. Once Congress, exercising its delegated powers, has decided the order of priorities in a given area, it is for the Executive to administer the laws and for the courts to enforce them when enforcement is sought. Here we are urged to view the Endangered Species Act “reasonably,” and hence shape a remedy “that accords with some modicum of common sense and the public weal.” Post, at 196. But is that our function? We have no expert knowledge on the subject of endangered species, much less do we have a mandate from the people to strike a balance of equities on the side of the Tellico Dam. Congress has spoken in the plainest of words, making it abundantly clear that the balance has been struck in favor of affording endangered species the highest of priorities, thereby adopting a policy which it described as “institutionalized caution.” Our individual appraisal of the wisdom or unwisdom of a particular course consciously selected by the Congress is to be put aside in the process of interpreting a statute. Once the meaning of an enactment is discerned and its constitutionality determined, the judicial process comes to an end. We do not TVA v. HILL 195 153 Powell, J., dissenting sit as a committee of review, nor are we vested with the power of veto. The lines ascribed to Sir Thomas More by Robert Bolt are not without relevance here: “The law, Roper, the law. I know what’s legal, not what’s right. And I’ll stick to what’s legal. . . . I’m not God. The currents and eddies of right and wrong, which you find such plain-sailing, I can’t navigate, I’m no voyager. But in the thickets of the law, oh there I’m a forester.. . . What would you do? Cut a great road through the law to get after the Devil? . . . And when the last law was down, and the Devil turned round on you—where would you hide, Roper, the laws all being flat? . . . This country’s planted thick with laws from coast to coast—Man’s laws, not God’s—and if you cut them down . . . d’you really think you could stand upright in the winds that would blow them? ... Yes, I’d give the Devil benefit of law, for my own safety’s sake.” R. Bolt, A Man for All Seasons, Act I, p. 147 (Three Plays, Heinemann ed. 1967). We agree with the Court of Appeals that in our constitutional system the commitment to the separation of powers is too fundamental for us to pre-empt congressional action by judicially decreeing what accords with “common sense and the public weal.” Our Constitution vests such responsibilities in the political branches. Affirmed. Mr. Justice Powell, with whom Mr. Justice Blackmun joins, dissenting. The Court today holds that § 7 of the Endangered Species Act requires a federal court, for the purpose of protecting an endangered species or its habitat, to enjoin permanently the operation of any federal project, whether completed or substantially completed. This decision casts a long shadow over the operation of even the most important projects, serving 196 OCTOBER TERM, 1977 Powell, J., dissenting 437 U. S. vital needs of society and national defense, whenever it is determined that continued operation would threaten extinction of an endangered species or its habitat. This result is said to be required by the “plain intent of Congress” as well as by the language of the statute. In my view § 7 cannot reasonably be interpreted as applying to a project that is completed or substantially completed1 when its threat to an endangered species is discovered. Nor can I believe that Congress could have intended this Act to produce the “absurd result”—in the words of the District Court—of this case. If it were clear from the language of the Act and its legislative history that Congress intended to authorize this result, this Court would be compelled to enforce it. It is not our province to rectify policy or political judgments by the Legislative Branch, however egregiously they may disserve the public interest. But where the statutory language and legislative history, as in this case, need not be construed to reach such a result, I view it as the duty of this Court to adopt a permissible construction that accords with some modicum of common sense and the public weal. I Although the Court has stated the facts fully, and fairly presented the testimony and action of the Appropriations Committees relevant to this case, I now repeat some of what has been said. I do so because I read the total record as compelling rejection of the Court’s conclusion that Congress intended the Endangered Species Act to apply to completed or substantially completed projects such as the dam and reservoir project that today’s opinion brings to an end—absent relief by Congress itself. 1 Attorney General Bell advised us at oral argument that the dam had been completed, that all that remains is to “[c]lose the gate,” and to complete the construction of “some roads and bridges.” The “dam itself is finished. All the landscaping has been done .... [I]t is completed.” Tr. of Oral Arg. 18. TVA v. HILL 197 153 Powell, J., dissenting In 1966, Congress authorized and appropriated initial funds for the construction by the Tennessee Valley Authority (TVA) of the Tellico Dam and Reservoir Project on the Little Tennessee River in eastern Tennessee. The Project is a comprehensive water resource and regional development project designed to control flooding, provide water supply, promote industrial and recreational development, generate some additional electric power within the TVA system, and generally improve economic conditions in an economically depressed area “characterized by underutilization of human resources and outmigration of young people.”2 Construction began in 1967, and Congress has voted funds for the Project in every year since. In August 1973, when the Tellico Project was half completed, a new species of fish known as the snail darter3 was discovered in the portion of the Little Tennessee River that would be impounded behind Tellico Dam. The Endangered Species Act was passed the following December. 87 Stat. 884, 16 U. S. C. § 1531 et seq. (1976 ed.). More than a year later, in January 1975, respondents joined others in petitioning the Secretary of the Interior to list the snail darter as an endangered species. On November 10, 1975, when the Tellico Project was 75% completed, the Secretary placed the snail darter on the endangered list and concluded that the “proposed impoundment of water behind 2 Hearings on Public Works for Water and Power Development and Energy Research Appropriation Bill, 1977, before a Subcommittee of the House Committee on Appropriations, 94th Cong., 2d Sess., pt. 5, p. 261 (1976).’ 3 Although the snail darter is a distinct species, it is hardly an extraordinary one. Even icthyologists familiar with the snail darter have difficulty distinguishing it from several related species. App. 107, 131. Moreover, new species of darters are discovered in Tennessee at the rate of about 1 a year; 8 to 10 have been discovered in the last five years. Id., at 131. All told, there are some 130 species of darters, 85 to 90 of which are found in Tennessee, 40 to 45 in the Tennessee River system, and 11 in the Little Tennessee itself. Id., at 38 n. 7, 130-131. 198 OCTOBER TERM, 1977 Powell, J., dissenting 437U.S. the proposed Tellico Dam would result in total destruction of the snail darter’s habitat.” 40 Fed. Reg. 47506 (1975). In respondents’ view, the Secretary’s action meant that completion of the Tellico Project would violate § 7 of the Act, 16 U. S. C. § 1536 (1976 ed.): “All . . . Federal departments and agencies shall, in consultation with and with the assistance of the Secretary, utilize their authorities in furtherance of the purposes of this chapter by carrying out programs for the conservation of endangered species ... listed pursuant to section 1533 of this title and by taking such action necessary to insure that actions authorized, funded, or carried out by them do not jeopardize the continued existence of such endangered species and threatened species or result in the destruction or modification of habitat of such species which is determined by the Secretary ... to be critical.” TVA nevertheless determined to continue with the Tellico Project in accordance with the prior authorization by Congress. In February 1976, respondents filed the instant suit to enjoin its completion. By that time the Project was 80% completed. In March 1976, TVA informed the House and Senate Appropriations Committees about the Project’s threat to the snail darter and about respondents’ lawsuit. Both Committees were advised that TVA was attempting to preserve the fish by relocating them in the Hiwassee River, which closely resembles the Little Tennessee. It stated explicitly, however, that the success of those efforts could not be guaranteed.4 4 Hearings on Public Works for Water and Power Development and Energy Research Appropriations Bill, 1977, before a Subcommittee of the House Committee on Appropriations, 94th Cong., 2d Sess., pt. 5, pp. 261-262 (1976); Hearings on Public Works for Water and Power Development and Energy Research Appropriations for Fiscal Year 1977, before a Sub TVA v. HILL 199 153 Powell, J., dissenting In a decision of May 25, 1976, the District Court for the Eastern District of Tennessee held that “the Act should not be construed as preventing completion of the project.”5 419 F. Supp. 753, 755 n. 2. An opposite construction, said the District Court, would be unreasonable : “At some point in time a federal project becomes so near completion and so incapable of modification that a court of equity should not apply a statute enacted long after inception of the project to produce an unreasonable result. Arlington Coalition on Transportation v. Volpe, 458 F. 2d 1323, 1331-32 (4th Cir.), cert. den. 409 U. S. 1000 . . . (1972). Where there has been an irreversible and irretrievable commitment of resources by Congress to a project over a span of almost a decade, the Court should proceed with a great deal of circumspection.” Id., at 760. Observing that respondents’ argument, carried to its logical extreme, would require a court to enjoin the impoundment of committee of the Senate Committee on Appropriations, 94th Cong., 2d Sess., pt. 4, pp. 3096-3099 (1976). 5 The Court of Appeals interpreted the District Court opinion as holding that TVA’s continuation of the Tellico Project would violate the Act, but that the requested injunction should be denied on equitable grounds. 549 F. 2d 1064, 1069-1070 (CA6 1977). This interpretation of the District Court opinion appears untenable in light of that opinion’s conclusion that the Act could “not be construed as preventing completion of the project,” 419 F. Supp. 753, 755 n. 2 (1976) (emphasis added). Moreover, the District Court stated the issue in the case as whether “[it is] reasonable to conclude that Congress intended the Act to halt the Tellico Project at its present stage of completion.” Id., at 760. It concluded that the “Act should be construed in a reasonable manner to effectuate the legislative purpose,” ibid., and “that the Act does not operate in such a manner as to halt the completion of this particular project,” id., at 763. From all this, together with the District Court’s reliance on cases interpreting the National Environmental Policy Act, 42 U. S. C. § 4321 et seq., as inapplicable to substantially completed projects, see 419 F. Supp., at 760-761, it seems clear that District Judge Taylor correctly interpreted § 7 as inapplicable to the Tellico Project. 200 OCTOBER TERM, 1977 Powell, J., dissenting 437 U.S. water behind a fully completed dam if an endangered species were discovered in the river on the day before the scheduled impoundment, the District Court concluded that Congress could not have intended such a result.6 Accordingly, it denied the prayer for an injunction and dismissed the action. In 1975, 1976, and 1977, Congress, with full knowledge of the Tellico Project’s effect on the snail darter and the alleged violation of the Endangered Species Act, continued to appropriate money for the completion of the Project. In doing so, the Appropriations Committees expressly stated that the Act did not prohibit the Project’s completion, a view that Congress presumably accepted in approving the appropriations each year. For example, in June 1976, the Senate Committee on Appropriations released a report noting the District Court decision and recommending approval of TVA’s full budget request for the Tellico Project. The Committee observed further that it did “not view the Endangered Species Act as prohibiting the completion of the Tellico project at its advanced stage,” and it directed “that this project be completed as promptly as possible in the public interest.” 7 The appropriations bill was passed by Congress and approved by the President. The Court of Appeals for the Sixth Circuit nevertheless reversed the District Court in January 1977. It held that the Act was intended to create precisely the sort of dramatic conflict presented in this case: “Where a project is on-going and substantial resources have already been expended, the conflict between national incentives to conserve living things and the pragmatic momentum to complete the project on schedule is most incisive.” 549 F. 2d 1064, 1071. Judicial reso 6 The District Court found that $53 million out of more than $78 million then expended on the Project would be unrecoverable if completion of the dam were enjoined. 419 F. Supp., at 760. As more than $110 million has now been spent on the Project, it seems probable that abandonment of the dam would entail an even greater waste of tax dollars. 7 S. Rep. No. 94-960, p. 96 (1976). TVA v. HILL 201 153 Powell, J., dissenting lution of that conflict, the Court of Appeals reasoned, would represent usurpation of legislative power. It quoted the District Court’s statement that respondents’ reading of the Act, taken to its logical extreme, would compel a court to halt impoundment of water behind a dam if an endangered species were discovered in the river on the day before the scheduled impoundment. The Court of Appeals, however, rejected the District Court’s conclusion that such a reading was unreasonable and contrary to congressional intent, holding instead that “[c]onscientious enforcement of the Act requires that it be taken to its logical extreme.” Ibid. It remanded with instructions to issue a permanent injunction halting all activities incident to the Tellico Project that would modify the critical habitat of the snail darter. In June 1977, and after being informed of the decision of the Court of Appeals, the Appropriations Committees in both Houses of Congress again recommended approval of TVA’s full budget request for the Tellico Project. Both Committees again stated unequivocally that the Endangered Species Act was not intended to halt projects at an advanced stage of completion: “[The Senate] Committee has not viewed the Endangered Species Act as preventing the completion and use of these projects which were well under way at the time the affected species were listed as endangered. If the act has such an effect, which is contrary to the Committee’s understanding of the intent of Congress in enacting the Endangered Species Act, funds should be appropriated to allow these projects to be completed and their benefits realized in the public interest, the Endangered Species Act notwithstanding.”8 “It is the [House] Committee’s view that the Endangered Species Act was not intended to halt projects such 8 S. Rep. No. 95-301, p. 99 (1977). 202 OCTOBER TERM, 1977 Powell, J., dissenting 437 U. S. as these in their advanced stage of completion, and [the Committee] strongly recommends that these projects not be stopped because of misuse of the Act.”9 Once again, the appropriations bill was passed by both Houses and signed into law. II Today the Court, like the Court of Appeals below, adopts a reading of § 7 of the Act that gives it a retroactive effect and disregards 12 years of consistently expressed congressional intent to complete the Tellico Project. With all due respect, I view this result as an extreme example of a literalist19 construction, not required by the language of the Act and adopted without regard to its manifest purpose. Moreover, it ignores established canons of statutory construction. A The starting point in statutory construction is, of course, the language of § 7 itself. Blue Chip Stamps v. Manor Drug Stores, 421 U. S. 723, 756 (1975) (Powell, J., concurring). I agree that it can be viewed as a textbook example of fuzzy language, which can be read according to the “eye of the beholder.” 11 The critical words direct all federal agencies to take “such action [as may be] necessary to insure that actions authorized, funded, or carried out by them do not jeopardize the continued existence of . . . endangered species ... or result in the destruction or modification of [a critical] habitat of such species . . . .” Respondents—as did 9 H. R. Rep. No. 95-379, p. 104 (1977). 10 See Frank, Words and Music: Some Remarks on Statutory Interpretation, 47 Colum. L. Rev. 1259, 1263 (1947); Hand, The Speech of Justice, 29 Harv. L. Rev. 617, 620 (1916). 11 The purpose of this Act is admirable. Protection of endangered species long has been neglected. This unfortunate litigation—wasteful for taxpayers and likely in the end to be counterproductive in terms of respondents’ purpose—may have been invited by careless draftsmanship of otherwise meritorious legislation. TVA v. HILL 203 153 Powell, J., dissenting the Sixth Circuit—read these words as sweepingly as possible to include all “actions” that any federal agency ever may take with respect to any federal project, whether completed or not. The Court today embraces this sweeping construction. Ante, at 184-188. Under the Court’s reasoning, the Act covers every existing federal installation, including great hydroelectric projects and reservoirs, every river and harbor project, and every national defense installation—however essential to the Nation’s economic health and safety. The “actions” that an agency would be prohibited from “carrying out” would include the continued operation of such projects or any change necessary to preserve their continued usefulness.12 The only precondition, according to respondents, to thus destroying the usefulness of even the most important federal project in our country would be a finding by the Secretary of the Interior 12 Ante, at 184-188. At oral argument, respondents clearly stated this as their view of § 7: “QUESTION: . . . Do you think—it is still your position, as I understand it, that this Act, Section 7, applies to completed projects? I know you don’t think it occurs very often that there’ll be a need to apply it. But does it apply if the need exists? “MR. PLATER: To the continuation— “QUESTION: To completed projects. Take the Grand Coulee dam— “MR. PTjATER: Right. Your Honor, if there were a species there— “—it wouldn’t be endangered by the dam. “QUESTION: I know that’s your view. I’m asking you not to project your imagination— “MR. PLATER: I see, your Honor. “QUESTION: —beyond accepting my assumption. “MR. PLATER: Right. “QUESTION: And that was that an endangered species might turn up at Grand Coulee. Does Section 7 apply to it? “MR. PLATER: I believe it would, Your Honor. The Secretary of the Interior— “QUESTION: That answers my question. “MR. PLATER: Yes, it would.” Tr. of Oral Arg. 57-58. 204 OCTOBER TERM, 1977 Powell, J., dissenting 437 U. S. that a continuation of the project would threaten the survival or critical habitat of a newly discovered species of water spider or amoeba.13 “ [Frequently words of general meaning are used in a statute, words broad enough to include an act in question, and yet a consideration of the whole legislation, or of the circumstances surrounding its enactment, or of the absurd results which follow from giving such broad meaning to the words, makes it unreasonable to believe that the legislator intended to include the particular act.” Church of the Holy Trinity v. United States, 143 U. S. 457, 459 (1892).14 The 13 Under the Court’s interpretation, the prospects for such disasters are breathtaking indeed, since there are hundreds of thousands of candidates for the endangered list: “ ‘The act covers every animal and plant species, subspecies, and population in the world needing protection. There are approximately 1.4 million full species of animals and 600,000 full species of plants in the world. Various authorities calculate as many as 10% of them—some 200,000—may need to be listed as Endangered or Threatened. When one counts in subspecies, not to mention individual populations, the total could increase to three to five times that number.’ ” Keith Shreiner, Associate Director and Endangered Species Program Manager of the U. S. Fish and Wildlife Service, quoted in a letter from A. J. Wagner, Chairman, TVA, to Chairman, House Committee on Merchant Marine and Fisheries, dated Apr. 25, 1977, quoted in Wood, On Protecting an Endangered Statute: The Endangered Species Act of 1973, 37 Federal B. J. 25, 27 (1978). 14 Accord, e. g., United States v. American Trucking Assns., 310 U. S. 534, 543 (1940); Armstrong Co. v. Nu-Enamel Corp., 305 U. S. 315, 333 (1938); Sorrells v. United States, 287 U. S. 435, 446-448 (1932) (collecting cases); United States v. Ryan, 284 U. S. 167, 175 (1931). The Court suggests, ante, at 187 n. 33, that the precept stated in Church of the Holy Trinity was somehow undermined in Crooks v. Harrelson, 282 U. S. 55, 60 (1930). Only a year after the decision in Crooks, however, the Court declared that a “literal application of a statute which would lead to absurd consequences is to be avoided whenever a reasonable application can be given which is consistent with the legislative purpose.” Ryan, supra, at 175. In the following year, the Court expressly relied upon Church of the Holy Trinity on this very point. Sorrells, supra, at 448. The real difference between the Court and myself on this issue arises from our per TVA v. HILL 205 153 Powell, J., dissenting result that will follow in this case by virtue of the Court’s reading of § 7 makes it unreasonable to believe that Congress intended that reading. Moreover, §7 may be construed in a way that avoids an “absurd result” without doing violence to its language. The critical word in § 7 is “actions” and its meaning is far from “plain.” It is part of the phrase: “actions authorized, funded or carried out.” In terms of planning and executing various activities, it seems evident that the “actions” referred to are not all actions that an agency can ever take, but rather actions that the agency is deciding whether to authorize, to fund, or to carry out. In short, these words reasonably may be read as applying only to prospective actions, i. e., actions with respect to which the agency has reasonable decisionmaking alternatives still available, actions not yet carried out. At the time respondents brought this lawsuit, the Tellico Project was 80% complete at a cost of more than $78 million. The Court concedes that as of this time and for the purpose of deciding this case, the Tellico Dam Project is “completed” or “virtually completed and the dam is essentially ready for operation,” ante, at 156, 157-158. See n. 1, supra. Thus, under a prospective reading of § 7, the action already had been “carried out” in terms of any remaining reasonable decisionmaking power. Cf. National Wildlife Federation v. Coleman, 529 F. 2d 359, 363, and n. 5 (CA5), cert, denied sub nom. Boteler y. National Wildlife Federation, 429 U. S. 979 (1976). This is a reasonable construction of the language and also is supported by the presumption against construing statutes to give them a retroactive effect. As this Court stated in ceptions of the character of today’s result. The Court professes to find nothing particularly remarkable about the result produced by its decision in this case. Because I view it as remarkable indeed, and because I can find no hint that Congress actually intended it, see infra, at 207-210,1 am led to conclude that the congressional words cannot be given the meaning ascribed to them by the Court. 206 OCTOBER TERM, 1977 Powell, J., dissenting 437 U. S. United States Fidelity & Guaranty Co. v. United States ex rel. Struthers Wells Co., 209 U. S. 306, 314 (1908), the “presumption is very strong that a statute was not meant to act retrospectively, and it ought never to receive such a construction if it is susceptible of any other.” This is particularly true where a statute enacts a new regime of regulation. For example, the presumption has been recognized in cases under the National Environmental Policy Act, 42 U. S. C. § 4321 et seq., holding that the requirement of filing an environmental impact statement cannot reasonably be applied to projects substantially completed. E. g., Pizitz, Inc. v. Volpe, 467 F. 2d 208 (CA5 1972); Ragland v. Mueller, 460 F. 2d 1196 (CA5 1972); Greene County Planning Board v. FPC, 455 F. 2d 412, 424 (CA2), cert, denied, 409 U. S. 849 (1972). The Court of Appeals for the Fourth Circuit explained these holdings. “Doubtless Congress did not intend that all projects ongoing at the effective date of the Act be subject to the requirements of Section 102. At some stage of progress, the costs of altering or abandoning the project could so definitely outweigh whatever benefits that might accrue therefrom that it might no longer be ‘possible’ to change the project in accordance with Section 102. At some stage, federal action may be so ‘complete’ that applying the Act could be considered a ‘retroactive’ application not intended by the Congress.” Arlington Coalition on Transportation v. Volpe, 458 F. 2d 1323,1331, cert, denied sub nom. Fugate v. Arlington Coalition on Transportation, 409 U. S. 1000 (1972). Similarly under § 7 of the Endangered Species Act, at some stage of a federal project, and certainly where a project has been completed, the agency no longer has a reasonable choice simply to abandon it. When that point is reached, as it was in this case, the presumption against retrospective interpretation is at its strongest. The Court today gives no weight to that presumption. TVA v. HILL 207 153 Powell, J., dissenting B The Court recognizes that the first purpose of statutory construction is to ascertain the intent of the legislature. E. g., United States v. American Trucking Assns., 310 U. S. 534, 542 (1940).15 The Court’s opinion reviews at length the legislative history, with quotations from Committee Reports and statements by Members of Congress. The Court then ends this discussion with curiously conflicting conclusions. It finds that the “totality of congressional action makes it abundantly clear that the result we reach today [justifying the termination or abandonment of any federal project] is wholly in accord with both the words of the statute and the intent of Congress.” Ante, at 184. Yet, in the same paragraph, the Court acknowledges that “there is no discussion in the legislative history of precisely this problem.” The opinion nowhere makes clear how the result it reaches can be “abundantly” self-evident from the legislative history when the result was never discussed. While the Court’s review of the legislative history establishes that Congress intended to require governmental agencies to take endangered species into account in the planning and execution of their programs,16 there is not 15 Landis, A Note on “Statutory Interpretation,” 43 Harv. L. Rev. 886 (1930). 16 The quotations from the legislative history relied upon by the Court are reasonably viewed as demonstrating that Congress was thinking about agency action in prospective situations, rather than actions requiring abandonment of completed projects. For example, the Court quotes Representative Dingell’s statement as a highly pertinent interpretation of what the Conference bill intended. In the statement relied upon, ante, at 183-184, Representative Dingell said that Air Force bombing activities along the gulf coast of Texas, if found to endanger whooping cranes, would have to be discontinued. With respect to grizzly bears, he noted that they may or may not be endangered, but under the Act it will be necessary “to take action to see . . . that these bears are not driven to extinction.” The Court also predicates its holding as to legislative intent upon the provision in the Act that instructs federal agencies not to “take” endangered 208 OCTOBER TERM, 1977 Powell, J., dissenting 437 U. S. even a hint in the legislative history that Congress intended to compel the undoing or abandonment of any project or program later found to threaten a newly discovered species.17 If the relevant Committees that considered the Act, and the Members of Congress who voted on it, had been aware that the Act could be used to terminate major federal projects authorized years earlier and nearly completed, or to require the abandonment of essential and long-completed federal instal- species, meaning that no one is “to harass, harm, pursue, hunt, shoot, wound, kill, trap, capture, or collect” such life forms. Ante, at 184—185. The Court quotes, ante, at 184-185, n. 30, the Secretary of the Interior’s definition of the term “harm” to mean—among other things—any act which “annoy [s wild life] to such an extent as to significantly disrupt essential behavioral patterns, which include, but are not limited to, breeding, feeding or sheltering; significant environmental modification or degradation which has such effects is included within the meaning of ‘harm.’ ” 50 CFR § 17.3 (1976). Two observations are pertinent. -First, the reach of this regulation—which the Court accepts as authorized by the Act—is virtually limitless. All one would have to find is that the “essential behavioral patterns” of any living species as to breeding, feeding, or sheltering are significantly disrupted by the operation of an existing project. I cannot believe that Congress would have gone this far to imperil every federal project, however important, on behalf of any living species however unimportant, without a clear declaration of that intention. The more rational interpretation is consistent with Representative Dingell’s obvious thinking: The Act is addressed to prospective action where reasonable options exist; no thought was given to abandonment of completed projects. 17 The Senate sponsor of the bill, Senator Tunney, apparently thought that the Act was merely precatory and would not withdraw from the agency the final decision on completion of the project: “[A]s I understand it, after the consultation process took place, the Bureau of Public Roads, or the Corps of Engineers, would not be prohibited from building a road if they deemed it necessary to do so. “[A]s I read the language, there has to be consultation. However, the Bureau of Public Roads or any other agency would have the final decision as to whether such a road should be built. That is my interpretation of the legislation at any rate.” 119 Cong. Rec. 25689-25690 (1973). See also Sierra Club v. Froehlke, 534 F. 2d 1289, 1303-1304 (CA8 1976). TVA v. HILL 209 153 Powell, J., dissenting lations and edifices,18 we can be certain that there would have been hearings, testimony, and debate concerning consequences so wasteful, so inimical to purposes previously deemed important, and so likely to arouse public outrage. The absence of any such consideration by the Committees or in the floor debates indicates quite clearly that no one participating in the legislative process considered these consequences as within the intendment of the Act. As indicated above, this view of legislative intent at the time of enactment is abundantly confirmed by the subsequent congressional actions and expressions. We have held, properly, that post-enactment statements by individual Members of Congress as to the meaning of a statute are entitled to little or no weight. See, e. g., Regional Rail Reorganization Act Cases, 419 U. S. 102, 132 (1974). The Court also has recognized that subsequent Appropriations Acts themselves are not necessarily entitled to significant weight in determining whether a prior statute has been superseded. See United States v. Langston, 118 U. S. 389, 393 (1886). But these precedents are inapposite. There was no effort here to “bootstrap” a post-enactment view of prior legislation by isolated statements of individual Congressmen. Nor is this a case where Congress, without explanation or comment upon the statute in question, merely has voted apparently inconsistent finan 18 The initial proposed rulemaking under the Act made it quite clear that such an interpretation was not intended: “Neither [the Fish and Wildlife Service of the Department of the Interior] nor [the National Marine Fisheries Service of the Department of Commerce] intends that section 7 bring about the waste that can occur if an advanced project is halted. . . . The affected agency must decide whether the degree of completion and extent of public funding of particular projects justify an action that may be otherwise inconsistent with section 7.” 42 Fed. Reg. 4869 (1977). After the decision of the Court of Appeals in this case, however, the cjuoted language was withdrawn, and the agencies adopted the view of the court. 43 Fed. Reg. 870, 872, 875 (1978). 210 OCTOBER TERM, 1977 Powell, J., dissenting 437 U. S. cial support in subsequent Appropriations Acts. Testimony on this precise issue was presented before congressional committees, and the Committee Reports for three consecutive years addressed the problem and affirmed their understanding of the original congressional intent. We cannot assume—as the Court suggests—that Congress, when it continued each year to approve the recommended appropriations, was unaware of the contents of the supporting Committee Reports. All this amounts to strong corroborative evidence that the interpretation of § 7 as not applying to completed or substantally completed projects reflects the initial legislative intent. See, e. g., Fleming v. Mohawk Wrecking & Lumber Co., 331 U. S. Ill, 116 (1947); Brooks v. Dewar, 313 U. S. 354 (1941). Ill I have little doubt that Congress will amend the Endangered Species Act to prevent the grave consequences made possible by today’s decision. Few, if any, Members of that body will wish to defend an interpretation of the Act that requires the waste of at least $53 million, see n. 6, supra, and denies the people of the Tennessee Valley area the benefits of the reservoir that Congress intended to confer.19 There will be little sentiment to leave this dam standing before an empty reservoir, serving no purpose other than a conversation piece for incredulous tourists. But more far reaching than the adverse effect on the people of this economically depressed area is the continuing threat to the operation of every federal project, no matter how important to the Nation. If Congress acts expeditiously, as may be anticipated, the Court’s decision probably will have no lasting adverse consequences. But I had not thought it to be the province of this Court to force Congress into otherwise 19 The Court acknowledges, as it must, that the permanent injunction it grants today will require “the sacrifice of the anticipated benefits of the project and of many millions of dollars in public funds.” Ante, at 174. TVA v. HILL 211 153 Rehnquist, J., dissenting unnecessary action by interpreting a statute to produce a result no one intended. Mr. Justice Rehnquist, dissenting. In the light of my Brother Powell’s dissenting opinion, I am far less convinced than is the Court that the Endangered Species Act of 1973, 16 U. S. C. § 1531 et seq. (1976 ed.), was intended to prohibit the completion of the Tellico Dam. But the very difficulty and doubtfulness of the correct answer to this legal question convinces me that the Act did not prohibit the District Court from refusing, in the exercise of its traditional equitable powers, to enjoin petitioner from completing the Dam. Section 11 (g)(1) of the Act, 16 U. S. C. § 1540 (g) (1) (1976 ed.), merely provides that “any person may commence a civil suit on his own behalf ... to enjoin any person, including the United States and any other governmental instrumentality or agency . . . , who is alleged to be in violation of any provision of this chapter.” It also grants the district courts “jurisdiction, without regard to the amount in controversy or the citizenship of the parties, to enforce any such provision.” This Court had occasion in Hecht Co. v. Bowles, 321 U. S. 321 (1944), to construe language in an Act of Congress that lent far greater support to a conclusion that Congress intended an injunction to issue as a matter of right than does the language just quoted. There the Emergency Price Control Act of 1942 provided that “[u]pon a showing by the Administrator that [a] person has engaged or is about to engage in any [acts or practices violative of this Act] a permanent or temporary injunction, restraining order, or other order shall be granted without bond.” 56 Stat. 33 (emphasis added). But in Hecht this Court refused to find even in such language an intent on the part of Congress to require that a 212 OCTOBER TERM, 1977 Rehnquist, J., dissenting 437 U. S. district court issue an. injunction as a matter of course without regard to established equitable considerations, saying: “Only the other day we stated that ‘An appeal to the equity jurisdiction conferred on federal district courts is an appeal to the sound discretion which guides the determinations of courts of equity.’. . . The essence of equity jurisdiction has been the power of the Chancellor to do equity and to mould each decree to the necessities of the particular case. Flexibility rather than rigidity has distinguished it. The qualities of mercy and practicality have made equity the instrument for nice adjustment and reconciliation between the public interest and private needs as well as between competing private claims. We do not believe that such a major departure from that long tradition as is here proposed should be lightly implied. ... [I]f Congress desired to make such an abrupt departure from traditional equity practice as is suggested, it would have made its desire plain.” 321 U. S., at 329-330. Only by sharply retreating from the principle of statutory construction announced in Hecht Co. could I agree with the Court of Appeals’ holding in this case that the judicial enforcement provisions contained in § 11 (g)(1) of the Act require automatic issuance of an injunction by the district courts once a violation is found. I choose to adhere to Hecht Co.’s teaching: “A grant of jurisdiction to issue compliance orders hardly suggests an absolute duty to do so under any and all circumstances. We cannot but think that if Congress had intended to make such a drastic departure from the traditions of equity practice, an unequivocal statement of its purpose would have been made.” 321 U. S., at 329. Since the District Court possessed discretion to refuse injunctive relief even though it had found a violation of the Act, the TVA v. HILL 213 153 Rehnquist, J., dissenting only remaining question is whether this discretion was abused in denying respondents’ prayer for an injunction. Locomotive Engineers v. Missouri, K. & T. R. Co., 363 U. S. 528, 535 (1960). The District Court denied respondents injunctive relief because of the significant public and social harms that would flow from such relief and because of the demonstrated good faith of petitioner. As the Court recognizes, ante, at 193, such factors traditionally have played a central role in the decisions of equity courts whether to deny an injunction. See also 7 J. Moore, Federal Practice 65.18 [3] (1972); Yakus v. United States, 321 U. S. 414, 440-441 (1944). This Court has specifically held that a federal court can refuse to order a federal official to take specific action, even though the action might be required by law, if such an order “would work a public injury or embarrassment” or otherwise “be prejudicial to the public interest.” United States ex rel. Greathouse v. Dern, 289 U. S. 352, 360 (1933). Here the District Court, confronted with conflicting evidence of congressional purpose, was on even stronger ground in refusing the injunction. Since equity is “the instrument for nice adjustment and reconciliation between the public interest and private needs,” Hecht Co., supra, at 329-330, a decree in one case will seldom be the exact counterpart of a decree in another. See, e. g., Eccles v. People’s Bank, 333 U. S. 426 (1948); Penn Mutual Life Ins. Co. v. Austin, 168 U. S. 685 (1898). Here the District Court recognized that Congress, when it enacted the Endangered Species Act, made the preservation of the habitat of the snail darter an important public concern. But it concluded that this interest on one side of the balance was more than outweighed by other equally significant factors. These factors, further elaborated in the dissent of my Brother Powell, satisfy me that the District Court’s refusal to issue an injunction was not an abuse of its discretion. I therefore dissent from the Court’s opinion holding otherwise. 214 OCTOBER TERM, 1977 Syllabus 437 U. S. NATIONAL LABOR RELATIONS BOARD v. ROBBINS TIRE & RUBBER CO. CERTIORARI TO THE UNITED STATES COURT OF APPEALS FOR THE FIFTH CIRCUIT No. 77-911. Argued April 26, 1978—Decided June 15, 1978 After the National Labor Relations Board (NLRB) filed an unfair labor practice complaint against respondent employer, respondent requested, pursuant to the Freedom of Information Act (FOIA), that the NLRB make available prior to the hearing copies of all potential witnesses’ statements collected during the NLRB’s investigation. This request was denied on the ground that the statements were exempt from disclosure under, inter alia, Exemption 7 (A) of the FOIA, which provides that disclosure is not required of “investigatory records compiled for law enforcement purposes, but only to the extent that the production of such records . . . would interfere with enforcement proceedings.” Respondent then filed an action in District Court seeking disclosure of the statements and injunctive relief. That court held that Exemption 7 (A) did not apply because the NLRB did not claim that release of the statements would pose any unique or unusual danger of interference with the particular enforcement proceeding, and hence directed the NLRB to provide the statements for copying prior to any hearing. The Court of Appeals affirmed, holding that the NLRB had failed to sustain its burden of demonstrating the availability of Exemption 7 (A) because it had introduced no evidence that interference with the unfair labor practice proceeding in the form of witness intimidation was likely to occur in this particular case. Held: The Court of Appeals erred in holding that the NLRB was not entitled to withhold the witness statements under Exemption 7 (A). Pp. 220-243. (a) Exemption 7 (A)’s language does not support an interpretation that determination of “interference” under the Exemption can be made only on an individual, case-by-case basis, and, indeed, the language of Exemption 7 as a whole tends to suggest the contrary. Nor is such an interpretation supported by other portions of the FOIA providing for disclosure of segregable portions of records and for in camera review of documents, and placing the burden of justifying nondisclosure on the Government. Pp. 223-224. (b) Exemption 7 (A)’s legislative history indicates that Congress did not intend to prevent federal courts from determining that, with respect NLRB v. ROBBINS TIRE & RUBBER CO. 215 214 Syllabus to particular kinds of enforcement proceedings, disclosure of particular kinds of investigatory records while a case is pending would generally “interfere with enforcement proceedings,” and, more particularly, did not intend to overturn the NLRB’s longstanding rule against prehearing disclosure of witnesses’ statements. Pp. 224-236. (c) Witness statements in pending unfair labor practice proceedings are exempt from FOIA disclosure at least until completion of the NLRB’s hearing, since the release of such statements necessarily would involve the kind of harm that Congress believed would constitute an “interference” with NLRB enforcement proceedings—that of giving a party litigant earlier and greater access to the NLRB’s case than he would otherwise have. Thus, here the NLRB met its burden of demonstrating that disclosure of the witnesses’ statements in question “would interfere with enforcement proceedings,” since the dangers posed by premature release of the statements would involve precisely the kind of “interference with enforcement proceedings” that Exemption 7 (A) was designed to avoid, the most obvious risk of such “interference” being that employers or, in some cases, unions will coerce or intimidate employees and others who have given statements, in an effort to make them change their testimony or not testify at all. Pp. 236-242. 563 F. 2d 724, reversed. Marshall, J., delivered the opinion of the Court, in which Burger, C. J., and Stewart, White, Blackmun, Rehnquist, and Stevens, JJ., joined. Stevens, J., filed a concurring opinion, in which Burger, C. J., and Rehnquist, J., joined, post, p. 243. Powell, J., filed an opinion concurring in part and dissenting in part, in which Brennan, J., joined, post, p. 243. Carl L. Taylor argued the cause for petitioner. With him on the brief were Solicitor General McCree, John S. Irving, Norton J. Come, and Carol A. De Deo. William M. Earnest argued the cause for respondent. With him on the brief was Charles A. Poellnitz* *Briefs of amici curiae urging affirmance were filed by Stephen A. Bokat and Stanley T. Kaleczyc for the Chamber of Commerce of the United States; by Robert E. Williams, Douglas S. McDowell, and Frank C. Morris, Jr., for the Equal Employment Advisory Council; and by Alan B. Morrison for the Freedom of Information Clearinghouse. 216 OCTOBER TERM, 1977 Opinion of the Court 437U.S. Mr. Justice Marshall delivered the opinion of the Court. The question presented is whether the Freedom of Information Act (FOIA), 5 U. S. C. § 552 (1976 ed.), requires the National Labor Relations Board to disclose, prior to its hearing on an unfair labor practice complaint, statements of witnesses whom the Board intends to call at the hearing. Resolution of this question depends on whether production of the material prior to the hearing would “interfere with enforcement proceedings” within the meaning of Exemption 7 (A) of FOIA, 5 U. S. C. § 552 (b)(7)(A) (1976 ed.). I Following a contested representation election in a unit of respondent’s employees, the Acting Regional Director of the NLRB issued an unfair labor practice complaint charging respondent with having committed numerous violations of § 8 (a)(1) of the National Labor Relations Act (NLRA), 29 U. S. C. §158 (a)(1), during the pre-election period.1 A hearing on the complaint was scheduled for April 27, 1976. On March 31, 1976, respondent wrote to the Acting Regional Director and requested, pursuant to FOIA, that he make available for inspection and copying, at least seven days prior to the hearing, copies of all potential witnesses’ statements collected during the Board’s investigation. The Acting Regional Director denied this request on April 2, on the ground that this material was exempt from the disclosure requirements of 1 After investigating the union’s objections to the election, the Acting Regional Director not only issued an unfair labor practice charge but also recommended that seven challenged ballots be counted and, if they did not result in the union’s receiving a majority, that a hearing be held on certain of the union’s objections. The Board adopted the Acting Regional Director’s recommendations and, when a count of the challenged ballots failed to give the union a majority, the hearing on its objections to the election was consolidated with the hearing on the unfair labor practice charge. NLRB v. ROBBINS TIRE & RUBBER CO. 217 214 Opinion of the Court FOIA by various provisions of the Act, see 5 U. S. C. §§ 552 (b)(5), (7)(A), (C), (D) (1976 ed.). He placed particular reliance on Exemption 7 (A), which provides that disclosure is not required of “matters that are . . . investigatory records compiled for law enforcement purposes, but only to the extent that the production of such records would . . . interfere with enforcement proceedings.” 5 U. S. C. § 552 (b)(7)(A) (1976 ed.). Respondent appealed to the Board’s General Counsel. Before expiration of the 20-day period within which FOIA requires such appeals to be decided, 5 U. S. C. § 552 (a)(6) (A)(ii) (1976 ed.), respondent filed this action in the United States District Court for the Northern District of Alabama, pursuant to 5 U. S. C. § 552 (a)(4)(B) (1976 ed.). The complaint sought not only disclosure of the statements, but also a preliminary injunction against proceeding with the unfair labor practice hearing pending final adjudication of the FOIA claim and a permanent injunction against holding the hearing until the documents had been disclosed. At argument in the District Court, the Board contended, inter alia, that these statements were exempt from disclosure under Exemption 7 (A), because their production would “interfere” with a pending enforcement proceeding. The District Court held that, since the Board did not claim that release of the documents at issue would pose any unique or unusual danger of interference with this particular enforcement proceeding, Exemption 7 (A) did not apply. App. 62, 91. It therefore directed the Board to provide the statements for copying on or before April 22, 1976, or at least five days before any hearing where the person making the statement would be called as a witness. On the Board’s appeal, the United States Court of Appeals for the Fifth Circuit commenced its discussion by observing that while “[t]his is a [FOIA] case, ... it takes on the troubling coloration of a dispute about the discovery rights . . . 218 OCTOBER TERM, 1977 Opinion of the Court 437 U. S. in [NLRB] proceedings.” 563 F. 2d 724, 726 (1977).2 It concluded first that the legislative history of certain amendments to FOIA in 1974 demonstrated that Exemption 7 (A) was to be available only where there was a specific evidentiary showing of the possibility of actual interference in an individual case. Id., at 728. It therefore framed the Exemption 7 (A) issue as “whether pre-hearing disclosure of the contents of statements made by those prepared to testify in support of the Board’s case would actually ‘interfere’ with the Board’s case.” Id., at 727. In addressing this question, the Court of Appeals rejected the Board’s argument that the premature revelation of its case that would flow from production of the statements prior to the hearing was the kind of “interference” that would justify nondisclosure under the 1974 amendments. Reasoning that the only statements sought were those of witnesses whose prior statements would, under the Board’s own rules, be disclosed to respondent following the witnesses’ hearing testimony, the court also rejected as inapplicable the argument that potential witnesses would refrain from giving statements at all if prehearing disclosure were available. Id., at 729-731. Finally, while the Court of Appeals agreed with the Board that there was “some risk of interference ... in the form of witness intimidation” during the five-day period between disclosure and the hearing under the District Court’s order, it held that the Board had failed to sustain its burden of demonstrating the availability of Exemption 7 (A), because it had “introduced [no] evidence tending to show that this kind of intimidation” 2 As a preliminary matter, the Court of Appeals rejected the Board’s argument that the District Court had, in effect, granted an injunction against the Board proceeding, thereby erroneously refusing to require respondent to exhaust its administrative remedies. The court concluded that the District Court had not enjoined the Board proceeding, but had simply conditioned its right to proceed on the Board’s complying with respondent’s discovery request. 563 F. 2d, at 727. NLRB v. ROBBINS TIRE & RUBBER CO. 219 214 Opinion of the Court was in fact likely to occur in this particular case. Id., at 732. Rejecting the Board’s other claimed bases of exemption,3 the Court of Appeals affirmed. The Board filed a petition for a writ of certiorari, seeking review, inter alia,4 of the Exemption 7(A) ruling below, on the ground that the decision was in conflict with the weight of Circuit authority that had followed the lead of the United States Court of Appeals for the Second Circuit in Title Guarantee Co. v. NLRB, 534 F. 2d 484, cert, denied, 429 U. S. 834 (1976).5 There, on similar facts, the court held that 3 The Board argued that the statements were within the “attorney-work-product” privilege embodied in Exemption 5, which applies to “interagency or intra-agency memorandums or letters which would not be available by law to a party other than an agency in litigation with the agency.” 5 U. S. C. §552 (b)(5) (1976 ed.). The Court of Appeals concluded, however, that the witnesses’ statements were neither “memorandums” nor “letters” within the meaning of Exemption 5. The Board also suggested that the statements were covered by Exemption 7 (C) or (D), which apply to “investigatory records compiled for law enforcement purposes,” to the extent that their production would “constitute an unwarranted invasion of personal privacy [or] disclose the identity of a confidential source . . . .” The Court of Appeals rejected these claims, noting first that there is “nothing unusual in the nature of personal or family details in these affidavits” that would bring them within the scope of Exemption 7 (C). 563 F. 2d, at 733. With respect to Exemption 7 (D), the court concluded that the Board had failed to prove that the statements sought had been given only by one receiving an assurance of confidentiality, and that it could not so prove since the only statements sought were of witnesses scheduled to testify at the trial. Id., at 733-734. 4 The second question in the Board’s petition for certiorari seeks review of the holding below that Exemption 5 did not protect these witnesses’ statements from disclosure. See n. 3, supra. In light of our disposition of the case in the Board’s favor on the basis of our interpretation of Exemption 7, we have no occasion to address the Exemption 5 question. 5 Those decisions that have followed Title Guarantee include New England Medical Center Hospital n. NLRB, 548 F. 2d 377 (CAI 1976); Roger J. Au & Son v. NLRB, 538 F. 2d 80 (CA3 1976); NLRB v. Hardeman Garment Corp., 557 F. 2d 559 (CA6 1977); Abrahamson Chrysler-Plymouth, Inc. v. NLRB, 561 F. 2d 63 (CA7 1977); Harvey’s 220 OCTOBER TERM, 1977 Opinion of the Court 437 U. S. statements of employees and union representatives obtained in an NLRB investigation leading to an unfair labor practice charge were exempt from disclosure under Exemption 7 (A) until the completion of all reasonably foreseeable administrative and judicial proceedings on the charge. Rejecting the employer’s contention that the Board must make a particularized showing of likely interference in each individual case, the Second Circuit found that such interference would “necessarily” result from the production of the statements. 534 F. 2d, at 491. We granted certiorari to resolve the conflict among the Circuits on this important question of federal statutory law. 434 U. S. 1061 (1978). We now reverse the judgment of the Fifth Circuit. II We have had several occasions recently to consider the history and purposes of the original FOIA of 1966. See EPA v. Mink, 410 U. S. 73, 79-80 (1973); Renegotiation Board v. Bannercraft Clothing Co., 415 U. S. 1 (1974); NLRB v. Sears, Roebuck & Co., 421 U. S. 132 (1975); Department of Air Force v. Rose, 425 U. S. 352 (1976). As we have repeatedly emphasized, “the Act is broadly conceived,” EPA v. Mink, supra, at 80, and its “basic policy” is in favor of disclosure, Department of Air Force v. Rose, supra, at 361. In 5 U. S. C. § 552 (b) (1976 ed.), Congress carefully structured nine exemptions from the otherwise mandatory disclosure requirements in order to protect specified confidentiality and privacy Wagon Wheel, Inc. n. NLRB, 550 F. 2d 1139 (CA9 1976); Climax Molybdenum Co. n. NLRB, 539 F. 2d 63 (CAIO 1976). In a case involving witnesses’ statements obtained during a pending Equal Employment Opportunity Commission investigation, the Fourth Circuit has recently followed the basic approach of the Fifth Circuit in this case and rejected the Title Guarantee rationale. Charlotte-Mecklenburg Hospital Authority v. Perry, 571 F. 2d 195 (1978). NLRB v. ROBBINS TIRE & RUBBER CO. 221 214 Opinion of the Court interests.6 But unless the requested material falls within one of these nine statutory exemptions, FOIA requires that records and material in the possession of federal agencies be made available on demand to any member of the general public. Exemption 7 as originally enacted permitted nondisclosure of “investigatory files compiled for law enforcement purposes except to the extent available by law to a private party.” 80 6 Section 552 (b) in its entirety provides: “This section does not apply to matters that are— “(1)(A) specifically authorized under criteria established by an Executive order to be kept secret in the interest of national defense or foreign policy and (B) are in fact properly classified pursuant to such Executive order; “(2) related solely to the internal personnel rules and practices of an agency; “(3) specifically exempted from disclosure by statute (other than section 552b of this title), provided that such statute (A) requires that the matters be withheld from the public in such a manner as to leave no discretion on the issue, or (B) establishes particular criteria for withholding or refers to particular types of matters to be withheld; “(4) trade secrets and commercial or financial information obtained from a person and privileged or confidential; “(5) inter-agency or intra-agency memorandums or letters which would not be available by law to a party other than an agency in litigation with the agency; “(6) personnel and medical files and similar files the disclosure of which would constitute a clearly unwarranted invasion of personal privacy; “(7) investigatory records compiled for law enforcement purposes, but only to the extent that the production of such records would (A) interfere with enforcement proceedings, (B) deprive a person of a right to a fair trial or an impartial adjudication, (C) constitute an unwarranted invasion of personal privacy, (D) disclose the identity of a confidential source and, in the case of a record compiled by a criminal law enforcement authority in the course of a criminal investigation, or by an agency conducting a lawful national security intelligence investigation, confidential information furnished only by the confidential source, (E) disclose investigative techniques and procedures, or (F) endanger the life or physical safety of law enforcement personnel; “(8) contained in or related to examination, operating, or condition 222 OCTOBER TERM, 1977 Opinion of the Court 437 U. S. Stat. 251. In 1974, this exemption was rewritten to permit the nondisclosure of “investigatory records compiled for law enforcement purposes,” but only to the extent that producing such records would involve one of six specified dangers. The first of these, with which we are here concerned, is that production of the records would “interfere with enforcement proceedings.” The Board contends that the original language of Exemption 7 was expressly designed to protect existing NLRB policy forbidding disclosure of statements of prospective witnesses until after they had testified at unfair labor practice hearings. In its view, the 1974 amendments preserved Congress’ original intent to protect witness statements in unfair labor practice proceedings from premature disclosure, and were directed primarily at case law that had applied Exemption 7 too broadly to cover any material, regardless of its nature, in an investigatory file compiled for law enforcement purposes. The Board urges that a particularized, case-by-case showing is neither required nor practical, and that witness statements in pending unfair labor practice proceedings are exempt as a matter of law from disclosure while the hearing is pending. Respondent disagrees with the Board’s analysis of the 1974 amendments. It argues that the legislative history conclusively demonstrates that the determination of whether disclosure of any material would “interfere with enforcement proceedings” must be made on an individual, case-by-case basis. While respondent agrees that the statements sought reports prepared by, on behalf of, or for the use of an agency responsible for the regulation or supervision of financial institutions; or “(9) geological and geophysical information and data, including maps, concerning wells. “Any reasonably segregable portion of a record shall be provided to any person requesting such record after deletion of the portions which are exempt under this subsection.” NLRB v. ROBBINS TIRE & RUBBER CO. 223 214 Opinion of the Court here are “investigatory files compiled for law enforcement purposes,” and that they are related to an imminent enforcement proceeding, it argues that the Board’s failure to make a specific factual showing that their release would interfere with this proceeding defeats the Board’s Exemption 7 claim. A The starting point of our analysis is with the language and structure of the statute. We can find little support in the language of the statute itself for respondent’s view that determinations of “interference” under Exemption 7 (A) can be made only on a case-by-case basis. Indeed, the literal language of Exemption 7 as a whole tends to suggest that the contrary is true. The Exemption applies to: “investigatory records compiled for law enforcement purposes, but only to the extent that the production of such records would (A) interfere with enforcement proceedings, (B) deprive a person of a right to a fair trial or an impartial adjudication, (C) constitute an unwarranted invasion of personal privacy, (D) disclose the identity of a confidential source and, in the case of a record compiled by a criminal law enforcement authority in the course of a criminal investigation, or by an agency conducting a lawful national security intelligence investigation, confidential information furnished only by the confidential source, (E) disclose investigative techniques and procedures, or (F) endanger the life or physical safety of law enforcement personnel.” There is a readily apparent difference between subdivision (A.) and subdivisions (B), (C), and (D). The latter subdivisions refer to particular cases—“a person,” “an unwarranted invasion,” “a confidential source”—and thus seem to require a showing that the factors made relevant by the statute are present in each distinct situation. By contrast, since subdivision (A) speaks in the plural voice about “enforcement 224 OCTOBER TERM, 1977 Opinion of the Court 437 U. S. proceedings,” it appears to contemplate that certain generic determinations might be made. Respondent points to other provisions of FOIA in support of its interpretation. It suggests that, because FOIA expressly provides for disclosure of segregable portions of records and for in camera review of documents, and because the statute places the burden of justifying nondisclosure on the Government, 5 U. S. C. §§552 (a)(4)(B), (b) (1976 ed.), the Act necessarily contemplates that the Board must specifically demonstrate in each case that disclosure of the particular witness’ statement would interfere with a pending enforcement proceeding. We cannot agree. The in camera review provision is discretionary by its terms, and is designed to be invoked when the issue before the District Court could not be otherwise resolved; it thus does not mandate that the documents be individually examined in every case. Similarly, although the segregability provision requires that nonexempt portions of documents be released, it does not speak to the prior question of what material is exempt. Finally, the mere fact that the burden is on the Government to justify nondisclosure does not, in our view, aid the inquiry as to what kind of burden the Government bears. We thus agree with the parties that resolution of the question cannot be achieved through resort to the language of the statute alone. Accordingly, we now turn to an examination of the legislative history. B In originally enacting Exemption 7, Congress recognized that law enforcement agencies had legitimate needs to keep certain records confidential, lest the agencies be hindered in their investigations or placed at a disadvantage when it came time to present their cases. Foremost among the purposes of this Exemption was to prevent “harm [to] the Government’s case in court,” S. Rep. No. 813, 89th Cong., 1st Sess. (1965), reprinted in Freedom of Information Act Source Book, Sub- NLRB v. ROBBINS TIRE & RUBBER CO. 225 214 Opinion of the Court committee on Administrative Practice & Procedure, Senate Judiciary Committee, S. Doc. No. 93-82, p. 44 (1974) (hereinafter cited as 1974 Source Book), by not allowing litigants “earlier or greater access” to agency investigatory files than they would otherwise have, H. R. Rep. No. 1497, 89th Cong., 2d Sess. (1966), reprinted in 1974 Source Book 32. Indeed, in an unusual, post-passage reconsideration vote, the Senate modified the language of this Exemption specifically to meet Senator Humphrey’s concern that it might be construed to require disclosure of “statements of agency witnesses” prior to the time they were called on to testify in agency proceedings. Id., at 110. Senator Humphrey was particularly concerned that the initial version of the Exemption passed by the Senate might be “susceptible to the interpretation that once a complaint of unfair labor practice is filed by the General Counsel of the NLRB, access could be had to the statements of all witnesses, whether or not these statements are relied upon to support the complaint.” Ibid. He argued against this, noting that “(witnesses would be loath to give statements if they knew that their statements were going to be made known to the parties before the hearing,” id., at 111, and proposed adding another exemption to make clear that “statements of agency witnesses” would be exempt “until such witnesses are called to testify in an action or proceeding,” id., at 110.7 In direct response to what he described as Senator Humphrey’s “valu 7 Senator Humphrey’s amendment would have exempted from disclosure “statements of agency witnesses until such witnesses are called to testify in an action or proceeding and request is timely made by a private party for the production of relevant parts of such statements for purposes of cross examination.” 1974 Source Book 110. Colloquy on the floor made clear that the Senators thought it desirable to extend the so-called “Jencks” rule to agency proceedings, requiring the disclosure of witnesses’ statements only after the witnesses testified at the agency proceedings. See id., at 111. 226 OCTOBER TERM, 1977 Opinion of the Court 437 U. S. able suggestion,” Senator Long offered an amendment resulting in the version of Exemption 7 actually passed in 1966, which Senator Humphrey agreed would “take care of the situation.” Id., at 111. In light of this history, the Board is clearly correct that the 1966 Act was expressly intended to protect against the mandatory disclosure through FOIA of witnesses’ statements prior to an unfair labor practice proceeding. From one of the first reported decisions under FOIA, Barceloneta Shoe Corp. v. Compton, 271 F. Supp. 591 (PR 1967), through the time of the 1974 amendments, the courts uniformly recognized this purpose. Thus, in Wellman Industries, Inc. v. NLRB, 490 F. 2d 427 (CA4), cert, denied, 419 IL S. 834 (1974), the Court of Appeals held that affidavits obtained by an NLRB investigator during an inquiry into union objections to a representation election, which ultimately led to the filing of an unfair labor practice charge, were exempt from disclosure sought by the employer prior to the hearing on the complaint. It noted that employees might become unwilling to make “ ‘uninhibited and non-evasive statementfs]’ ” if disclosure were granted, 490 F. 2d, at 431, quoting NLRB v. National Survey Service, Inc., 361 F. 2d 199, 206 (CA7 1966), and emphasized that application of the exemption was “necessary in order to prevent premature disclosure of an investigation so that the Board can present its strongest case in court.” 490 F. 2d, at 431. Accord, NLRB v. Clement Bros. Co., 407 F. 2d 1027, 1031 (CA5 1969). C In 1974 Congress acted to amend FOIA in several respects. The move to amend was prompted largely by congressional disapproval of our decision in EPA v. Mink, 410 U. S. 73 (1973), regarding the availability of in camera review of classified documents. Congress was also concerned that administrative agencies were being dilatory in complying with the NLRB v. ROBBINS TIRE & RUBBER CO. 227 214 Opinion of the Court spirit of the Act and with court decisions interpreting FOIA to mandate disclosure of information to the public. See, e. g., Administration of the Freedom of Information Act, H. R. Rep. No. 92-1419 (1972),8 reprinted in 1975 Source Book 18, 79-80. As the amending legislation was reported out of the respective Committees, no change in Exemption 7 was recommended. See n. 14, infra. The 1974 amendment of Exemption 7 resulted instead from a proposal on the floor by Senator Hart during Senate debate. Senator Hart, in introducing his floor amendment, noted that the original intent of the 1966 Congress “was to prevent harm to the Government’s case in court by not allowing an opposing litigant earlier or greater access to investigatory files than he would otherwise have.” 1975 Source Book 332. He indicated his continued agreement with this purpose, id., at 333, but stated that recent court decisions had gone beyond this original intent by shielding from disclosure information that Congress had not intended to protect. Senator Hart emphasized his concern that “material cannot be and ought not be exempt merely because it can be categorized as an investigatory file compiled for law enforcement purposes.” Ibid. In colloquy with Senator Kennedy on the floor, Senator Hart stated specifically, id., at 349, that the amendment’s purpose was to respond to four decisions of the District of 8 This 89-page Report resulted from several days of hearings held by the House Government Operations Committee. Its focus was primarily on the procedural aspects of FOIA, and it manifested little discontent with the substantive disclosure and exemption requirements of the Act. See Administration of the Freedom of Information Act, H. R. Rep. No. 92-1419 (1972), reprinted in House Committee on Government Operations and Senate Committee on the Judiciary, Freedom of Information Act and Amendments of 1974 (Pub. L. 93-502) Source Book, 94th Cong., 1st Sess., 15 (Joint Comm. Print 1975) (identification of “major problem areas”) (hereinafter cited as 1975 Source Book). 228 OCTOBER TERM, 1977 Opinion of the Court 437 U. S. Columbia Circuit* commencing with the en banc decision in Weisberg v. United States Dept, of Justice, 160 U. S. App. D. C. 71, 489 F. 2d 1195 (1973), cert, denied, 416 U. S. 993 (1974). There, the plaintiff had sought disclosure of certain material in investigatory files relating to the assassination of President Kennedy, files that had been compiled 10 years before. Although the court acknowledged that no enforcement proceedings were then pending or contemplated, it held that all the agency need show to be entitled to withhold under Exemption 7 was that the records were investigatory in nature and had been compiled for law enforcement purposes. 160 U. S. App. D. C., at 74, 489 F. 2d, at 1198. The court adhered to this holding in Aspin v. Department of Defense, 160 U. S. App. D. C. 231, 237, 491 F. 2d 24, 30 (1973), stating that even “after the termination of investigation and enforcement proceedings,” material found in an investigatory file is entirely exempt. In Ditlow v. Brinegar, 161 U. S. App. D. C. 154, 494 F. 2d 1073 (1974), the court indicated that, after Weisberg, the only question before it was whether the requested material was found in an investigatory file compiled for law enforcement purposes. Finally, in Center for National Policy Review on Race and Urban Issues v. Weinberger, 163 U. S. App. D. C. 368, 502 F. 2d 370 (1974), the court held that the investigatory file exemption was available even if an enforcement proceeding 9 In response to Senator Hruska’s remarks that the amendment of Exemption 7 was likely to result in lawlessness due to ineffective law enforcement activities, Senator Kennedy stated that there had “been a gross misinterpretation of the actual words of the amendment and its intention.” 1975 Source Book 349. In order “for the record to be extremely clear,” he continued, what the amendment sought to do was “be specific about safeguarding . . . legitimate investigations ... by the Federal agencies.” He then asked Senator Hart whether its “impact and effect [was] to override” the four decisions discussed in the text. Ibid. The Conference Report on the 1974 amendments similarly states that the Exemption 7 amendment was designed to clarify Congress’ intent to disapprove of certain court decisions. Id., at 229. NLRB v. ROBBINS TIRE & RUBBER CO. 229 214 Opinion of the Court were neither imminent nor likely either at the time of the compilation or at the time disclosure was sought. These four cases, in Senator Hart’s view, erected a “stone wall” against public access to any material in an investigatory file. 1975 Source Book 332.10 Senator Hart believed that his amendment would rectify these erroneous judicial interpretations and clarify Congress’ original intent in two ways. First, by substituting the word “records” for “files,” it would make clear that courts had to consider the nature of the particular document as to which exemption was claimed, in order to avoid the possibility of 10 Although much of the debate on this amendment focused on the problems of access to “closed files,” two of the four D. C. Circuit cases involved files in still-pending investigations. Ditlow v. Brinegar; Center for National Policy Review of Race and Urban Issues v. Weinberger. But we do not understand the thrust of the Board’s argument to depend solely on its file being “open.” Instead, the Board points to the particular nature of these proceedings and the imminence of an actual adjudicatory proceeding on the charge. Since Senators Kennedy and Hart carefully explained the amendment’s purpose as being to eliminate a “wooden” and overly literal approach to the language of the Exemption, we do not read their reference to these two cases to mean that consideration of the pendency of an as-yet-unresolved charge to which the material sought relates is a factor that cannot be considered. Assuming, arguendo, that the references to Ditlow and W einberger mean that Congress disapproved of their holdings, as well as their reasoning, we do not think this disapproval undercuts our conclusion that the records sought here are protected In Ditlow, Exemption 7 was held to protect correspondence between automobile manufacturers and the National Highway Safety Traffic Administration concerning an apparently extended investigation of possible defects. Similarly, in Weinberger, Exemption 7 protection was extended to material in investigatory files of the Department of Health, Education, and Welfare relating to desegregation of the public schools in the North. In each of these cases, no enforcement proceeding was contemplated, much less imminent. Here, by contrast, an imminent adjudicatory proceeding is involved, in which the special dangers of interference with enforcement proceedings from prehearing disclosure are necessarily of a finite duration. 230 OCTOBER TERM, 1977 Opinion of the Court 437U.S. impermissible “commingling” by an agency’s placing in an investigatory file material that did not legitimately have to be kept confidential. Id., at 451. Second, it would explicitly enumerate the purposes and objectives of the Exemption, and thus require reviewing courts to “loo[k] to the reasons” for allowing withholding of investigatory files before making their decisions. Id., at 334. The “woode[n] and mechanica[l]” approach taken by the D. C. Circuit and disapproved by Congress would thereby be eliminated. Id., at 335 (remarks of Sen. Kennedy). As Congressman Moorhead explained to the House, the Senate amendment was needed to address “recent court decisions” that had applied the exemptions to investigatory files “even if they ha[d] long since lost any requirement for secrecy.” Id., at 378. Thus, the thrust of congressional concern in its amendment of Exemption 7 was to make clear that the Exemption did not endlessly protect material simply because it was in an investigatory file. Although, as indicated previously, no change in this section was reported out of committee, both Senate and House Committees had considered proposals to amend the provision.11 The Hart amendment was identical in respects 11 Both S. 1142 and H. R. 5425, as introduced in the 93d Congress, would have amended Exemption 7 to read as follows: “‘(7) investigatory records compiled for any specific law-enforcement purpose the disclosure of which is not in the public interest, except to the extent that— “'(A) any such investigatory records are available by law to a party other than an agency, or “ ‘(B) any such investigatory records are— “‘(i) scientific tests, reports, or data. “‘(ii) inspection reports of any agency which relate to health, safety, environmental protection, or “‘(iii) records which serve as a basis for any public policy statement made by any agency or officer or employee of the United States or which serve as a basis for rulemaking by any agency.’ ” See 1 Hearings on S. 858 et al. before the Subcommittee on Intergovem- NLRB v. ROBBINS TIRE & RUBBER CO. 23.1 214 Opinion of the Court here relevant to a proposal submitted during the hearings by the Administrative Law Division of the American Bar Association.12 2 Senate Hearings 158. The purpose of this proposal, mental Relations of the Senate Committee on Government Operations and the Subcommittees on Separation of Powers and Administrative Practice and Procedure of the Senate Committee on the Judiciary, 93d Cong., 1st Sess., 507 (1973) (hereinafter Senate Hearings); Hearings on H. R. 5425 et al. before a Subcommittee of the House Committee on Government Operations, 93d Cong., 1st Sess., 7 (1973) (hereinafter House Hearings). In addition, H. R. 4960 would have amended the Exemption with the following language: “investigatory records complied [sic] for law enforcement purposes, but only to the extent that production of such records would constitute (A) a genuine risk to enforcement proceedings. (B) a clearly unwarranted invasion of personal privacy, or (c) [sic] a threat to life.” House Hearings 12. The hearings on these proposals reflected Senator Hart’s concern that the courts were applying the language of the Exemption too literally and without regard for its underlying purposes. One witness from the American Civil Liberties Union, for example, emphasized that “[w]hat is being gotten at here ... is the old investigatory files, the dead files, the files that are yellowing in the Justice Department and the FBI . . . .” 2 Hearings on S. 1142 et al. before the Subcommittees on Administrative Practice and Procedure and Separation of Powers of the Senate Judiciary Committee and the Subcommittee on Intergovernmental Relations of the Senate Committee on Government Operations, 93d Cong., 1st Sess., 40 (1973) (hereinafter cited as 2 Senate Hearings) (statement of John Shattuck, ACLU staff counsel). See also House Hearings 28 (remarks of Rep. Erlenborn); id., at 78 (remarks of Rep. Horton). Senator Kennedy at one point proposed an amendment that would protect only actively pending cases, 2 Senate Hearings 2; the proposal was similar to a Justice Department proposal that would exempt all files in pending cases, and closed files but to a more limited extent. Id., at 227. 12 The ABA proposal exempted: “Investigatory records compiled for law enforcement purposes, but only to the extent that the production of such records would (A) interfere with enforcement proceedings, (B) deprive a person of a right to a fair trial or an impartial adjudication, (C) disclose the identity of an 232 OCTOBER TERM, 1977 Opinion of the Court 437U.S. according to the Chairman of the ABA Administrative Law Division, was to indicate that “with passage of time, . . . when the investigation is all over and the purpose and point of it has expired, it would no longer be an interference with enforcement proceedings and there ought to be disclosure.” Id., at 149. The tenor of this description of the statutory-language clearly suggests that the release of information in investigatory files prior to the completion of an actual, contemplated enforcement proceeding was precisely the kind of interference that Congress continued to want to protect against. Indeed, Senator Hart stated specifically that Exemption 7 (A) would apply “whenever the Government’s case in court— a concrete prospective law enforcement proceeding—would be harmed by the premature release of evidence or information . . . 1975 Source Book 333. That the 1974 Congress did not mean to undercut the intent of the 1966 Congress with respect to Senator Humphrey’s concern about interference with pending NLRB enforcement proceedings is apparent from the emphasis that both Senators Kennedy and Hart, the leaders in the debate on Exemption 7, placed on the fact that the amendment represented no radical departure from prior case law. While the D. C. Circuit decisions discussed above were repeatedly mentioned and condemned in the debates, nowhere do the floor debates or informer, or (D) disclose investigative techniques and procedures.” Id., at 158. The Hart amendment, proposed on the floor, incorporated most of this language and all of the language found in Exemption 7 (A): “Investigatory records compiled for law enforcement purposes, but only to the extent that the production of such records would (A) interfere with enforcement proceedings, (B) deprive a person of a right to a fair trial or an impartial adjudication or constitute a clearly unwarranted invasion of personal privacy, (C) disclose the identity of an informer, or (D) disclose investigative techniques and procedures.” After passing the Senate in this form, the amendment was modified to its present form, see supra, at 223, in Conference Committee. NLRB v. ROBBINS TIRE & RUBBER CO. 233 214 Opinion of the Court Committee Reports condemn the decisions holding that Exemption 7 protected witnesses’ statements in pending NLRB proceedings from disclosure, see supra, at 226, although Congress was clearly aware of these decisions.13 As Senator Hart concluded in his introductory remarks in support of the amendment: “This amendment is by no means a radical departure from existing case law under the Freedom of Information Act. Until a year ago the courts looked to the reasons for the seventh exemption before allowing the withholding of documents. That approach is in keeping with the intent of Congress and by this amendment we wish to reinstall it as the basis for access to information.” 1975 Source Book 334.14 13 Congress had prepared for its use a detailed case summary of the first 200 decisions under FOIA, see 1974 Source Book 116-183, a summary that included such cases as Barceloneta Shoe Corp. v. Compton, 271 F. Supp. 591 (PR 1967), and NLRB v. Clement Bros. Co., 407 F. 2d 1027 (CA5 1969), discussed supra, at 226. Wellman Industries, Inc. n. NLRB, 490 F. 2d 427 (CA4), cert, denied, 419 U. S. 834 (1974), followed the holdings of these two earlier decisions, but was apparently decided after the case summary was prepared and is not cited therein. 14 Senator Hart’s comments are in accord with Senator Kennedy’s explanation of why the Committees, after considering similar proposals to amend Exemption 7, see n. 11, supra, failed to report out an amendment. Senator Kennedy stated that the Committees had concluded that the courts were, by and large, giving that Exemption an appropriately narrow construction, and that any amendment of the Exemption would serve only to create confusion. See 1975 Source Book 335; S. Rep. No. 93-854 (1974), reprinted in 1975 Source Book 159. Senator Kennedy then stated that in light of the recent series of cases in the last 9-12 months, the “initial appraisal” of the case law had “turned out to be short lived.” Id., at 335. The Senator may have been mistaken as to the year of the first decision extending Exemption 7 protection automatically even in closed-file cases. In Frankel n. SEC, 460 F. 2d 813 (CA2 1972), over the strong dissent of Judge Oakes (the author of the later Title Guarantee opinion), the court held that material in an investigatory file was exempt from 234 OCTOBER TERM, 1977 Opinion of the Court 437 U. S. Senator Kennedy confirmed that “by accepting [this] amendment we will be reemphasizing and clarifying what the law presently requires.” Id., at 336. The emphasis that was placed on the limited scope of the amendment makes it more than reasonable to conclude that Congress intended to preserve existing law relating to NLRB proceedings—case law that had looked to the “reasons” for the Exemption and found them to be present where an unfair labor practice proceeding was pending and the documents sought were potential witnesses’ statements. D In the face of this history, respondent relies on Senator Hart’s floor statement that “it is only relevant” to determine whether an interference would result “in the context of the particular enforcement proceeding.” Id., at 333. Respondent argues that this statement means that in each case the court must determine whether the material of which disclosure is sought would actually reveal the Government’s case prematurely, result in witness intimidation, or otherwise create a demonstrable interference with the particular case. We believe that respondent’s reliance on this statement is misplaced. Although Congress could easily have required in so many words that the Government in each case show a particularized risk to its individual “enforcement proceed-in[g],” it did not do so;15 the statute, if anything, seems to draw a distinction in this respect between subdivision (A) and subdivisions (B), (C), and (D), see supra, at 223-224. Senator Hart’s words are ambiguous, moreover, and must be disclosure even though the investigation was complete and no enforcement proceedings were pending. Given the long history of cases construing NLRB witness statements as nondisclosable, see supra, at 226, we may assume that these decisions were not the object of the Senator’s amendment. 15 Indeed, Congress failed to enact proposals that might have had this effect. See n. 11, supra. NLRB v. ROBBINS TIRE & RUBBER CO. 235 214 Opinion of the Court read in light of his primary concern: that by extending blanket protection to anything labeled an investigatory file, the D. C. Circuit had ignored Congress’ original intent. His remarks plainly do not preclude a court from considering whether “particular” types of enforcement proceedings, such as NLRB unfair labor practice proceedings, will be interfered with by particular types of disclosure. Respondent also relies on President Ford’s message accompanying his veto of this legislation, and on the debate which led to Congress’ override of the veto. The President’s primary concern was with the congressional response to this Court’s decision in EPA v. Mink, 410 U. S. 73 (1973), concerning in camera judicial review of classified documents under Exemption 1. In addition, however, the President cited what in his view were the onerous new requirements of Exemption 7 that would require the Government to “prove . . .—separately for each paragraph of each document—that disclosure ‘would’ cause” a specific harm. 1975 Source Book 484. The leading supporters of the 1974 amendments, however, did not accept the President’s characterization; instead they indicated, with regard to the amended Exemption 7, that the President’s suggestions were “ludicrous,” id., at 406 (remarks of Rep. Moorhead), and that the “burden is substantially less than we would be led to believe by the President’s message,” id., at 450 (remarks of Sen. Hart). What Congress clearly did have in mind was that Exemption 7 permit nondisclosure only where the Government “specif [ies]” that one of the six enumerated harms is present, id., at 413 (remarks of Rep. Reid), and the court, reviewing the question de novo, agrees that one of those six “reasons” for nondisclosure applies. See supra, at 232. Thus, where an agency fails to “demonstrat [e] that the ... documents [sought] relate to any ongoing investigation or . . . would jeopardize any future law enforcement proceedings,” Exemption 7 (A) would not provide protection to the agency’s decision. 1975 236 OCTOBER TERM, 1977 Opinion of the Court 437 U. S. Source Book 440 (remarks of Sen. Kennedy). While the Court of Appeals was correct that the amendment of Exemption 7 was designed to eliminate “blanket exemptions” for Government records simply because they were found in investigatory files compiled for law enforcement purposes, we think it erred in concluding that no generic determinations of likely interference can ever be made. We conclude that Congress did not intend to prevent the federal courts from determining that, with respect to particular kinds of enforcement proceedings, disclosure of particular kinds of investigatory records while a case is pending would generally “interfere with enforcement proceedings.” Ill The remaining question is whether the Board has met its burden of demonstrating that disclosure of the potential witnesses’ statements at this time “would interfere with enforcement proceedings.” A proper resolution of this question requires us to weigh the strong presumption in favor of disclosure under FOIA against the likelihood that disclosure at this time would disturb the existing balance of relations in unfair labor practice proceedings, a delicate balance that Congress has deliberately sought to preserve and that the Board maintains is essential to the effective enforcement of the NLRA. Although reasonable arguments can be made on both sides of this issue, for the reasons that follow we conclude that witness statements in pending unfair labor practice proceedings are exempt from FOIA disclosure at least until completion of the Board’s hearing. Historically, the NLRB has provided little prehearing discovery in unfair labor practice proceedings and has relied principally on statements such as those sought here to prove its case. While the NLRB’s discovery policy has been criticized, the Board’s position that § 6 of the NLRA, 29 U. S. C. § 156, commits the formulation of discovery practice to its NLRB v. ROBBINS TIRE & RUBBER CO. 237 214 Opinion of the Court discretion has generally been sustained by the lower courts.16 A profound alteration in the Board’s trial strategy in unfair labor practice cases would thus be effectuated if the Board were required, in every case in which witnesses’ statements were sought under FOIA prior to an unfair labor practice proceeding, to make a particularized showing that release of these statements would interfere with the proceeding.17 Not only would this change the substantive discovery rules, but it would do so through mechanisms likely to cause substantial delays in the adjudication of unfair labor practice 16 Section 6 of the NLRA provides that the Board may “make such rules and regulations as may be necessary to carry out the provisions of this Act.” Most Circuits have held that prehearing discovery questions are committed to the Board’s discretion. See, e. g., NLRB v. Vapor Blast Mfg. Co., 287 F. 2d 402 (CA7 1961); Electromec Design & Development Co. v. NLRB, 409 F. 2d 631, 635 (CA9 1969); NLRB v. Interboro Contractors, Inc., 432 F. 2d 854, 858 (CA2 1970), cert, denied, 402 U. S. 915 (1971); D’YouviUe Manor, Lowell, Mass., Inc. v. NLRB, 526 F. 2d 3, 7 (CAI 1975); NLRB v. Valley Mold Co., 530 F. 2d 693, 695 (CA6 1976). Contrary to these authorities, the Fifth Circuit has held that “when good cause is shown [the NLRB] should permit discovery” in unfair labor practice proceedings. NLRB n. Rex Disposables, 494 F. 2d 588, 592 (1974), citing NLRB v. Safway Steel Scaffolds Co., 383 F. 2d 273 (CA5 1967), cert, denied, 390 U. S. 955 (1968) (relying on § 10 (b) of the NLRA, 29 U. S. C. § 160 (b)). This view of discovery in Board proceedings may have influenced the decision of the court below, since it noted that, under the Fifth Circuit’s approach to NLRB discovery, granting the FOIA request here might not have given the employer any more information about the Board’s case than it could otherwise have obtained. Since the court below did not rest on this ground, but instead indicated that the prospect of premature revelation of the Board’s case was not, of itself, an “interference” with enforcement proceedings, see supra, at 218, we intimate no view as to the validity of the Fifth Circuit’s approach to Board discovery. 17 If the Court of Appeals’ ruling below were not reversed, the Board anticipated that prehearing requests for witnesses’ statements under FOIA would be made by employer-respondents in virtually all unfair labor practice proceedings. See Pet. for Cert. 9. 238 OCTOBER TERM, 1977 Opinion of the Court 437 U. S. charges.18 In addition to having a duty under FOIA to provide public access to its processess, the NLRB is charged with the duty of effectively investigating and prosecuting violations of the labor laws. See 29 U. S. C. §§ 160,. 161. To meet its latter duty, the Board can be expected to continue to claim exemptions with regard to prehearing FOIA discovery requests, and numerous court contests will thereby ensue. Unlike ordinary discovery contests, where rulings are generally not appealable until the conclusion of the proceedings, an agency’s denial of a FOIA request is immediately reviewable in the district court, and the district court’s decision can then be reviewed in the court of appeals. The potential for delay and for restructuring of the NLRB’s routine adjudications of unfair labor practice charges from requests like respondent’s is thus not insubstantial. See n. 17, supra. In the absence of clear congressional direction to the contrary, we should be hesitant under ordinary circumstances to interpret an ambiguous statute to create such dislocations. Not only is such direction lacking, but Congress in 1966 was particularly concerned that premature production of witnesses’ statements in NLRB proceedings would adversely affect that agency’s ability to prosecute violations of the NLRA, and, as indicated above, the legislative history of the 1974 amendments affords no basis for concluding that Con- 18 We believe that delay of adjudicatory proceedings is a relevant factor, because Exemption 7 requires us to look at the interference that would flow from the “production,” and not merely the disclosure, of records. Since Congress had before it proposals that would have exempted only those investigatory records whose “disclosure” would create specified harms, see 1975 Source Book 338 (proposal of Assn, of Bar of City of New York), it is not unreasonable to attribute some significance to the use of the word “production” as defining the scope of activities from which the “interferences” justifying nondisclosure might flow. See also 5 U. S. C. § 552 (b) (6) (1976 ed.) (exempting personnel and medical files the “disclosure of which” would invade privacy) (emphasis added). NLRB v. ROBBINS TIRE & RUBBER CO. 239 214 Opinion of the Court gress at that time intended to create any radical departure from prior, court-approved Board practice. See supra, at 224^234. Our reluctance to override a long tradition of agency discovery, based on nothing more than an amendment to a statute designed to deal with a wholly different problem, is strengthened by our conclusion that the dangers posed by premature release of the statements sought here would involve precisely the kind of “interference with enforcement proceedings” that Exemption 7 (A) was designed to avoid. A The most obvious risk of “interference” with enforcement proceedings in this context is that employers or, in some cases, unions will coerce or intimidate employees and others who have given statements, in an effort to make them change their testimony or not testify at all. This special danger flowing from prehearing discovery in NLRB proceedings has been recognized by the courts for many years, see, e. g., NLRB v. Vapor Blast Mjg. Co., 287 F. 2d 402, 407 (CA7), cert, denied, 368 U. S. 823 (1961); NLRB v. National Survey Service, Inc., 361 F. 2d 199, 206 (CA7 1966); NLRB v. Lizdale Knitting Mills, 523 F. 2d 978, 980 (CA2 1975), and formed the basis for Senator Humphrey’s particular concern, see supra, at 225. Indeed, Congress recognized this danger in the NLRA itself, and provided in § 8 (a)(4) that it is an unfair labor practice for an employer “to discharge or otherwise discriminate against an employee because he has filed charges or given testimony under this subchapter.” 29 U. S. C. § 158 (a)(4). See NLRB v. Scrivener, 405 U. S. 117, 121 (1972). Respondent’s argument that employers will be deterred from improper intimidation of employees who provide statements to the NLRB by the possibility of a § 8 (a) (4) charge misses the point of Exemption 7 (A); the possibility of deterrence arising from post hoc disciplinary action is no substitute for a prophylactic 240 OCTOBER TERM, 1977 Opinion of the Court 437U.S. rule that prevents the harm to a pending enforcement proceeding which flows from a witness’ having been intimidated.19 The danger of witness intimidation is particularly acute with respect to current employees—whether rank and file, supervisory, or managerial—over whom the employer, by virtue of the employment relationship, may exercise intense leverage. Not only can the employer fire the employee, but job assignments can be switched, hours can be adjusted, wage and salary increases held up, and other more subtle forms of influence exerted. A union can often exercise similar authority over its members and officers. As the lower courts have recognized, due to the “peculiar character of labor litigation[,] the witnesses are especially likely to be inhibited by fear of the employer’s or—in some cases—the union’s capacity for reprisal and harassment.” Roger J. Au & Son, Inc. v. NLRB, 538 F. 2d 80, 83 (CA3 1976). Accord, NLRB v. Hardeman Garment Corp., 557 F. 2d 559 (CA6 1977). While the risk of intimidation (at least from employers) may be somewhat diminished with regard to statements that are favorable to the employer, those known to have already given favorable statements are then subject to pressure to give even more favorable testimony. Furthermore, both employees and nonemployees may be reluctant to give statements to NLRB investigators at all, absent assurances that unless called to testify in a hearing, their statements will be exempt from disclosure until the unfair labor practice charge has been adjudicated. Such reluctance may flow less from a witness’ desire to maintain complete confidentiality—the concern of Exemption 7 (D)—than from an all too familiar unwillingness to “get too involved” unless 19 Respondent argues that the relatively small percentage of unfair labor practice charges filed under § 8 (a) (4) demonstrates that the Board’s justifications for its nondisclosure rules are illusory. Brief for Respondent 38. But the small percentage may reflect the effectiveness of the intimidation, rather than any lack thereof. It may also reflect the success of the Board’s current policy. NLRB v. ROBBINS TIRE & RUBBER CO. 241 214 Opinion of the Court absolutely necessary. Since the vast majority of the Board’s unfair labor practice proceedings are resolved short of hearing, without any need to disclose witness statements, those currently giving statements to Board investigators can have some assurance that in most instances their statements will not be made public (at least until after the investigation and any adjudication is complete).20 The possibility that a FOIA-induced change in the Board’s prehearing discovery rules will have a chilling effect on the Board’s sources cannot be ignored.21 In short, prehearing disclosure of witnesses’ statements would involve the kind of harm that Congress believed would constitute an “interference” with NLRB enforcement proceedings: that of giving a party litigant earlier and greater access to the Board’s case than he would otherwise have. As the lower courts have noted, even without intimidation or harassment a suspected violator with advance access to the Board’s case could “ ‘construct defenses which would permit violations to go unremedied.’ ” New England Medical Center Hosp. v. NLRB, 548 F. 2d 377, 382 (CAI 1976), quoting Title Guarantee Co. v. NLRB, 534 F. 2d, at 491. This possibility arises simply from the fact of prehearing disclosure of any witness 20 According to the Board, 94% of all unfair labor practice charges filed are resolved short of hearing; in the remaining 6% that go to hearing, many potential witnesses are not actually called to testify, since their testimony is cumulative. Brief for Petitioner 17-18, n. 4. 21 Respondent argues that the Court of Appeals was correct in concluding that this danger is nonexistent with respect to a witness scheduled to testify, since the Board under its own discovery rules will turn over those statements once the witness has actually testified. See 29 CFR § 102.118 (b) (1) (1977). This argument falters, first, on the fact that only those portions of the witness’ statements relating to his direct examination or the issues raised in the pleadings are disclosed under the Board’s discovery rules. In addition, to uphold respondent’s FOIA request would doubtless require the Board in many cases to turn over statements of persons whom it did not actually call at the adjudicatory hearings. See n. 20, supra. 242 OCTOBER TERM, 1977 Opinion of the Court 437 U. S. statements, whether the witness is favorable or adverse, employee or nonemployee. While those drafting discovery rules for the Board might determine that this “interference” is one that should be tolerated in order to promote a fairer decisionmaking process, that is not our task in construing FOIA. B The basic purpose of FOIA is to ensure an informed citizenry, vital to the functioning of a democratic society, needed to check against corruption and to hold the governors accountable to the governed. 1974 Source Book 38; see also NLRB v. Sears, Roebuck & Co., 421 U. S., at 152. Respondent concedes that it seeks those statements solely for litigation discovery purposes, and that FOIA was not intended to function as a private discovery tool, see Renegotiation Board v. Bannercrajt Clothing Co., 415 U. S., at 22.22 Most, if not all, persons who have sought prehearing disclosure of Board witnesses’ statements have been in precisely this posture— parties respondent in Board proceedings.23 Since we are dealing here with the narrow question whether witnesses’ statements must be released five days prior to an unfair labor practice hearing, we cannot see how FOIA’s purposes would be defeated by deferring disclosure until after the Government has “presented its case in court.” Cf. NLRB v. Sears, Roebuck <& Co., supra, at 159-160. Consideration of the underlying policy of the Act as it applies in this case thus reinforces our conclusion that Congress, having given no explicit attention to this problem in its 1974 legislation, could not have intended to overturn the NLRB’s longstanding rule against prehearing disclosure of 22 Tr. of Oral Arg. 31, 34. 23 This is not to suggest that respondent’s rights are in any way diminished by its being a private litigant, but neither are they enhanced by respondent’s particular, litigation-generated need for these materials. See EPA v. Mink, 410 U. S. 73, 86 (1973). NLRB v. ROBBINS TIRE & RUBBER CO. 243 214 Opinion of Powell, J. witness statements. It was Congress’ understanding, and it is our conclusion, that release of such statements necessarily “would interfere” in the statutory sense with the Board’s “enforcement proceedings.” We therefore conclude that the Court of Appeals erred in holding that the Board was not entitled to withhold such statements under Exemption 7 (A). The judgment of the Court of Appeals is, accordingly, Reversed. Mr. Justice Stevens, with whom The Chief Justice and Mr. Justice Rehnquist join, concurring. The “act of meddling in” a process is one of Webster’s accepted definitions of the word “interference.”* A statute that authorized discovery greater than that available under the rules normally applicable to an enforcement proceeding would “interfere” with the proceeding in that sense. The Court quite correctly holds that the Freedom of Information Act does not authorize any such interference in Labor Board enforcement proceedings. Its rationale applies equally to any enforcement proceeding. On that understanding, I join the opinion. Mr. Justice Powell, with whom Mr. Justice Brennan joins, concurring in part and dissenting in part. I join the Court’s opinion to the extent that it holds that Exemption 7 (A) of the Freedom of Information Act (Act or FOIA), 5 U. S. C. § 552 (b)(7)(A) (1976 ed.), permits the federal courts to determine that “with respect to particular kinds of enforcement proceedings, disclosure of particular kinds of investigatory records while a case is pending would generally ‘interfere with enforcement proceedings.’ ” Ante, at 236. *One of the definitions of “interference” is “the act of meddling in or hampering an activity or process.” Webster’s Third New International Dictionary 1178 (1961). 244 OCTOBER TERM, 1977 Opinion of Powell, J. 437U.S. I endorse the limitation of such “generic determinations of likely interference,” ibid., to “an imminent adjudicatory proceeding” that is “necessarily of a finite duration,” ante, at 229 n. 10. I also agree that the National Labor Relations Board (Board) has sustained its burden of justifying nondisclosure of statements by current employees that are unfavorable to their employer’s cause in an unfair labor practice proceeding against that employer. But I cannot accept the Court’s approval of the application of the Board’s rule of nondisclosure to all witness statements, unless and until a witness gives direct testimony before an administrative law judge. And I disagree with the Court’s apparent interpretation of Exemption 7 (A) as providing no “earlier or greater access” to records than that available under the discovery rules that an agency chooses to promulgate. See concurring opinion of Mr. Justice Stevens, ante, p. 243. There is no persuasive evidence that Congress in 1974 intended to authorize federal agencies to withhold all FOIA-requested material in pending proceedings by invoking restrictive rules of discovery promulgated under their “housekeeping” rulemaking authority.1 I The starting point is the language of Exemption 7 (A). Congress provided for the nondisclosure of “investigatory records compiled for law enforcement purposes, but only to the extent that the production of such records would (A) interfere with enforcement proceedings . .. .” Establishing a presumption of disclosure, the Act “does not authorize withholding of information or limit the availability of records to the public, 1 The FOIA was enacted in 1966 as a remedy for agency “housekeeping” rules that had restricted unduly public information about the operations of Government. See H. R. Rep. No. 1497, 89th Cong., 2d Sess., 3-6 (1966); S. Rep. No. 813, 89th Cong., 1st Sess., 3, 5 (1965). Congress intended to establish legislative standards for nondisclosure of official information and to empower the federal courts to review claims of agency non-compliance with those standards. NLRB v. ROBBINS TIRE & RUBBER CO. 245 214 Opinion of Powell, J. except as specifically stated in this section.” 5 U. S. C. § 552 (c) (1976 ed.). Moreover, “[a]ny reasonably segregable portion of a record shall be provided to any person requesting such record after deletion of the portions which are exempt under this subsection.” § 552 (b). The language of Exemption 7 (A) simply cannot be squared with the Court’s conclusion that “giving a party litigant earlier and greater access to the Board’s case than he would otherwise have” under agency rules is “the kind of harm that Congress believed would constitute an ‘interference’ with NLRB enforcement proceedings . . . .” Ante, at 241. It is instructive to compare the 1974 amendment with the 1966 version of the “investigatory files” exemption. Exemption 7 as originally enacted permitted nondisclosure of “investigatory files compiled for law enforcement purposes except to the extent available by law to a private party.” 80 Stat. 251.2 Congress in 1974 abandoned the language that keyed the standard of disclosure to that available generally to private litigants.3 In its place, Congress prescribed that the withholding of investigatory records be based upon one or more of six specified types of harm. That change in language suggests that Congress may have intended a more focused inquiry into the likelihood of harm resulting from disclosure of investigatory records than was possible under a standard defining the scope of disclosure in terms of an agency’s rules of discovery.4 2 The exception clause first appeared in a post-passage amendment on the floor of the Senate to accommodate Senator Humphrey’s desire that the investigatory files exemption shield from disclosure prehearing statements of NLRB witnesses. 110 Cong. Rec. 17666-17668 (1964), reprinted in Subcommittee on Administrative Practice and Procedure, Senate Judiciary Committee, Freedom of Information Act Source Book, S. Doc. No. 93-82, pp. 109, 111 (1974). 3 Congress did not disturb similar language contained in Exemption 5, 5 U. S. C. § 552 (b) (5) (1976 ed.). See EPA v. Mink, 410 U. S. 73, 85-86 (1973). 4 Although the Committee Reports and the debates appear to be silent oh 246 OCTOBER TERM, 1977 Opinion of Powell, J. 437U.S. The Court of Appeals in this case observed that “[i]f the mere fact that one could not have obtained the document in private discovery were enough, the Board would have made naught of the requirement that nondisclosure be permitted ‘only to the extent that . . . production . . . would . . . interfere’ in some way” with the proceeding. 563 F. 2d 724, 730 (CA5 1977). There also is force to the Court of Appeals’ view that such a standard is unworkable because the courts have not accorded uniform recognition to the Board’s authority to deny rights of discovery to litigants in proceedings before it. Moreover, that court noted that a discovery standard may require an assessment of the particular needs of the FOIA plaintiff when the Act mandates release of information “to any person,” 5 U. S. C. § 552 (a)(3) (1976 ed.), incorporating the principle that “anyone’s case is as strong (or as weak) as the point, the deletion of the exception clause has been viewed as evidence of an intent to broaden the scope of disclosure under Exemption 7. See Fuselier & Moeller, NLRB Investigatory Records: Disclosure Under the Freedom of Information Act, 10 U. Rich. L. Rev. 541, 546 (1976). Others have attached little significance to this change in language. See Attorney General’s Memorandum on the 1974 Amendments to the Freedom of Information Act 5 n. 3 (1975), reprinted in House Committee on Government Operations and Senate Committee on the Judiciary, Freedom of Information Act and Amendments of 1974 (Pub. L. 93-502) Source Book, 94th Cong., 1st Sess., 515 (Joint Comm. Print 1975) (hereinafter cited as 1975 Source Book); Ellsworth, Amended Exemption 7 of the Freedom of Information Act, 25 Am. U. L. Rev. 37, 45-46, n. 39 (1975). In an early decision, the clause had been construed “to limit persons charged with violations of federal regulatory statutes to the discovery available to persons charged with violations of federal criminal law.” Bristol-Myers Co. v. FTC, 138 U. S. App. D. C. 22, 26, 424 F. 2d 935, 939, cert, denied, 400 U. S. 824 (1970). See Note, The Freedom of Information Act: A Seven-Year Assessment, 74 Colum. L. Rev. 895, 948, and n. 291 (1974). The proviso later was relied on by the same court to deny disclosure to an FOIA litigant who would not have been a “party” engaged in litigation with an agency. See Weisberg v. United States Dept, of Justice, 160 U. S. App. D. C. 71, 79 n. 15, 489 F. 2d 1195, 1203 n. 15 (1973) (en banc), cert, denied, 416 U. S. 993 (1974). NLRB v. ROBBINS TIRE & RUBBER CO. 247 214 Opinion of Powell, J. anyone else’s.” 563 F. 2d, at 730; see NLRB v. Sears, Roebuck & Co., 421 U. S. 132,143 n. 10 (1975). Nor does the legislative history provide more than ambiguous support for the Court’s reading. There are statements by Senator Hart, the principal sponsor of the Exemption 7 amendment, that appear favorable. But these statements, made on the floor of the Senate, are not very clear on the point in dispute. Thus while Senator Hart noted that the original intent of the 1966 provision was to deny “an opposing litigant earlier or greater access to investigative files than he would otherwise have,” 120 Cong. Rec. 17033 (1974), reprinted in 1975 Source Book 332, he also said that Exemption 7 (A) “would apply whenever the Government’s case in court— a concrete prospective enforcement proceeding—would be harmed by the premature release of evidence or information not in the possession of known or potential defendants.” Id., at 333. If Exemption 7 (A) were intended to authorize nondisclosure in every pending proceeding, it is doubtful that Senator Hart would have spoken in terms of “whenever the Government’s case in court . . . would be harmed by the premature release . . . .” I find equally unilluminating statements to the effect that the 1974 amendment was not intended to work “a radical departure from existing case law under the Freedom of Information Act.” Id., at 334 (remarks of Sen. Hart). The one point that emerges with clarity is that Congress intended that “the courts look ... to the reasons for the seventh exemption before allowing the withholding of documents.” Ibid. But it is difficult to reconcile that principle with the underlying rationale of the Court’s opinion that “the release of information in investigatory files prior to the completion of an actual, contemplated enforcement proceeding was precisely the kind of interference that Congress continued to want to protect against.” Ante, at 232. Congress had before it several proposals that would have drawn the line between 248 OCTOBER TERM, 1977 Opinion of Powell, J. 437U.S. files in “pending or contemplated” proceedings and files in “closed” cases. These were not adopted.5 One must assume that a deliberate policy decision informed Congress’ rejection of these alternatives in favor of the language presently contained in Exemption 7 (A). Moreover, as the Court notes, ante, at 229 n. 10, at least two of the decisions of the Court of Appeals for the District of Columbia Circuit that Congress intended to overrule “involved files in still-pending investigations.” See Ditlow v. Brinegar, 161 U. S. App. D. C. 154, 494 F. 2d 1073, cert, denied, 419 U. S. 974 (1974); Center for National Policy Review v. Weinberger, 163 U. S. App. D. C. 368, 502 F. 2d 370 (1974).6 Senator Hart stated that these cases, among others, were wrongly decided because the courts failed to approach the disclosure issue “on a balancing basis, which is exactly what this amendment seeks to do.” 1975 Source Book 349. The Court’s approach in this case also is in tension with Congress’ most recent amendment to the Act. Congress in 1976 overturned our decision in FAA Administrator v. Robertson, 422 U. S. 255 (1975), which held that Exemption 3, 5 U. S. C. § 552 (b) (3), should not be interpreted to disturb a broad delegation of authority to an agency to withhold information from the public. Pub. L. No. 94-409, § 5 (b)(3), 90 Stat. 1247. Congress tightened the standard for Exemp- 5 See 2 Hearings on S. 1142 et al. before the Subcommittees on Administrative Practice and Procedure and Separation of Powers of the Senate Judiciary Committee and the Subcommittee on Intergovernmental Relations of the Senate Committee on Government Operations, 93d Cong., 1st Sess., 2 (1973) (Sen. Kennedy); id., at 227 (Dept, of Justice), discussed in 1975 Source Book 339; id., at 338 (Committee on Federal Legislation of the Assn, of Bar of City of New York). 6 In Center for National Policy Review, for example, the court held that Exemption 7 permitted the Secretary of Health, Education, and Welfare to resist disclosure of the material of 22 “open and active” files involving agency review of public school discrimination practices in northern localities. NLRB v. ROBBINS TIRE & RUBBER CO. 249 214 Opinion of Powell, J. tion 3 “to exempt only material required to be withheld from the public by any statute establishing particular criteria or referring to particular types of information,” and rejected Robertson, which was viewed as “afford [ing] the FAA Administrator cart[e] blanche to withhold any information he pleases . . . .” H. R. Rep. No. 94-880, pt. 1, p. 23 (1976). The Court’s ruling today appears to afford an agency similar carte blanche authority to withhold witness statements in investigatory files, at least during the pendency of an enforcement proceeding. The Court appropriately recognizes the danger that FOIA claims are “likely to cause substantial delays in the adjudication of unfair labor practice charges.” Ante, at 237-238. But Congress had a right to insist, as I believe it did in the 1974 legislation, that nondisclosure of investigatory records be grounded in one of the six specific categories of harm set out in Exemption 7, even though litigation may ensue over disputed claims of exemption. II As the Court demonstrates, the congressional requirement of a specific showing of harm does not prevent determinations of likely harm with respect to prehearing release of particular categories of documents. The statements of the Act’s sponsors in urging an override of President Ford’s veto of the 1974 amendments shed light on this point. The President’s message to Congress explained that “confidentiality would not be maintained if many millions of pages of FBI and other investigatory law enforcement files would be subject to compulsory disclosure at the behest of any person unless the Government could prove to a court—separately for each paragraph of each document—that disclosure ‘would’ cause a type of harm specified in the amendment.” 1975 Source Book 484. The bill’s proponents discounted the President’s concern. See id., at 405-406 (remarks of Rep. Moorhead); id., at 451-452 250 OCTOBER TERM, 1977 Opinion of Powell, J. 437U.S. (remarks of Sen. Hart). As then Attorney General Levi observed: “This legislative history suggests that denial can be based upon a reasonable possibility, in view of the circumstances, that one of the six enumerated consequences would result from disclosure.” Attorney General’s Memorandum on the 1974 Amendments to the Freedom of Information Act 13 (1975), reprinted in 1975 Source Book 523. A In my view, the Board has demonstrated a “reasonable possibility” that harm will result from prehearing disclosure of statements by current employees that are damaging to their employer’s case in an unfair labor practice proceeding. The Courts of Appeals have recognized with virtual unanimity that due to the “peculiar character of labor litigation[,] the witnesses are especially likely to be inhibited by fear of the employer’s or—in some cases—the union’s capacity for reprisal and harassment.” Roger J. Au & Son, Inc. v. NLRB, 538 F. 2d 80, 83 (CA3 1976).7 The “delicate” relationship between employer and employee—or between union and employee-member—suggests that “ [t]he labor case is peculiarly susceptible to employer [or union] retaliation, coercion, or influence to the point that it can be concluded that there is no need for an express showing of interference in each case to justify giving effect to the exemption contained in Section 7 (A) in 7 The Court of Appeals in this case also recognized that “there may be some risk of interference with Board proceedings in the form of witness intimidation from harassment of an employee-witness during the five days prior to the hearing, done in an effort to silence him or dilute the nature of his testimony.” 563 F. 2d 724, 732 (CA5 1977). It determined, however, that the Board had failed to introduce any evidence tending to show that such intimidation was likely, and declined to accept the Board’s assertion that “in every case the potential for intimidation is so great as to require nondisclosure of all statements and affidavits.” Id., at 732-733 (emphasis supplied). NLRB v. ROBBINS TIRE & RUBBER CO. 251 214 Opinion of Powell, J. Labor Board proceedings.” Climax Molybdenum Co. v. NLRB, 539 F. 2d 63, 65 (CAIO 1976). The Board knows from experience that an employer or a union charged with an unfair labor practice often can exercise special influence—either through threats or promises of benefit—over employees or members whose welfare and opportunity for advancement depend on remaining in the good graces of the charged party. Accordingly, the Court has construed § 8 (a)(4) of the National Labor Relations Act, as amended, 61 Stat. 140, 29 U. S. C. § 158 (a)(4), to protect employees who give written sworn statements to a Board field examiner even when they do not file a charge or testify at a formal hearing on the charge. NLRB v. Scrivener, 405 U. S. 117 (1972).8 Although the Board may be able to impose post hoc sanctions for interference with its witnesses, see 29 U. S. C. §§ 158 (a)(4) and 162; 18 U. S. C. § 1505 (1976 ed.), these remedies cannot safeguard fully the integrity of ongoing unfair labor practice proceedings. Intimidation or promise of benefit may be subtle and not susceptible of proof. As the Board cannot proceed without a charge filed by knowledgeable individuals, see Nash v. Florida Industrial Comm’n, 389 U. S. 235, 238 (1967), many instances of interference could go undetected. Even if interference is detected and a complaint is filed, appropriate sanctions often cannot be imposed until after the initial unfair labor practice proceeding has terminated. Moreover, as the Court notes, many employees, mindful of the 8 The Court’s substantive labor law rulings have “takefn] into account the economic dependence of the employees on their employers, and the necessary tendency of the former, because of that relationship, to pick up intended implications of the latter that might be more readily dismissed by a more disinterested ear.” NLRB v. Gissel Packing Co., 395 U. S. 575, 617 (1969); see Textile Workers v. Darlington Co., 380 U. 8. 263 (1965); NLRB v. Exchange Parts Co., 375 U. S. 405 (1964). Similar considerations apply to statements made or inducements offered by labor unions. See, e. g., NLRB v. Savair Mjg. Co., 414 U. S. 270 (1973). 252 OCTOBER TERM, 1977 Opinion of Powell, J. 437U.S. Board’s prehearing settlement practice, may be willing to cooperate with the Board because they know that their identity will not be revealed and they will not be called to give public testimony adverse to their employer’s interest unless such a course is absolutely necessary. Until the Board’s view here is proved unfounded, as an empirical matter, I agree that the danger of altered testimony— through intimidation or promise of benefit—provides sufficient justification for the judgment that disclosure of unfavorable statements by current employees prior to the time when they are called to give testimony before an administrative law judge, “would interfere with enforcement proceedings . ...”9 B But the Court holds that all “witness statements in pending unfair labor practice proceedings are exempt from FOIA disclosure at least until completion of the Board’s hearing. . . .” Ante, at 236. I find no warrant for that sweeping conclusion in the expressed intention of the 93d Congress. Exemption 7 (A) requires that the Board demonstrate a reasonable possibility that disclosure would “interfere with enforcement proceedings . . . .” In my view, absent a particularized showing of likely interference, statements of all witnesses—other than current employees in proceedings against employers (or union members in proceedings against unions)—are subject to the statutory presumption in favor of disclosure. In contrast to the situation of current employees or union members, there simply is no basis for presuming a particular likelihood of employer interference with union representatives or others not employed by the charged party, or, in a proceeding against a union, of union interference with employer representatives and other nonmembers of the union or the bargaining unit. Simi- $ Similarly, the Board may protect against prehearing disclosure statements by union members and employees unfavorable to the union’s cause in an unfair labor practice proceeding. NLRB v. ROBBINS TIRE & RUBBER CO. 253 214 Opinion of Powell, J. larly, I am unwilling to presume interference with respect to disclosure of favorable statements by current employees, and would require the Board to show a reasonable possibility of employer reprisal. See Temple-Eastex, Inc. v. NLRB, 410 F. Supp. 183,186 (ED Tex. 1976). I do not read the Act to authorize agencies to adopt or adhere to nonstatutory rules10 barring all prehearing disclosure of investigatory records. The Court reasons, ante, at 241, that such disclosure—which is deemed “premature” only because it is in advance of the time of release set by the agency—will enable “suspected violators ... to learn the Board’s case in advance and frustrate the proceedings or construct defenses which would permit violations to go unremedied . . . .” Title Guarantee Co. v. NLRB, 534 F. 2d 484, 491 (CA2), cert, denied, 429 U. S. 834 (1976). This assumption is not only inconsistent with the congressional judgment expressed in the Federal Rules of Civil Procedure that “trial by ambush,” New England Medical Center Hosp. v. NLRB, 548 F. 2d 377, 387 (CAI 1976); Capital Cities Communications, Inc. v. NLRB, 409 F. Supp. 971, 977 (ND Cal. 1976), well may disserve the cause of truth, but it also threatens to undermine the Act’s overall presumption of disclosure, at least during the pendency of enforcement proceedings.11 10 It may be that criminal law enforcement agencies will be able to resist pretrial disclosure of witness statements on the theory that the Jencks Act, 18 U. S. C. §3500 (a) (1976 ed.), falls within the terms of Exemption 3 of the Act; see supra, at 248-249. 111 do not construe the Court’s ruling today to authorize agencies to withhold disclosure of materials generated in closed or otherwise inactive proceedings, absent a particularized showing of harm, even though the Board itself would like this authority. Brief for Petitioner 33 n. 17. The Board has advanced this view in the Courts of Appeals with some success. Compare New England Medical Center Hosp. v. NLRB, 548 F. 2d, at 385-386 (records generated in a related, inactive investigation held protected against disclosure), with Poss v. NLRB, 565 F. 2d 654, 657 (CAIO 1977) (statements taken in an investigation that ended in a decision not to issue a complaint held not protected). 254 OCTOBER TERM, 1977 Opinion of Powell, J. 437U.S. There may be exceptional cases that would permit the Board to withhold all witness statements for the duration of an unfair labor practice proceeding. Such a situation could arise where prehearing revelation would divulge incompletely developed information which, if prematurely disclosed, may interfere with the proceedings before the Board, or where the facts of a case suggest a strong likelihood that the charged party will attempt to interfere with any and all of the Board’s witnesses. The Act requires, however, that the Board convince a federal court that there is a reasonable possibility of this kind of interference.12 I would reverse the judgment of the Court of Appeals to the extent that it requires prehearing disclosure of unfavorable statements by respondent’s current employees, but affirm as to any remaining statements in dispute.13 12 In light of my view of the limits of Exemption 7 (A), I reach the Board’s alternative argument that the witness affidavits in dispute are protected against disclosure by Exemption 5, 5 U. S. C. § 552 (b) (5) (1976 ed.). That section provides that the Act does not apply to “interagency or intra-agency memorandums or letters which would not be available by law to a party other than any agency in litigation with the agency. ...” I agree generally with the analysis of the Court of Appeals that the purpose of this Exemption is to protect agency litigation strategy and decisionmaking processes, and not to incorporate fully the “work product” privilege recognized in Hickman v. Taylor, 329 U. S. 495 (1947), and Fed. Rule Civ. Proc. 26 (b)(3). Our decision in NLRB n. Sears, Roebuck & Co., 421 U. S. 132, 154-155, 159-160 (1975), provides support for this view. In this case, by contrast, the Board does not suggest that the witness affidavits in question are anything other than verbatim transcripts of statements made by witnesses to Board personnel. 13 There is no need for a remand in this case, cf. Harvey’s Wagon Wheel, Inc. v. NLRB, 550 F. 2d 1139, 1143 (CA9 1976), for the Board conceded in the District Court that “[t] here’s nothing unique in Board proceedings in these statements . . . .” App. 91. FIRST FEDERAL S. & L. v. MASSACHUSETTS TAX COMM’N 255 Syllabus FIRST FEDERAL SAVINGS & LOAN ASSOCIATION OF BOSTON et al. v. TAX COMMISSION OF MASSACHUSETTS et al. APPEAL FROM THE SUPREME JUDICIAL COURT OF MASSACHUSETTS No. 77-334. Argued March 21, 1978—Decided June 15, 1978 Appellants brought suit in a Massachusetts court challenging the State’s power to impose an excise tax on federal savings and loan associations as measured by their net operating income, claiming that the tax violates § 5 (h) of the Home Owners’ Loan Act of 1933, which provides that no tax on a federal savings and loan association shall be “greater than that imposed” by the State on similar local thrift and home financing institutions. Appellants claimed that the state tax on their net operating income exceeds that imposed on similar local institutions because the deduction available under the state tax statute for “minimum additions to its guaranty fund or surplus required by law or the ap; propriate federal and state supervisory authorities” is generally lower for federal savings and loan associations than for similar state savings institutions. Appellants also contended that because the Massachusetts tax does not apply to credit unions, which, appellants maintained, are “similar” to federal savings and loan associations, the associations are entitled to the credit unions’ exemptions. The Supreme Judicial Court of Massachusetts upheld the statute. Held: 1. The Massachusetts tax is not discriminatory on its face. The amount of the deduction depends on varying regulatory practices as to the reserves that must be maintained, but a tax is not invalid because it recognizes that state and federal regiilations may differ. Nor does the record show any discrimination in fact, or in statutory purpose (federal reserve requirements were as high as the State’s when the tax was enacted). Pp. 257-260. 2. Credit unions are not “similar” to federal savings and loan associations within the meaning of § 5 (h), as is clear not only from distinctions between the two under both federal and state law but also from the fact that Massachusetts savings banks and cooperative banks are more competitive with federal associations than credit unions are. Congress recognized that States might classify their own institutions in various ways, as Massachusetts has done in excluding credit unions from a large 256 OCTOBER TERM, 1977 Opinion of the Court 437 U. S. classification that includes state institutions more closely resembling the federal associations. Pp. 260-262. 372 Mass. 478, 363 N. E. 2d 474, affirmed. Stevens, J., delivered the opinion of the Court, in which Burger, C. J., and Brennan, Stewart, White, Marshall, Powell, and Rehnquist, JJ., joined. Blackmun, J., filed an opinion concurring in part and dissenting in part, post, p. 263. Chester M. Howe argued the cause for appellants. With him on the briefs was Maxwell D. Solet. S.'Stephen Rosenfeld, Assistant Attorney General of Massachusetts, argued the cause for appellees. With him on the brief were Francis X. Bellotti, Attorney General, and John E. Bowman, Jr., and Margot Botsford, Assistant Attorneys General.* Mr. Justice Stevens delivered the opinion of the Court. This appeal challenges the power of the State of Massachusetts to impose a tax on federal savings and loan associations. Relying on a federal law forbidding States to tax federal associations more heavily than “similar” state institutions, appellants contend that the State’s tax discriminates against federal associations because: (1) the state institutions subject to the tax are allowed a larger deduction for required additions to reserves than federal associations, and (2) the state tax does not apply to credit unions, which appellants believe to be “similar” to federal savings and loan associations. In the Home Owners’ Loan Act of 1933, Congress authorized the creation of federally chartered savings and loan associations. 48 Stat. 128. Section 5 (h) of that Act, as amended, 76 Stat. 984, 12 U. S. C. § 1464 (h) (1976 ed.), provides: “No State, county, municipal, or local taxing authority * Solicitor General McCree, Assistant Attorney General Ferguson, Stuart A. Smith, and David English Carmack filed a brief for the United States as amicus curiae urging reversal. FIRST FEDERAL S. & L. v. MASSACHUSETTS TAX COMM’N 257 255 Opinion of the Court shall impose any tax on such associations or their franchise, capital, reserves, surplus, loans, or income greater than that imposed by such authority on other similar local mutual or cooperative thrift and home financing institutions.” As enacted in 1966, the Massachusetts statute imposed an excise tax, measured by deposits and income, on state cooperative banks, state savings banks, and state and federal savings and loan associations. 1966 Mass. Acts, ch. 14, § 11. In 1973, the deposits aspect of the tax was invalidated as discriminatory. United States v. State Tax Comm’n, 481 F. 2d 963 (CAI 1973). See n. 3, infra. The present case, brought in state court in 1975, challenges the income aspect of the tax. It was presented on stipulated facts to the Supreme Judicial Court of Massachusetts, which upheld the statute. 372 Mass. 478, 363 N. E. 2d 474 (1977). We affirm. I The state tax statute allows a financial institution to deduct from its taxable income any “minimum additions ... to its guaranty fund or surplus required by law or the appropriate federal and state supervisory authorities.” Mass. Gen. Laws Ann., ch. 63, § 11 (b) (West Supp. 1977). As might be expected, the reserves required by state and federal regulators are not precisely the same. Before 1970, each federal association was required to adopt a charter providing for a minimum reserve equal to 10% of the association’s capital. See 12 CFR §544.1 (1977). This reserve was as large as, or larger than, the reserves that Massachusetts required its institutions to maintain.1 In 1970, federal associations were allowed 1 Massachusetts savings banks must set aside 7^% of deposits. Mass. Gen. Laws Ann., ch. 168, § 58 (West 1971). State cooperative banks must reserve 10% of their assets. Ch. 170, § 38. The reserve requirement for state savings and loan associations is not spelled out by statute. Cf. ch. 93, § 34 (West Supp. 1977). 258 OCTOBER TERM, 1977 Opinion of the Court 437U.S. to delete the reserve provision from their charters, a change that dropped their reserve requirement to 5% of checking and savings account balances. 35 Fed. Reg. 4044 (1970) ; 12 CFR §§ 544.8 (c)(1), 563.13 (1977); 12 U. S. C. § 1726 (b) (1976 ed.). More than three-quarters of the federal associations in Massachusetts adopted the change within a few months of the new regulation, and all but four have now amended their charters. The new requirement is lower than those set for state institutions. For this reason, the federal associations argue, their tax deductions are smaller than those of state institutions; they contend that this disparity in deductions is the sort of discrimination that has been proscribed by federal law. Section 5 (h) of the Home Owners’ Loan Act of 1933 “unequivocally bars discriminatory state taxation of the Federal Savings and Loan Associations.” Laurens Federal Savings & Loan Assn. v. South Carolina Tax Common, 365 U. S. 517, 523. It is one of several laws passed by Congress to protect federally chartered financial institutions from “unequal and unfriendly competition” caused by state tax laws favoring state-chartered institutions.2 On its face, however, Massachusetts’ tax scheme is not unfriendly or discriminatory. It applies a single neutral standard to state and federal institutions alike. The amount of the deduction depends on varying regulatory practices, but a tax is not invalid because it recognizes that state and federal regulations may differ. There is no reason to believe that § 5 (h) was intended to force state and federal regulation into the same mold.3 2 Mercantile Bank v. New York, 121 U. S. 138, 155. See 12 U. S. C. §548 (1976 ed.) (national banks); 12 U. S. C. §627 (1976 ed.) (corporations federally authorized to engage in foreign banking). 3 Indeed, the federal statute protects federal associations from being forced into the state regulatory mold. The deposits aspect of the tax was invalidated partly because its apparently neutral provisions were FIRST FEDERAL S. & L. v. MASSACHUSETTS TAX COMM’N 259 255 Opinion of the Court Notwithstanding its neutral language, the federal associations argue that the tax is discriminatory in fact. They have not, however, established that it is unfairly burdensome in “practical operation.” Michigan Nat. Bank v. Michigan, 365 U. S. 467, 476. The record does not indicate that federal associations have suffered a significant handicap in competing with state institutions, or that any other federal policies have been thwarted.4 The lower reserve requirement, by making more funds available for dividends, may well give the associations a competitive advantage, despite the tax. Certainly the associations’ rush to amend their charters in 1970 lends support to that conclusion. Any suggestion of discriminatory purpose calculated to impose state regulatory requirements on federal associations. The statute permitted an institution to take a deduction for loans secured by out-of-state real estate but only if the property was within 50 miles of the institution’s home office. Mass. Gen. Laws Ann., ch. 63, § 11 (West Supp. 1977). This limit reflected state restrictions on making out-of-state loans more than 50 miles from the home office. United States v. State Tax Common, 481 F. 2d 963, 968-969, n. 6 (CAI 1973). But federal associations are empowered by federal law to make such loans up to 100 miles from home. 12 U. S. C. § 1464 (c) (1976 ed.). By treating the state and federal institutions as though they were subject to the same regulatory limits, the statute exacted a higher tax from federal associations and tended at the same time to force federal associations to follow state rather than federal regulations. It is difficult to conceive of a non discriminatory reason for the 50-mile limit on deductions. For these reasons, the Court of Appeals for the First Circuit held the tax discriminatory under §5 (h). 481 F. 2d, at 970. 4 Cf. n. 3, supra. The sparse evidence introduced on this point by the associations is ambiguous at best. For example, in three of the seven years from 1968 to 1975, federal associations put a larger proportion of their assets into required reserves than did state savings banks, which are the dominant state mutual institutions. From 1970 through 1973, federal associations made smaller contributions to surplus than state savings banks, but in these years the federal associations may have been simply consuming reserves built up under the stringent requirements of their pre-1970 charters. 260 OCTOBER TERM, 1977 Opinion of the Court 437U.S. is foreclosed by the fact that the tax was enacted when federal reserve requirements were as high as state requirements. II Massachusetts does not impose its tax on credit unions. Arguing that credit unions in Massachusetts are “similar” to federal savings and loan associations, the associations claim entitlement to the credit unions’ exemption. There are indeed similarities between these two kinds of financial institutions. For example, both are characterized by mutual ownership and control; 12 CFR § 544.1 (1977); Mass. Gen. Laws Ann., ch. 171, §§ 10, 13, and 24 (West 1971 and Supp. 1977); and both are empowered to make loans secured by real estate. 12 U. S. C. § 1464 (c) (1976 ed.); Mass. Gen. Laws Ann., ch. 171, § 24 (West Supp. 1977). But the institutions are far from identical. Congress has long treated federally chartered credit unions differently from federally chartered savings and loan associations, giving the credit unions, but not the savings and loan associations, an exemption from state taxes. See 12 U. S. C. § 1768 (1976 ed.). In establishing insurance programs to protect members’ deposits, Congress distinguished state and federal credit unions from state and federal savings and loan associations. See 12 U. S. C. §§ 1726 (a) and 1781 (a) (1976 ed.). Moreover, courts in other jurisdictions have generally rejected the claim that credit unions are “similar” under § 5 (h) to federal savings and loan associations.5 The distinctions found in those jurisdictions have validity in Massachusetts as well. By law, Massachusetts credit unions must give preference to small personal loans, Mass. Gen. Laws 5 See Manchester Federal Savings & Loan Assn. n. State Tax Comm’n, 105 N. H. 17, 191 A.. 2d 529 (1963); First Federal, Savings & Loan Assn. v. Connelly, 142 Conn. 483, 115 A. 2d 455 (1955), appeal dismissed, 350 U. S. 927; State v. Minnesota Federal Savings & Loan Assn., 218 Minn. 229, 15 N. W. 2d 568 (1944). FIRST FEDERAL S. & L. v. MASSACHUSETTS TAX COMM’N 261 255 Opinion of the Court Ann., ch. 171, §24 (West Supp. 1977), while the primary lending role of federal savings and loan associations is “to provide for the financing of homes.” 12 U. S. C. § 1464 (a) (1976 ed.). Massachusetts credit unions may lend only to members, Mass. Gen. Laws Ann., ch. 171, § 24 (West Supp. 1977), while federal associations are not so limited. And, despite individual exceptions, there are major differences between the actual lending practices of state credit unions as a class and federal associations as a class.6 Of greater importance than these differences, however, is the fact that Massachusetts credit unions are not the federal associations’ closest state-chartered competitors. Massachusetts savings banks and cooperative banks have much more in common with federal associations than do state credit unions; their business is unquestionably similar to that of the federal associations.7 These institutions are an important segment of Massachusetts’ financial community.8 Any favoritism shown 6 As the Supreme Judicial Court noted: “In 1972, . . . credit unions placed 30.1% of their total investments (in dollars) in real estate mortgages. Federal savings and loan associations had 87.7% of their total investments (in dollars) in real estate mortgages. . . . Federal savings and loan associations had almost 98% of their total loans in real estate mortgages .... Credit unions, on the other hand, had only about 42% of their total loans in real estate mortgages.” 372 Mass. 478, 493-494, 363 N. E. 2d 474, 484 (1977). 7 See, e. g., Commissioner of Corporations & Taxation v. Flaherty, 306 Mass. 461, 28 N. E. 2d 433 (1940); Springfield Institution for Savings v. Worcester Federal Savings & Loan Assn., 329 Mass. 184, 107 N. E. 2d 315 (1952). Massachusetts cooperative banks had more than 97% of their total loans in real estate mortgages in 1972, while state savings banks had 95% of their loans in real estate mortgages. Federal associations had almost 98% of their loans in real estate mortgages. Cooperative banks had 80.4% of their total dollar investments in real estate mortgages, and savings banks had 65.3% in such mortgages. The figure for federal associations was 87.7%. See 372 Mass., at 493, 363 N. E. 2d, at 484. 8 Their assets greatly exceed those of state credit unions. State savings banks had assets of almost $18.5 billion in 1973; cooperative banks had almost $3 billion in assets; federal associations had almost $2.5 billion; and 262 OCTOBER TERM, 1977 Opinion of the Court 437U.S. to Massachusetts credit unions falls as harshly on them as on the federal associations. Nonetheless, the Massachusetts Legislature has concluded that credit unions are not similar to state cooperative and savings banks or to state and federal savings and loan associations. When Congress required that federal savings and loan associations be placed in the same classification as “similar” state institutions, it certainly did not assume that every local and mutual or cooperative thrift and home-financing institution is similar to a federal association. See 12 U. S. C. § 1464 (h) (1964 ed.). It recognized that States might classify their own institutions in various ways. Massachusetts has excluded credit unions from a large classification, that includes the institutions most closely resembling federal savings and loan associations. The composition of the class in which Massachusetts has placed the federal associations satisfies the federal statute’s central purpose of protecting federal associations from discriminatory treatment. We conclude that Massachusetts has not imposed a greater tax on the federal associations than that imposed on other “similar” institutions.9 credit unions had over $1 billion. App. 131-132; Annual Report of the Commissioner of Banks, Commonwealth of Massachusetts, Division of Banks and Loan Agencies, Sec. B (Credit Unions), iv (1973). 9 Only two of the associations’ remaining attacks on the statute deserve mention. They claim that Massachusetts’ tax is not one of the enumerated taxes approved by § 5 (h), which allows a nondiscriminatory “tax on [federal] associations or their franchise, capital, reserves, surplus, loans, or income.” 12 U. S. C. § 1464 (h) (1976 ed.). Whether or not this tax may be characterized as a “franchise” or an “income” tax, it is certainly a tax “on” federal associations and therefore within the ambit of § 5 (h). The federal associations also argue that the state statute violates the Commerce Clause by creating a risk of multiple taxation. They claim that some neighboring State may at some time in the future attempt to tax the income from loans secured by property in that State. This argument is wholly speculative and unsupported by evidence in the record. FIRST FEDERAL S. & L. v. MASSACHUSETTS TAX COMM’N 263 255 Opinion of Blackmun, J. Accordingly, the judgment of the Supreme Judicial Court is affirmed. So ordered. Mr. Justice Blackmun, concurring in part and dissenting in part. Section 5 (h) of the Home Owners’ Loan Act of 1933, as amended, 76 Stat. 984, 12 U. S. C. § 1464 (h) (1976 ed.), reads: “No State, county, municipal, or local taxing authority shall impose any tax on such associations or their franchise, capital, reserves, surplus, loans, or income greater than that imposed by such authority on other similar local mutual or cooperative thrift and home financing institutions.” The Court, in speaking of this statute, has said : “This provision unequivocally bars discriminatory state taxation of the Federal Savings and Loan Associations.” Laurens Federal Savings & Loan Assn. v. South Carolina Tax Comm’n, 365 U. S. 517, 523 (1961). I agree with the Court’s ruling today on the first issue, namely, that the lesser reserve deduction available for federal savings and loan associations of itself does not demonstrate that the- associations pay a greater tax than similar Massachusetts savings banks. On the second issue, however, I am in disagreement with the Court and, to that extent, dissent from its opinion. For this issue, the important focus of the statute is on the word “similar,” and the measure of the Commonwealth’s allowable tax is only that imposed “on other similar local mutual or cooperative thrift and home financing institutions.” There is no argument here that Massachusetts credit unions are not “local mutual or cooperative thrift and home financing institutions,” within the meaning of § 5 (h). See Mass. Gen. Laws Ann., ch. 171, § 2 (West 1971). The Supreme Judicial 264 OCTOBER TERM, 1977 Opinion of Blackmun, J. 437U.S. Court so found, 372 Mass. 478, 492, 363 N. E. 2d 474, 483 (1977), and no challenge to that finding is made here. The question, then, is whether Massachusetts credit unions are “similar” to federal savings and loan associations. If they are similar, the tax Massachusetts would impose on the federal entities, see Mass. Gen. Laws Ann., ch. 63, § 11 (West Supp. 1977), violates the statute, for the Commonwealth’s excise does not apply at all to Massachusetts credit unions. The Court, in construing a similar federal statute, Rev. Stat. § 5219, as amended, 12 U. S. C. § 548 (1) (b), which had barred state taxation of the shares of national banks “at a greater rate than is assessed upon other moneyed capital . . . coming into competition with the business of national banks,” and at a rate higher than the highest rates assessed upon business corporations, observed that Congress intended “to prohibit only those systems of state taxation which discriminate in practical operation against national banking associations or their shareholders as a class.” Tradesmens Nat. Bank v. Oklahoma Tax Comm’n, 309 U. S. 560, 567 (1940); Michigan Nat. Bank v. Michigan, 365 U. S. 467, 473 (1961). The policy of § 5 (h) obviously is to assure that the States do not put federal associations to any competitive disadvantage with respect to local savings institutions. The statutory term “similar” usually, and certainly here, does not mean “identical.” 1 The Massachusetts credit union and the federal savings and loan association are “similar” with respect to their fundamental elements. Each has mutuality of ownership and control. Each has the pronounced ability to attract savings. Each is empowered to make first mortgage residential real estate loans on substantially the same terms 1 See Commonwealth v. Fontain, 127 Mass. 452, 454 (1879); Chicago v. Vaccarro, 408 Ill. 587, 601, 97 N. E. 2d 766, 773 (1951); Thomas v. Consumers Power Co., 58 Mich. App. 486, 493-494, 228 N. W. 2d 786, 790 (1975); Miller n. Allstate Ins. Co., 66 Wash. 2d 871, 875, 405 P. 2d 712, 714 (1965). FIRST FEDERAL S. & L. v. MASSACHUSETTS TAX COMM’N 265 255 Opinion of Blackmun, J. and to approximately the same extent. The Massachusetts credit union has the statutory authority to make loans secured by first mortgages on real estate for terms up to 30 years, for 90% of the value of the property, and to a maximum amount of $40,000. See Mass. Gen. Laws Ann., ch. 171, §§ 24 (B)(a) (4) and (b)(8) (West Supp. 1977), and 1977 Mass. Acts, ch. 20. A federal association may make real estate loans for terms up to 30 years, for 80% of the value of the property, and to a maximum amount of $55,000. See 12 U. S. C. § 1464 (c) (1976 ed.); 12 CFR §545.0-1 (a)(l)(i) (1977). Although the Massachusetts credit union, to be sure, may make loans only to members and is required to give “preference” to “personal loans,” see Mass. Gen. Laws Ann., ch. 171, § 24 (West Supp. 1977), this distinction is minor and does not demonstrate that the credit union is not “similar” to the federal association, within the meaning of § 5 (h). There is no statutory limitation on the membership of the Massachusetts credit union, other than self-imposed conditions of residence, occupation, or association, see Mass. Gen. Laws Ann., ch. 171, § 7 (c) (West 1971), and a small deposit will qualify a prospective borrower as a member. In addition, there is no statutory enforcement of the “preference” in favor of personal loans. The Supreme Judicial Court observed, 372 Mass., at 493-494, 363 N. E. 2d, at 484, that in 1972 Massachusetts credit unions placed 30.1% of their total dollar investments in real estate mortgages, and 42% of their total loans in real estate mortgages.2 As of the end of 1973, they had $329 million as outstanding mortgage loans. Large Massachusetts credit unions may invest up to 80% of their assets in real estate loans, see Mass. Gen. Laws Ann., ch. 171, § 24 (B) (6) (7) (West Supp. 1977). All this leads me to conclude that the Massachusetts credit union in all pertinent respects is “similar,” and not dissimilar, 2 Federal associations had 87.7% of their total dollar investments in real estate mortgages and almost 98% of their total loans in such mortgages. 266 OCTOBER TERM, 1977 Opinion of Blackmun, J. 437 U. S. to the federal savings and loan association.3 Both perform the same functions in that they attract savings upon which they pay interest, and they make loans, substantial amounts of which are first mortgage residential loans. It follows, in my view, that, because of these similarities, the exemption of Massachusetts credit unions from the Massachusetts excise tax to which federal savings and loan associations are subject renders the tax invalid, under § 5 (h), as applied to the federal institutions. I therefore would reverse the judgment of the Supreme Judicial Court of Massachusetts. 3 See Message of the President to the Congress on Tax Reduction and Reform, Jan. 20, 1978, 14 Weekly Comp, of Pres. Docs. 158, 172. MOORMAN MFG. CO. v. BAIR 267 Syllabus MOORMAN MANUFACTURING CO. v. BAIR, DIRECTOR OF REVENUE OF IOWA APPEAL FROM THE SUPREME COURT OF IOWA No. 77-454. Argued March 21, 1978—Decided June 15, 1978 An Iowa statute prescribes a so-called single-factor sales formula for apportioning an interstate corporation’s income for state income tax purposes. Under this formula, the part of income from such a corporation’s sale of tangible personal property attributable to business within the State and hence subject to the state income tax is deemed to be in that proportion which the corporation’s gross sales made within the State bear to its total gross sales. Appellant, an Illinois corporation that sells animal feed it manufactures in Illinois to Iowa customers through Iowa salesmen and warehouses, brought an action in an Iowa court challenging the constitutionality of the single-factor formula. The trial court held the formula invalid under the Due Process Clause of the Fourteenth Amendment and the Commerce Clause, but the Iowa Supreme Court reversed. Held: 1. Iowa’s single-factor formula is not invalid under the Due Process Clause. Pp. 271-275. (a) Any assumption that at least some portion of appellant’s income from Iowa sales was generated by Illinois activities is too speculative to support a claim that Iowa in fact taxed profits not attributable to activities within the State. P. 272. (b) An apportionment formula, such as the single-factor formula, that is necessarily employed as a rough approximation of a corporation’s income reasonably related to the activities conducted within the taxing State will only be disturbed when the taxpayer has proved by “clear and cogent evidence” that the income attributed to the State is in fact “out of all reasonable proportion to the business transacted ... in that State,” Hans Rees’ Sons v. North Carolina ex rel. Maxwell, 283 U. S. 123, 135, or has “led to a grossly distorted result,” Norfolk & Western R. Co. v. State Tax Comm’n, 390 U. S. 317, 326. Here, the Iowa statute afforded appellant an opportunity to demonstrate that the single-factor formula produced an arbitrary result in its case, but the record contains no such showing. Pp. 272-275. 2. Nor is Iowa’s single-factor formula invalid under the Commerce Clause. Pp. 276-281. 268 OCTOBER TERM, 1977 Syllabus 437 U. S. (a) On this record, the existence of duplicative taxation as between Iowa and Illinois (which uses the so-called three-factor—property, payroll, and sales—formula) is speculative, but even assuming some overlap, appellant’s argument that Iowa, rather than Illinois, was necessarily at fault in a constitutional sense cannot be accepted. Where the record does not reveal the sources of appellant’s profits, its Commerce Clause claim cannot rest on the premise that profits earned in Illinois were included in its Iowa taxable income and therefore the Iowa formula was at fault for whatever overlap may have existed. Pp. 276-277. (b) The Commerce Clause itself, without implementing legislation by Congress, does not require, as appellant urges, that Iowa compute ■corporate net income under the Illinois three-factor formula. If the Constitution were read to mandate a prohibition against any overlap in the computation of taxable income by the States, the consequences would extend far beyond this particular case and would require extensive judicial lawmaking. Pp. 277-281. 254 N. W. 2d 737, affirmed. Stevens, J., delivered the opinion of the Court, in which Burger, C. J., and Stewart, White, Marshall, and Rehnquist, JJ., joined. Brennan, J., post, p. 281, and Blackmun, J., post, p. 282, filed dissenting opinions. Powell, J., filed a dissenting opinion, in which Blackmun, J., joined, post, p. 283. Donald K. Barnes argued the cause for appellant. With him on the briefs were Walter R. Brown, John V. Donnelly, Carl G. Schmiedeskamp, and Robert W. Cook. Harry M. Griger, Assistant Attorney General of Iowa, argued the cause for appellee. With him on the brief was Richard C. Turner, Attorney General.* ^Ernest S. Christian, Jr., and Allan Abbot Tuttle filed a brief for the Committee on State Taxation of the Council of State Chambers of Commerce as amicus curiae urging reversal. James L. Rogers, John R. Phillips, and Philip B. Kurland filed a brief for the Iowa Manufacturers Assn, et al. as amici curiae urging affirmance. Wiliam D. Dexter, James A. Redden, Attorney General of Oregon, and Theodore W. deLooze, Assistant Attorney General, filed a brief for the Multistate Tax Comm’n et al. as amici curiae. MOORMAN MFG. CO. v. BAIR 269 267 Opinion of the Court Mr. Justice Stevens delivered the opinion of the Court. The question in this case is whether the single-factor sales formula employed by Iowa to apportion the income of an interstate business for income tax purposes is prohibited by the Federal Constitution. I Appellant, Moorman Manufacturing Co., is an Illinois corporation engaged in the manufacture and sale of animal feeds. Although the products it sells to Iowa customers are manufactured in Illinois, appellant has over 500 salesmen in Iowa and it owns six warehouses in the State from which deliveries are made to Iowa customers. Iowa sales account for about 20% of appellant’s total sales. Corporations, both foreign and domestic, doing business in Iowa are subject to the State’s income tax. The taxable income for federal income tax purposes, with certain adjustments, is treated as the corporation’s “net income” under the Iowa statute. If a corporation’s business is not conducted entirely within Iowa, the statute imposes a tax only on the portion of its income “reasonably attributable” to the business within the State. There are essentially two steps in computing the share of a corporation’s income “reasonably attributable” to Iowa. First, certain income, “the geographical source of which is easily identifiable,” is attributed entirely to a particular State.1 xThe statute provides: “Interest, dividends, rents, and royalties (less related expenses) received in connection with business in the state, shall be allocated to the state, and where received in connection with business outside the state, shall be allocated outside of the state.” Iowa Code §422.33 (l)(a) (1977). In describing this section, the Iowa Supreme Court stated that “certain income, the geographical source of which is easily identifiable, is allocated to the appropriate state.” 254 N. W. 2d 737, 739. Thus, for example, rental income would be attributed to the State where the property was located. And in appellant’s case, this section operated to exclude its investment income from the tax base. 270 OCTOBER TERM, 1977 Opinion of the Court 437U.S. Second, if the remaining income is derived from the manufacture or sale of tangible personal property, “the part thereof attributable to business within the state shall be in that proportion which the gross sales made within the state bear to the total gross sales.” 2 This is the single-factor formula that appellant challenges in this case. If the taxpayer believes that application of this formula subjects it to taxation on a greater portion of its net income than is “reasonably attributable” to business within the State, it may file a statement of objections and submit an alternative method of apportionment. If the evidence submitted by the taxpayer persuades the Director of Revenue that the statute is “inapplicable and inequitable” as applied to it, he may recalculate thei corporation’s taxable income. During the fiscal years 1949 through 1960, the State Tax Commission allowed appellant to compute its Iowa income on the basis of a formula consisting of three, equally weighted factors—property, payroll, and sales—rather than the formula prescribed by statute.3 For the fiscal years 1961 through 1964, appellant complied with a directive of the State Tax Commission to compute its income in accordance with the statutory formula. Since 1965, however, appellant has resorted to the three-factor formula without the consent of the commission. In 1974, the Iowa Director of Revenue revised appellant’s tax assessment for the fiscal years 1968 through 1972. This assessment was based on the statutory formula, which pro- 2 Iowa Code §422.33 (1) (6) (1977). 3 The operation of the two formulas may be briefly described. The single-factor sales formula yields a percentage representing a ratio of gross sales in Iowa to total gross sales. The three-factor formula yields a percentage representing an average of three ratios: property within the State to total property, payroll within the State to total payroll, and sales within the State to total sales. These percentages are multiplied by the adjusted total net income to arrive at Iowa taxable net income. This net income figure is then multiplied by the tax rate to compute the actual tax obligation of the taxpayer. MOORMAN MFG. CO. v. BAIR 271 267 Opinion of the Court duced a higher percentage of taxable income than appellant, using the three-factor formula, had reported on its return in each of the disputed years.4 The higher percentages, of course, produced a correspondingly greater tax obligation for those years.6 After the Tax Commission had rejected Moorman’s appeal from the revised assessment, appellant challenged the constitutionality of the single-factor formula in the Iowa District Court for Polk County. That court held the formula invalid under the Due Process Clause of the Fourteenth Amendment and the Commerce Clause. The Supreme Court of Iowa reversed, holding that an apportionment formula that is necessarily only a rough approximation of the income properly attributable to the taxing State is not subject to constitutional attack unless the taxpayer proves that the formula has produced an income attribution “out of all proportion to the business transacted” within the State. The court concluded that appellant had not made such a showing. We noted probable jurisdiction of Moorman’s appeal, 434 U. S. 953, and now affirm. II Appellant contends that Iowa’s single-factor formula results in extraterritorial taxation in violation of the Due Process 4 For those years the two formulas resulted in the following percentages : Fiscal Year Sales Factor Three-Factor Ended Percentage Percentage 3/31/68 21.8792% 14.1088% 3/31/69 21.2134% 14.3856% 3/31/70 19.9492% 14.0200% 3/31/71 18.9544% 13.2186% 3/31/72 18.6713% 12.2343% For a description of how these percentages are computed, see n. 3, supra. 5 Thus, in 1968, for example, Moorman’s three-factor computation resulted in a tax of $81,466, whereas the Director’s single-factor computation resulted in a tax of $121,363. 272 OCTOBER TERM, 1977 Opinion of the Court 437 U. S. Clause. This argument rests on two premises: first, that appellant’s Illinois operations were responsible for some of the profits generated by sales in Iowa; and, second, that a formula that reaches any income not in fact earned within the borders of the taxing State violates due process. The first premise is speculative and the second is foreclosed by prior decisions of this Court. Appellant does not suggest that it has shown that a significant portion of the income attributed to Iowa in fact was generated by its Illinois operations; the record does not contain any separate accounting analysis showing what portion of appellant’s profits was attributable to sales, to manufacturing, or to any other phase of the company’s operations. But appellant contends that we should proceed on the assumption that at least some portion of the income from Iowa sales was generated by Illinois activities. Whatever merit such an assumption might have from the standpoint of economic theory or legislative policy, it cannot support a claim in this litigation that Iowa in fact taxed profits not attributable to activities within the State during the years 1968 through 1972. For all this record reveals, appellant’s manufacturing operations in Illinois were only marginally profitable during those years and the high-volume sales to Iowa customers from Iowa warehouses were responsible for the lion’s share of the income generated by those sales. Indeed, a separate accounting analysis might have revealed that losses in Illinois operations prevented appellant from earning more income from exploitation of a highly favorable Iowa market. Yet even were we to assume that the Illinois activities made some contribution to the profitability of the Iowa sales, appellant’s claim that the Constitution invalidates an apportionment formula whenever it may result in taxation of some income that did not have its source in the taxing State is incorrect. The Due Process Clause places two restrictions on a State’s power to tax income generated by the activities of an interstate MOORMAN MFG. CO. v. BAIR 273 267 Opinion of the Court business. First, no tax may be imposed unless there is some minimal connection between those activities and the taxing State. National Bellas Hess, Inc. v. Department of Revenue, 386 U. S. 753, 756. This requirement was plainly satisfied here. Second, the income attributed to the State for tax purposes must be rationally related to “values connected with the taxing State.” Norfolk & Western R. Co. v. State Tax Comm’n, 390 U. S. 317, 325. Since 1934 Iowa has used the formula method of computing taxable income. This method, unlike separate accounting, does not purport to identify the precise geographical source of a corporation’s profits; rather, it is employed as a rough approximation of a corporation’s income that is reasonably related to the activities conducted within the taxing State. The single-factor formula used by Iowa, therefore, generally will not produce a figure that represents the actual profits earned within the State. But the same is true of the Illinois three-factor formula. Both will occasionally over-reflect or under-reflect income attributable to the taxing State. Yet despite this imprecision, the Court has refused to impose strict constitutional restraints on a State’s selection of a particular formula.0 Thus, we have repeatedly held that a single-factor formula is presumptively valid. In Underwood Typewriter Co. v. Chamberlain, 254 U. S. 113, for example, the taxpayer challenged Connecticut’s use of such a formula to apportion its net income. Underwood’s manufacturing operations were conducted entirely within Connecticut. Its main office, however, was in New York City and it had branch offices in many States where its typewriters were sold and repaired. Applying a single-factor property formula, Connecticut taxed 47% of the company’s net income. Claiming that 97% of its profits were 6 See, e. g., Underwood Typewriter Co. v. Chamberlain, 254 U. S. 113; Bass, Ratcliff & Gretton, Ltd. n. State Tax Comm’n, 266 U. S. 271; Ford Motor Co. v. Beauchamp, 308 U. S. 331. 274 OCTOBER TERM, 1977 Opinion of the Court 437U.S. generated by transactions in tangible personal property outside Connecticut, Underwood contended that the formula taxed “income arising from business conducted beyond the boundaries of the State” in violation of the Due Process Clause. Id., at 120. Rejecting this claim, the Court noted that Connecticut “adopted a method of apportionment which, for all that appears in this record, reached, and was meant to reach, only the profits earned within the State,” id., at 121, and held that the taxpayer had failed to carry its burden of proving that “the method of apportionment adopted by the State was inherently arbitrary, or that its application to this corporation produced an unreasonable result.” Ibid, (footnote omitted).7 In individual cases, it is true, the Court has found that the application of a single-factor formula to a particular taxpayer violated due process. See Hans Rees’ Sons, Inc. v. North Carolina ex rel. Maxwell, 283 U. S. 123; Norfolk de Western R. Co. v. State Tax Comm’n, supra. In Hans Rees’, for example, the Court concluded that proof that the formula produced a tax on 83% of the taxpayer’s income when only 17% of that income actually had its source in the State would suffice to invalidate the assessement under the Due Process Clause. But in neither Hans Rees’ nor Norfolk do Western did the Court depart from the basic principles that the States have wide latitude in the selection of apportionment formulas and that a formula-produced assessment will only be disturbed when the taxpayer has proved by “clear and cogent evidence” that the income attributed to the State is in fact “out of all appropriate proportions to the business transacted ... in that State,” 283 U. S., at 135, or has “led to a grossly distorted result,” 390 U. S., at 326. General Motors Corp. v. District of Columbia, 380 U. S. 553, 7 See also Bass, Ratcliff & Gretton, Ltd. n. State Tax Comm’n, supra; Norfolk & Western R. Co. v. North Carolina ex rel. Maxwell, 297 U. S. 682. MOORMAN MFG. CO. v. BAIR 275 267 Opinion of the Court on which appellant relies, does not suggest a contrary result. In that case the Court held that a regulation prescribing a single-factor sales formula was not authorized by the District of Columbia Code. It concluded that the formula violated the statutory requirement that the net income of a corporation doing business both inside and outside the District must be deemed to arise from “sources” both inside and outside the District. But that statutory requirement has no counterpart in the Constitution, and the Court in General Motors made clear that it did “not mean to take any position on the constitutionality of a state income tax based on the sales factor alone.” Id., at 561.8 The Iowa statute afforded appellant an opportunity to demonstrate that the single-factor formula produced an arbitrary result in its case. But this record contains no such showing and therefore the Director’s assessment is not subject to challenge under the Due Process Clause.9 8 The Court, it is true, expressed doubts about the wisdom of the economic assumptions underlying the challenged formula and noted that its use in the context of the more prevalent three-factor formula would not advance the policies underlying the Commerce Clause. But these considerations were deemed relevant to the question of legislative intent, not constitutional interpretation. 9 In his concurring opinion, Justice McCormick of the Iowa Supreme Court made this point: “In the present case, Moorman did not attempt to prove the amount of its actual net income from Iowa activities in the years involved. Therefore no basis was presented for comparison of the corporation’s Iowa income and the income apportioned to Iowa under the formula. In this era of sophisticated accounting techniques, it should not be impossible for a unitary corporation to prove its actual income from activities in a particular state. However, Moorman showed only that its tax liability would be substantially less if Iowa employed a three-factor apportionment formula. We have no basis to assume that the three-factor formula produced a result equivalent to the corporation’s actual income from Iowa activities. Having failed to establish a basis for comparison of its actual income in Iowa with the income apportioned to Iowa under the single-factor formula, Moorman did not demonstrate that the single-factor formula 276 OCTOBER TERM, 1977 Opinion of the Court 437 IT. S. Ill Appellant also contends that during the relevant years Iowa and Illinois imposed a tax on a portion of the income derived from the Iowa sales that was also taxed by the other State in violation of the Commerce Clause.10 Since most States use the three-factor formula that Illinois adopted in 1970, appellant argues that Iowa’s longstanding single-factor formula must be held responsible for the alleged duplication and declared unconstitutional. We cannot agree. In the first place, this record does not establish the essential factual predicate for a claim of duplicative taxation. Appellant’s net income during the years in question was approximately $9 million. Since appellant did not prove the portion derived from sales to Iowa customers, rather than sales to customers in other States, we do not know whether Illinois and Iowa together imposed a tax on more than 100% of the relevant net income. The income figure that appellant contends was subject to duplicative taxation was computed by comparing gross sales in Iowa to total gross sales. As already noted, however, this figure does not represent actual profits earned from Iowa sales. Obviously, all sales are not equally profitable. Sales in Iowa, although only 20% of gross sales, may have yielded a much higher percentage of appellant’s profits. Thus, profits from Iowa sales may well have exceeded the $2.5 million figure that appellant contends was taxed by the two States. If so, there was no duplicative taxation of the net income generated by Iowa sales. In any event, on this record its existence is speculative.11 produced a grossly unfair result. Thus it did not prove unconstitutionality of the formula as applied.” 254 N. W. 2d, at 757. 10 Since Illinois did not adopt its income tax until 1970, there was no possibility of any overlap until that year. The alleged overlap in the three years following Illinois’ enactment of an income tax was 34.38% in 1970, 34.51% in 1971, and 37.01% in 1972. 11 Since there is no evidence in the record regarding the percentages of its total net income taxed in the other States in which it did business during MOORMAN MFG. CO. v. BAIR 277 267 Opinion of the Court Even assuming some overlap, we could not accept appellant’s argument that Iowa, rather than Illinois, was necessarily at fault in a constitutional sense. It is, of course, true that if Iowa had used Illinois’ three-factor formula, a risk of duplication in the figures computed by the two States might have been avoided. But the same would be true had Illinois used the Iowa formula. Since the record does not reveal the sources of appellant’s profits, its Commerce Clause claim cannot rest on the premise that profits earned in Illinois were included in its Iowa taxable income and therefore the Iowa formula was at fault for whatever overlap may have existed. Rather, the claim must be that even if the presumptively valid Iowa formula yielded no profits other than those properly attributable to appellant’s activities within Iowa, the importance of avoiding any risk of duplication in the taxable income of an interstate concern justifies invalidation of the Iowa statute. Appellant contends that, to the extent this overlap is permitted, the corporation that does business in more than one State shoulders a tax burden not shared by those operating entirely within a State.12 To alleviate the burden, appellant those years, any claim that appellant was taxed on more than 100% of its total net income would also be speculative. 12 Appellant also contends that the Iowa formula discriminates against interstate commerce in violation of the Commerce Clause and the Equal Protection Clause, because an Illinois corporation doing business in Iowa must pay tax on a greater portion of its income than a local Iowa company, and an Iowa company doing business in Illinois will pay tax on less of its income than an Illinois corporation doing business in Iowa. The simple answer, however, is that whatever disparity may have existed is not attributable to the Iowa statute. It treats both local and foreign concerns with an even hand; the alleged disparity can only be the consequence of the combined effect of the Iowa and Illinois statutes, and Iowa is not responsible for the latter. Thus, appellant’s “discrimination” claim is simply a way of describing the potential consequences of the use of different formulas by the two States. These consequences, however, could be avoided by the adoption of any uniform rule; the “discrimination” does not inhere in either State’s formula. 278 OCTOBER TERM, 1977 Opinion of the Court 437U.S. invites us to hold that the Commerce Clause itself, without implementing legislation by Congress, requires Iowa to compute corporate net income under the Illinois equally weighted, three-factor formula. For the reasons that follow, we hold that the Constitution does not require such a result. The only conceivable constitutional basis for invalidating the Iowa statute would be that the Commerce Clause prohibits any overlap in the computation of taxable income by the States. If the Constitution were read to mandate such precision in interstate taxation, the consequences would extend far beyond this particular case. For some risk of duplicative taxation exists whenever the States in which a corporation does business do not follow identical rules for the division of income. Accepting appellant’s view of the Constitution, therefore, would require extensive judicial lawmaking. Its logic is not limited to a prohibition on use of a single-factor apportionment formula. The asserted constitutional flaw in that formula is that it is different from that presently employed by a majority of States and that difference creates a risk of duplicative taxation. But a host of other division-of-income problems create precisely the same risk and would similarly rise to constitutional proportions. Thus, it would be necessary for this Court to prescribe a uniform definition of each category in the three-factor formula. For if the States in which a corporation does business have different rules regarding where a “sale” takes place, and each includes the same sale in its three-factor computation of the corporation’s income, there will be duplicative taxation despite the apparent identity of the formulas employed.13 A similar 13 Thus, while some States such as Iowa assign sales by destination, “sales can be assigned to the state ... of origin, the state in which the sales office is located, the state where an employee of the business making the sale carries on his activities or where the order is first accepted, or the state in which an interstate shipment is made.” Note, State Taxation of Interstate Businesses and the Multistate Tax Compact: The Search for a Delicate MOORMAN MFG. CO. v. BAIR 279 267 Opinion of the Court risk of multiple taxation is created by the diversity among the States in the attribution of “nonbusiness” income, generally defined as that portion of a taxpayer’s income that does not arise from activities in the regular course of its business.14 Some States do not distinguish between business and nonbusiness income for apportionment purposes. Other States, however, have adopted special rules that attribute nonbusiness income to specific locations. Moreover, even among the latter, there is diversity in the definition of nonbusiness income and in the designation of the locations to which it is deemed attributable. The potential for attribution of the same income to more than one State is plain.15 The prevention of duplicative taxation, therefore, would require national uniform rules for the division of income. Although the adoption of a uniform code would undeniably advance the policies that underlie the Commerce Clause, it would require a policy decision based on political and economic considerations that vary from State to State. The Constitution, however, is neutral with respect to the content of any uniform rule. If division-of-income problems were to be constitutionalized, therefore, they would have to be resolved in the manner suggested by appellant for resolution of formula diversity—the prevalent practice would be endorsed as the constitutional rule. This rule would at best be an amalgam of independent state decisions, based on considerations unique to each State. Of most importance, it could not reflect the Uniformity, 11 Colum. J. Law & Soc. Prob. 231, 237 n. 20 (1975) (citation omitted). 14See, e. g., Uniform Division of Income for Tax Purposes Act § 1 (a). 15 Thus, one State in which a corporation does business may consider a particular type of income business income and simply include it in its apportionment formula; a second State may deem that same income nonbusiness income and attribute it to itself as the “commercial domicile” of the company; and a third State, though also considering it nonbusiness income, may attribute it to itself as the “legal domicile” of the company. See Note, supra n. 13, at 239. 280 OCTOBER TERM, 1977 Opinion of the Court 437U.S. national interest, because the interests of those States whose policies are subordinated in the quest for uniformity would be excluded from the calculation.16 While the freedom of the States to formulate independent policy in this area may have to yield to an overriding national interest in uniformity, the content of any uniform rules to which they must subscribe should be determined only after due consideration is given to the interests of all affected States. It is clear that the legislative power granted to Congress by the Commerce Clause of the Constitution would amply justify the enactment of legislation requiring all States to adhere to uniform rules for the division of income. It is to that body, and not this Court, that the Constitution has committed such policy decisions. Finally, it would be an exercise in formalism to declare appellant’s income tax assessment unconstitutional based on speculative concerns with multiple taxation. For it is evident that appellant would have had no basis for complaint if, instead of an income tax, Iowa had imposed a more burdensome gross-receipts tax on the gross receipts from sales to Iowa customers. In Standard Pressed Steel Co. v. Washington Revenue Dept., 419 U. S. 560, the Court sustained a tax on the entire gross receipts from sales made by the taxpayer into Washington State. Because receipts from sales made to States other than Washington were not included in Standard Pressed Steel’s taxable gross receipts, the Court concluded that the tax was 11 ‘apportioned exactly to the activities taxed.’ ” Id., at 564. In this case appellant’s actual income tax obligation was the rough equivalent of a 1 % tax on the entire gross receipts from its Iowa sales. Thus, the actual burden on interstate commerce would have been the same had Iowa imposed a plainly 16 This process is especially unsettling if a longstanding tax policy of one State, such as Iowa’s, becomes the object of constitutional attack simply because it is different from the recently adopted practice of its neighbor. MOORMAN MFG. CO. v. BAIR 281 267 Brennan, J., dissenting valid gross-receipts tax instead of the challenged income tax. Of more significance, the gross-receipts tax sustained in Standard Pressed Steel and General Motors Corp. v. Washington, 377 LT. S. 436, is inherently more burdensome than the Iowa income tax. It applies whether or not the interstate concern is profitable and its imposition may make the difference between profit and loss. In contrast, the income tax is only imposed on enterprises showing a profit and the tax obligation is not heavy unless the profits are high. Accordingly, until Congress prescribes a different rule, Iowa is not constitutionally prohibited from requiring taxpayers to prove that application of the single-factor formula has produced arbitrary results in a particular case. The judgment of the Iowa Supreme Court is affirmed. So ordered. Mr. Justice Brennan, dissenting. I agree with the Court that, for purposes of constitutional review, there is no distinction between a corporate income tax and a gross-receipts tax. I do not agree, however, that Iowa’s single-factor sales apportionment formula meets the Commerce Clause requirement that a State’s taxation of interstate business must be “fairly apportioned to the commerce carried on within the taxing state.” Western Live Stock v. Bureau of Revenue, 303 U. S. 250, 256 (1938). As I have previously explained: “[Where a sale] exhibits significant contacts with more than one State ... it is the commercial activity within the State, and not the sales volume, which determines the State’s power to tax, and by which the tax must be apportioned. While the ratio of in-state to out-of-state sales is often taken into account as one factor among others in apportioning a firm’s total net income, see, e. g., the description of the ‘Massachusetts Formula’ in Note, 75 Harv. L. Rev. 953, 1011 (1962), it nevertheless remains true that 282 OCTOBER TERM, 1977 Blackmun, J., dissenting 437 U. S. if commercial activity in more than one State results in a sale in one of them, that State may not claim as all its own the gross receipts to which the activity within its borders has contributed only a part. Such a tax must be apportioned to reflect the business activity within the taxing State.” General Motors Corp. n. Washington, 377 U. S. 436,450—451 (1964) (dissenting opinion). I would therefore reverse. Mr. Justice Blackmun, dissenting. The unspoken, but obvious, premise of the majority opinion is the fear that a Commerce Clause invalidation of Iowa’s single-factor sales formula will lead the Court into problems and difficulties in other cases yet to come. I reject that premise. I agree generally with the content of Mr. Justice Powell’s opinion in dissent. I join that opinion because I, too, feel that the Court has a duty to resolve, not to avoid, these problems of “delicate adjustment,” Boston Stock Exchange v. State Tax Comm’n, 429 U. S. 318, 329 (1977), and because the opinion well demonstrates that Iowa’s now anachronistic single-factor sales formula runs headlong into overriding Commerce Clause considerations and demands. Today’s decision is bound to be regressive.1 Single-factor formulas are relics of the early days of state income taxation.2 The three-factor formulas were inevitable improvements and, while not perfect, reflect more accurately the realities of the business and tax world. With their almost universal adoption by the States, the Iowa system’s adverse and parochial impact on commerce comes vividly into focus. But with its 1 Iowa is not a member of the Multistate Tax Commission. Tr. of Oral Arg. 33. See United States Steel Corp. v. Multistate Tax Comm’n, 434 U. S. 452 (1978). 2 Iowa’s income tax was first adopted in 1934. 1933-1934 Iowa Acts, Ex. Sess., ch. 82; Tr. of Oral Arg. 29. Its single-factor sales formula was embraced in § 28 of that original Act. MOORMAN MFG. CO. v. BAIR 283 267 Powell, J., dissenting single-factor formula now upheld by the Court, there is little reason why other States, perceiving or imagining a similar advantage to local interests, may not go back to the old ways. The end result, in any event, is to exacerbate what the Commerce Clause, absent governing congressional action, was devised to avoid. Mr. Justice Powell, with whom Mr. Justice Blackmun joins, dissenting. It is the duty of this Court “to make the delicate adjustment between the national interest in free and open trade and the legitimate interest of the individual States in exercising their taxing powers.” Boston Stock Exchange v. State Tax Comm’n, 429 U. S. 318, 329 (1977). This duty must be performed with careful attention to the settings of particular cases and consideration of their special facts. See Raymond Motor Transp., Inc. v. Rice, 434 U. S. 429, 447-448, n. 25 (1978). Consideration of all the circumstances of this case leads me to conclude that Iowa’s use of a single-factor sales formula to apportion the net income of multistate corporations results in the imposition of “a tax which discriminates against interstate commerce ... by providing a direct commercial advantage to local business.” Northwestern States Portland Cement Co. v. Minnesota, 358 U. S. 450, 458 (1959). I therefore dissent. I Iowa’s use of single-factor sales-apportionment formula— though facially neutral—operates as a tariff on goods manufactured in other States and as a subsidy to Iowa manufacturers selling their goods outside of Iowa. Because 44 of the 45 other States (including the District of Columbia) which impose corporate income taxes use a three-factor formula involving property, payroll, and sales,1 Iowa’s practice insures that out- 1 Those 44 States are as follows: Alabama, Alaska, Arizona, Arkansas, California, Colorado, Connecticut, Delaware, District of Columbia, Florida, 284 OCTOBER TERM, 1977 Powell, J., dissenting 437 U. S. of-state businesses selling in Iowa will have higher total tax payments than local businesses. This result follows from the fact that Iowa attributes to itself all of the income derived from sales in Iowa, while other taxing States—using the three-factor formula—are also taxing some portion of the same income through attribution to property or payroll in those States. This surcharge on Iowa sales increases to the extent that a business’ plant and labor force are located outside Iowa. It can be avoided altogether only by locating all property and payroll in Iowa; an Iowa manufacturer selling only in Iowa will never have any portion of its income attributed to any other State. And to the extent that an Iowa manufacturer makes its sales in States other than Iowa, its overall state tax liability will be reduced. Assuming comparable tax rates, its liability to other States, in which sales constitute only one-third of the apportionment formula, will be far less than the amount it would have owed with a comparable volume of sales in Iowa, where sales are the exclusive mode of apportioning income. The effect of Iowa’s formula, then, is to penalize out-of-state manufacturers for selling in Iowa and to subsidize Iowa manufacturers for selling in other States.2 Georgia, Hawaii, Idaho, Illinois, Indiana, Kansas, Kentucky, Louisiana, Maine, Maryland, Massachusetts, Michigan, Minnesota, Mississippi, Missouri, Montana, Nebraska, New Hampshire, New Jersey, New Mexico, New York, North Carolina, North Dakota, Ohio, Oklahoma, Oregon, Pennsylvania, Rhode Island, South Carolina, Tennessee, Utah, Vermont, Virginia, and Wisconsin. West Virginia, the 45th State, uses a two-factor formula which omits the sales component. Colorado also has a two-factor property and sales formula, and Missouri a one-factor sales formula, which are available to taxpayers at their option as alternatives to the three-factor formula. 2 A simplified example demonstrates the economic effect of the Iowa formula on out-of-state corporations. Iowa Corp, is domiciled in Iowa, and its total property and payroll are located there. Illinois Corp, is domiciled in Illinois, with all its property and payroll in that State. Both corporations have $1 million in net income, MOORMAN MFG. CO. v. BAIR 285 267 Powell, J., dissenting This appeal requires us to determine whether these economic effects of the Iowa apportionment formula violate either the Due Process Clause or the Commerce Clause. I now turn to those questions. and both make half their sales in Iowa and half in Illinois. A 5% corporate income tax is levied in both States. If both States use a single-factor sales apportionment formula, both would go through the following calculation in determining the tax liability of both corporations: Sales in States ------------------------= %; i/2x $1,000,000X 0.05=425,000 Total Sales The pattern of payments and receipts would be as follows: Total Taxes Taxes Paid to Iowa Taxes Paid to Illinois Paid by each Corporation Illinois Corp. Iowa Corp. $25,000 25,000 $25,000 25,000 $50,000 50,000 TOTAL 50,000 50,000 If both Iowa and Illinois again levy the same 5% the three-factor formula, which is: Sales in Property in Payroll in State State State income tax but use Total Sales Total Property Total Payroll 3 then each corporation’s payment to its state of domicile would be 0.5+1+1 x$1 000x 0 os—$41667, 3 its payment to the state in which it is a foreign corporation would be 0.5+0+0 ----------X$l,000,000x 0.05=$8,333. 3 The pattern of tax payments and receipts would be as follows: Total Taxes Paid by each Corporation Taxes Paid to Iowa Taxes Paid to Illinois Iowa Corp. $41,667 $8,333 $50,000 Illinois Corp. 8,333 41,667 50,000 TOTAL 50,000 50,000 But where Iowa uses a single-factor sales formula and Illinois uses the 286 OCTOBER TERM, 1977 Powell, J., dissenting 437U.S. II For the reasons given by the Court, ante, at 271-275,1 agree that application of Iowa’s formula does not violate the Due Process Clause. The decisions of this Court make it clear that arithmetical perfection is not to be expected from apportionment formulae. International Harvester Co. v. Evatt, 329 U. S. 416 (1947). It has been said that the “apportionment theory is a mongrel one, a cross between desire not to interfere with state taxation and desire at the same time not utterly to crush out interstate commerce.” Northwest Airlines, Inc. v. Minnesota, 322 U. S. 292, 306 (1944) (Jackson, J., concurring). It owes its existence to the fact that with respect to a business earning income through a series of transactions beginning with manufacturing in one State and ending with a sale in another, a precise—or even wholly logical—determination of the State in which any specific portion of the income was earned is impossible. Underwood Typewriter Co. v. Chamberlain, 254 U. S. 113,120-121 (1920). Hence, the fact that a particular formula—like the one at issue here—may permit a State to tax some income actually “located” in another State is not in and of itself a basis for three-factor method, Illinois Corp, faces an increase in its overall state tax liability not encountered by Iowa Corp.: Total Taxes Taxes Paid Taxes Paid Paid by each to Iowa to Illinois Corporation Iowa Corp. $25,000 $8,333 $33,333 Illinois Corp. 25,000 41,667 66,667 TOTAL 50,000 50,000 These differences will be smaller or larger, depending upon the actual tax rates of the various States involved, and upon the actual proportions of domestic to foreign sales, the payrolls, and the properties of individual corporations. Only the magnitudes will change with these factors, however, and not the direction of the impact. The general principle will apply in all cases. MOORMAN MFG. CO. v. BAIR 287 267 Powell, J., dissenting finding a due process violation.3 Were it otherwise, any formula deviating in the smallest detail from that used in other States would be invalid. Because there is no ideal means of “locating” any State’s rightful share, such uniformity cannot be dictated by this Court. Hence, the decisions of this Court properly require the taxpayer claiming a due process violation to show that the apportionment is “out of all appropriate proportion to the business transacted.” Hans Rees’ Sons, Inc. v. North Carolina ex rel. Maxwell, 283 U. S. 123, 135 (1931). As appellant has failed to make any such showing, I agree with the Court that no due process violation has been made out here. This conclusion does not ipso facto mean that Commerce Clause strictures are satisfied as well. This Court’s decisions dealing with state levies that discriminate against out-of-state business, as Iowa’s formula does, compel a more detailed inquiry. Ill A It is a basic principle of Commerce Clause jurisprudence that “[n]either the power to tax nor the police power may be 3 This does not mean, as the Court suggests, ante, at 277-280, that this Court is disabled from ever determining whether a particular apportionment formula imposes multiple burdens upon or discriminates against interstate commerce. See General Motors Corp. v. District of Columbia, 380 U. S. 553 (1965); Bass, Ratcliff & Gretton, Ltd. v. State Tax Comm’n, 266 U. S. 271 (1924); Underwood Typewriter Co. v. Chamber-lain, 254 U. S. 113 (1920). Regardless of which formula more accurately locates the State in which any particular segment of income is earned, it is a mathematical fact that the use of different formulae may result in taxation on more than 100% of the corporation’s income under the State’s own definitions, as well as in skewed tax effects. See n. 2, supra. When this result has a predictably burdensome or discriminatory effect, Commerce Clause scrutiny is triggered. See Part III, infra. The effects of the challenged formula upon the particular corporation’s income is strictly related only to inquiry under the Due Process Clause, since Commerce Clause analysis focuses on the impact upon commerce in general. 288 OCTOBER TERM, 1977 Powell, J., dissenting 437U.S. used by the state of destination with the aim and effect of establishing an economic barrier against competition with the products of another state or the labor of the residents.” Baldwin v. G. A. F. Seelig, Inc., 294 U. S. 511, 527 (1935); accord, H. P. Hood Ac Sons v. Du Mond, 336 *U. S. 525, 532 (1949); Boston Stock Exchange, 429 U. S., at 335-336, and n. 14. Those barriers would constitute “an unreasonable clog upon the mobility of commerce.” Baldwin, supra, at 527. One form of such unreasonable restrictions is “discriminating State legislation.” Welton v. Missouri, 91 U. S. 275, 280 (1876). This Court consistently has struck down state and local taxes which unjustifiably benefit local businesses at the expense of out-of-state businesses. Ibid.; accord, Boston Stock Exchange; Halliburton Oil Well Co. v. Reily, 373 U. S. 64 (1963); Nippert v. Richmond, 327 U. S. 416 (1946); Hale v. Bimco Trading, Inc., 306 U. S. 375 (1939); I. M. Darnell & Son v. Memphis, 208 U. S. 113 (1908); Guy v. Baltimore, 100 U. S. 434 (1880). This ban applies not only to state levies that by their terms are limited to products of out-of-state business, or which explicitly tax out-of-state sellers at higher rates than local sellers. It also reaches those taxes that “in their practical operation [work] discriminatorily against interstate commerce to impose upon it a burden, either in fact or by the very threat of its incidence.” Nippert v. Richmond, supra, at 425. For example, this Court has invalidated a facially neutral fixed-fee license tax collected from all local and out-of-state “drummers,” where it appeared the tax fell far more heavily upon out-of-state businesses, since local businesses had little or no occasion to solicit sales in that manner. Robbins v. Shelby County Taxing Dist., 120 U. S. 489 (1887). See also West Point Wholesale Grocery Co. v. Opelika, 354 U. S. 390 (1957); Memphis Steam Laundry Cleaner, Inc. v. Stone, 342 U. S. 389 (1952); Best <& Co. v. Maxwell, 311 U. S. 454 (1940); Real MOORMAN MFG. CO. v. BAIR 289 267 Powell, J., dissenting Silk Hosiery Mills v. Portland, 268 U. S. 325 (1925); Corson v. Maryland, 120 U. S. 502 (1887). Thus, the constitutional inquiry relates not simply to the form of the particular tax, but to its effect on competition in the several States. As indicated in Part I above, application of Iowa’s singlefactor sales-apportionment formula, in the context of general use of three-factor formulae, inevitably handicaps out-of-state businesses competing for sales in Iowa. The handicap will diminish to the extent that the corporation locates its plant and labor force in Iowa, but some competitive disadvantage will remain unless all of the corporate property and payroll are relocated in Iowa.4 In the absence of congressional action, the Commerce Clause constrains us to view the State’s interest in retaining this particular levy as against the constitutional preference for an open economy. See, e. g., Raymond Motor Transp., Inc. v. Rice, 434 U. S., at 440-442; Pike v. Bruce Church, Inc., 397 U. S. 137, 142 (1970); Di Santo v. Pennsylvania, 273 U. S. 34, 44 (1927) (Stone, J., dissenting); Dowling, Interstate Commerce and State Power, 27 Va. L. Rev. 1, 14-15, and n. 20 (1940). 4 The clog on commerce present here is similar to the risk of imposing "multiple burdens” on interstate commerce against which the Court has warned in various decisions. See, e. g., Western Live Stock v. Bureau of Revenue, 303 U. S. 250, 255-256 (1938); J. D. Adams Mfg. Co. v. Stören, 304 U. S. 307, 311-312 (1938); Gwin, White & Prince, Inc. v. Hennejord, 305 U. S. 434, 439 (1939); Northwestern States Portland Cement Co. v. Minnesota, 358 U. S. 450, 458 (1959). Compare Evco v. Jones, 409 U. S. 91 (1972), with General Motors Corp. v. Washington, 377 U. S. 436 (1964). In this case, Iowa corporations will not risk additional burdens when they make out-of-state sales. Cf. Hunt v. Washington Apple Advertising Comm’n, 432 U. S. 333, 351 (1977). Indeed, to the extent that they shift sales out of Iowa, their overall state tax liability will decrease. Out-of-state corporations selling in Iowa, however, do face the prospect of multiple burdens. Hence, there is clear discrimination against out-of-state corporations, which is the consequence of the particular multiple burden imposed. 290 OCTOBER TERM, 1977 Powell, J., dissenting 437U.S. B Iowa’s interest in any particular level of tax revenues is not affected by the use of the single-factor sales formula. It cannot be predicted with certainty that its application will result in higher revenues than any other formula.5 If Iowa needs more revenue, it can adjust its tax rates. That adjustment would not have the discriminatory impact necessarily flowing from the choice of the single-factor sales formula.6 Hence, if Iowa’s choice is to be sustained, it cannot be by virtue of the State’s interest in protecting its fisc or its power to tax. No other justification is offered. If we are to uphold Iowa’s apportionment formula, it must be because no consistent principle can be developed that could account for the invalidation of the Iowa formula, yet support application of other States’ imprecise formulae. 5 For example, if Iowa switched to a three-factor formula and retained the same rates, revenues from out-of-state corporations would decrease, since Iowa would no longer be attributing to itself all of the income earned by Iowa sales of such corporations. Revenues from corporations located in Iowa, however, would increase, since Iowa would now be attributing to itself some portion of the income earned by those corporations’ out-of-state sales. See also n. 2, supra. 6 Given the nearly infinite variety of taxes, rates, and apportionment formulae, it might be possible for Iowa to alter its entire tax structure to effect a similar discrimination, and perhaps to do it in a way that avoids Commerce Clause scrutiny. See Barrett, “Substance” vs. “Form” in the Application of the Commerce Clause to State Taxation, 101 U. Pa. L. Rev. 740, 748 (1953). That speculative possibility cannot deter us from striking down an obvious discrimination against interstate commerce when one is presented. The Court has never shrunk from that duty in the past. To do so would be to abandon any effort of applying Commerce Clause principles to state tax measures. This is not to say that States are always forbidden to offer tax incentives to encourage local industry or to achieve other valid state goals. See, e. g., Hughes n. Alexandria Scrap Corp., 426 U. S. 794 (1976). Such programs, and the interests being served, must be considered on a case-by-case basis. MOORMAN MFG. CO. v. BAIR 291 267 Powell, J., dissenting c It is argued that since this Court on several occasions has upheld the use of single-factor formulae, Iowa’s scheme cannot be regarded as suspect simply because it does not embody the prevalent three-factor theory. Consideration of the decisions dealing with single-factor formulae, however, reveals that each is distinguishable. In Underwood Typewriter Co. v. Chamberlain, 254 U. S. 113 (1920), this Court upheld Connecticut’s use of a singlefactor property formula to apportion the net profits of a foreign corporation. Such a formula is not clearly discriminatory in Commerce Clause terms. The only competitive disadvantage inevitably resulting from it would attend a decision to locate a plant or office in the taxing State. The Commerce Clause does not concern itself with a State’s decision to place local business at a disadvantage. Cf. Allied Stores of Ohio, Inc. v. Bowers, 358 U. S. 522, 528 (1959). Bass, Ratcliff & Gretton, Ltd. v. State Tax Comm’n, 266 LT. S. 271 (1924), is similarly distinguishable. In Bass, New York apportioned the net income of foreign corporations using a single-factor property formula that comprised real and tangible personal property, bills and accounts receivable, and stock in other corporations. This Court upheld that formula, observing that plaintiff in error had not shown that “application of the statutory method of apportionment has produced an unreasonable result.” Id., at 283. As in Underwood Typewriter, however, the single-factor property formula did not necessarily discriminate against businesses carried on out of State; indeed, its impact would tend to increase to the extent that corporate business was carried on within the State. Cf. National Leather Co. v. Massachusetts, 277 U. S. 413 (1928); accord, e. g., International Shoe Co. v. Shartel, 279 U. S. 429 (1929); New York v. Latrobe, 279 U. S. 421 (1929); Hump Hairpin Co. v. Emmerson, 258 U. S. 290 (1922); United States Glue Co. v. Oak Creek, 247 U. S. 321 (1918). 292 OCTOBER TERM, 1977 Powell, J., dissenting 437U.S. Somewhat more troublesome is Ford Motor Co. v. Beauchamp, 308 U. S. 331 (1939). In that case, the Court sustained Texas’ use of a single-factor sales formula to apportion the outstanding capital stock, surplus, undivided profits, and longterm obligations of corporations subject to the state franchise tax. While this case may be seen as standing for the proposition that single-factor sales formulae are not per se illegal, it is not controlling in the present case.7 In Ford Motor Co., as in Underwood Typewriter and Bass, there was no showing of virtually universal use of a conflicting type of formula for determining the same tax. Thus, it could not be said that the Texas formula inevitably imposed a competitive disadvantage on out-of-state corporations. Discrimination not being shown, there was no basis for invalidating the Texas scheme under the Commerce Clause. The opposite is true here. In the context of virtually universal use of the basic three-factor formula, Iowa’s use of the single-factor sales formula necessarily discriminates against out-of-state manufacturers. The only remaining question, then, is whether Iowa’s scheme may be saved by the fact that its discriminatory nature depends on context: If other States were not virtually unanimous in their use of an opposing 7 Although overruling Ford Motor Co. would not be necessary in this case, the time may be ripe for its reconsideration. See, e. g., J. Hellerstein, State and Local Taxation 324 (3d ed. 1969). As suggested in General Motors Corp. v. District of Columbia, 380 U. S. 553, 561 (1965), a sales-only formula is probably the most illogical of all apportionment methods, since “the geographic distribution of a corporation’s sales is, by itself, of dubious significance in indicating the locus of either” a corporation’s sources of income or the social costs it generates. The Court’s willingness to uphold the sales-only formula in Ford Motor Co. may have been the result of its view that it was dealing solely with the “measure” of the tax rather than its “subject.” See 308 U. S7 at 336. This Court no longer adheres to the use of those formalistic labels, looking instead to “economic realities” in determining the constitutionality of state taxing schemes. Complete Auto Transit, Inc. v. Brady, 430 U. S. 274, 279 (1977). MOORMAN MFG. CO. v. BAIR 293 267 Powell, J., dissenting formula, past decisions would make it difficult to single out Iowa’s scheme as more offensive than any other. D On several occasions, this Court has compared a state statutory requirement against the practice in other States in determining the statute’s validity under the Commerce Clause. In Southern Pacific Co. v. Arizona ex rd. Sullivan, 325 U. S. 761 (1945), the Court struck down a state statute limiting passenger trains to 14 cars and freight trains to 70 cars. Noting that only one State other than Arizona enforced a restriction on train lengths,8 the Southern Pacific Court specifically considered the Arizona law against the background of the activities in other States: “Enforcement of the law in Arizona, while train lengths remain unregulated or are regulated by varying standards in other states, must inevitably result in an impairment of uniformity of efficient railroad operation because the railroads are subjected to regulation which is not uniform in its application. Compliance with a state statute limiting train lengths requires interstate trains of a length lawful in other states to be broken up and reconstituted as they enter each state according as it may impose varying limitations upon train lengths. The alternative is for the carrier to conform to the lowest train limit restriction of any of the states through which its trains pass, whose laws thus control the carriers’ operations both within and without the regulating state.” Id., at 773. (Emphasis added.) The clear implication is that the Court’s view of the Arizona length limit might have been different if practices in other States had been other than as the Court found them. Had 8 That State was Oklahoma. Southern Pacific Co. v. Arizona ex rel. Sullivan, 325 U. S., at 773-774, n. 3. 294 OCTOBER TERM, 1977 Powell, J., dissenting 437 U. S. other States adopted the Arizona rule, there might have been no basis for holding it unconstitutional. See also Morgan v. Virginia, 328 U. S. 373 (1946) • Hall v. DeCuir, 95 U. S. 485 (1878). The Court also looked to the practices of other States in holding unconstitutional Illinois’ mudguard requirement in Bibb v. Navajo Freight Lines, Inc., 359 U. S. 520 (1959). The type of mudguard banned on trucks operating in Illinois was required in Arkansas and permitted in 45 other States. The Court pointed out the conflict between the Illinois and Arkansas regulations and went on to consider the relevance of other States’ rules: “A State which insists on a design out of line with the requirements of almost all the other States may sometimes place a great burden of delay and inconvenience on those interstate motor carriers entering or crossing its territory. Such a new safety device—out of line with the requirements of the other States—may be so compelling that the innovating State need not be the one to give way. But the present showing—balanced against the clear burden on commerce—is far too inconclusive to make this mudguard meet that test.” Id., at 529-530. It seems clear from the Bibb Court’s discussion that the conflict between the Illinois regulation and that of Arkansas would not have led to the latter’s invalidation had it been the one before the Court. The Arkansas regulation merely required what was permitted in nearly all the other States. After looking to that virtually uniform practice opposed to that of Illinois, the conclusion that the Illinois requirement was “out of line” was a relatively simple one. Since it was not justified by any interest in increased safety, it was held unconstitutional. See also Raymond Motor Transp., Inc. v. Rice, 434 U. S., at 444 -446. Most nearly in point is General Motors Corp. v. District of Columbia, 380 U. S. 553 (1965). In that case, this Court held MOORMAN MFG. CO. v. BAIR 295 267 Powell, J., dissenting unlawful the District’s use of a single-factor sales apportionment formula under the District of Columbia Income and Franchise Tax Act of 1947. Although the decision turned on a question of statutory interpretation, the Court’s analysis is equally applicable to a Commerce Clause inquiry: “The great majority of States imposing corporate income taxes apportion the total income of a corporation by application of a three-factor formula which gives equal weight to the geographical distribution of plant, payroll, and sales. The use of an apportionment formula based wholly on the sales factor, in the context of general use of the three-factor approach, will ordinarily result in multiple taxation of corporate net income .... In any case, the sheer inconsistency of the District formula with that generally prevailing may tend to result in the unhealthy fragmentation of enterprise and an uneconomic pattern of plant location, and so presents an added reason why this Court must give proper meaning to the relevant provisions of the District Code.” Id., at 559-560 (footnote omitted). The General Motors Court, then, expressly evaluated the single-factor sales formula in the context of general use of the three-factor method and concluded that the former created dangers for interstate commerce. These cases lead me to believe that it is not only proper but essential to determine the validity of the Iowa formula against the background of practices in the other States. If one State’s regulatory or taxing statute is significantly “out of line” with other States’ rules, Bibb, supra, at 530, and if by virtue of that departure from the general practice it burdens or discriminates against interstate commerce, Commerce Clause scrutiny is triggered, and this Court must invalidate it unless it is justified by a legitimate local purpose outweighing the harm to interstate commerce, Pike v. Bruce Church, Inc., 397 U. S., at 142; accord, Hughes v. Alexandria Scrap Corp., 426 U. S. 794, 804 (1976). There probably can be no fixed rule 296 OCTOBER TERM, 1977 Powell, J., dissenting 437U.S. as to how nearly uniform the countervailing state policies must be; that is, there can be no rule of 26 States, of 35, or of 45. Commerce Clause inquiries generally do not run in such precise channels. The degree of conflict and its resulting impact on commerce must be weighed in the circumstances of each case. But the difficulty of engaging in that weighing process does not permit this Court to avoid its constitutional duty and allow an individual State to erect “an unreasonable clog upon the mobility of commerce,” Baldwin v. G. A. F. Seelig, Inc., 294 U. S., at 527, by taking advantage of the other States’ commendable trend toward uniformity. Such is the case before us. Forty-four of the forty-five States (including the District of Columbia), other than Iowa, that impose a corporate income tax utilize a similar three-factor apportionment formula.9 The 45th State, West Virginia, uses a two-factor formula based on property and payroll. See n. 1, supra. Those formulae individually may be no more rational as means of apportioning the income of a multistate business than Iowa’s single-factor sales formula. But see General Motors Corp. n. District of Columbia, supra, at 561. Past decisions upheld differing formulae because of this inability to determine that any of the various methods of apportionment in use was the best; so long as a State’s choice was not shown to be grossly unfair, it would be upheld. Com- 9 There are differences in definitions of the three factors among the States that use a three-factor formula. See, e. g., J. Hellerstein, State and Local Taxation 309-310, and n. 7 (3d ed. 1969); Note, State Taxation of Interstate Businesses and the Multistate Tax Compact: The Search for a Delicate Uniformity, 11 Colum. J. of Law & Soc. Prob. 231, 235-238 (1975). Such differences may tend in less dramatic fashion to impose burdens on out-of-state businesses not entirely dissimilar to the one presented here. It may be that any such effects do not work inevitably in one direction, as does the burden imposed here, or they may be de minimis in Commerce Clause terms. In any event, they are not presently before us. It suffices to dispose of this case that nearly all the other States use a basic three-factor formula, while Iowa clings to its sales-only method. MOORMAN MFG. CO. v. BAIR 297 267 Powell, J., dissenting pare Underwood Typewriter with Hans Rees’ Sons. The more recent trend toward uniformity, however, permits identification of Iowa’s formula, like the mudguard requirement in Bibb, as “out of line,” if not per se irrational. Since Iowa’s formula inevitably discriminates against out-of-state sellers, and since it has not been justified on any fiscal or administrative basis, I would hold it invalid under the Commerce Clause. 298 OCTOBER TERM, 1977 Syllabus 437 U. S. UNITED STATES et al. v. LaSALLE NATIONAL BANK ET AL. CERTIORARI TO THE UNITED STATES COURT OF APPEALS FOR THE SEVENTH CIRCUIT No. 77-365. Argued March 29, 1978—Decided June 19, 1978 Petitioner special agent of the Internal Revenue Service (IRS), in the process of investigating a taxpayer’s tax liability, issued summonses to respondent bank under authority of § 7602 of the Internal Revenue Code of 1954 (which permits use of a summons “[f]or the purpose of ascertaining the correctness of any return, . . . determining the liability of any person for any internal revenue tax . . . , or collecting any such liability”) to appear before the agent and produce files of certain land trusts, created for the benefit of the taxpayer. When respondent bank official appeared in response to the summonses but refused to produce the files, the United States and the agent petitioned the District Court for enforcement of the summonses. That court denied enforcement, finding that the summonses were not issued in good faith because they were issued “solely for the purpose of unearthing evidence of criminal conduct” by the taxpayer. The Court of Appeals affirmed. Held: The District Court erred in refusing to enforce the summonses, since its finding that the agent was investigating the taxpayer “solely for the purpose of unearthing evidence of criminal conduct” does not necessarily lead to the conclusion that the summonses were not issued in good-faith pursuit of the congressionally authorized purposes of § 7602. Pp. 307-319. (a) Congress has not categorized tax fraud investigation into civil and criminal components but has created a tax enforcement system in which criminal and civil elements are inherently intertwined, and any limitation on the good-faith use of an IRS summons must reflect this statutory premise. Pp. 308-311. (b) To enforce a summons under § 7602, the primary requirement is that it be issued before the IRS recommends to the Department of Justice the initiation of a criminal prosecution relating to the subject matter of the summons. This is a prophylactic rule designed to protect the standards of criminal litigation discovery and the role of the grand jury as a principal tool of criminal accusation. Pp. 311-313. (c) Enforcement of a summons is also conditioned upon the good-faith use of the summons authority by the IRS, which must not abandon its UNITED STATES v. LaSALLE NATIONAL BANK 299 298 Opinion of the Court institutional responsibility to determine and to collect taxes and civil fraud penalties. That a single special agent intends only to gather evidence for a criminal investigation is not dispositive of the good faith of the IRS as an institution. Those resisting enforcement of a summons must disprove the actual existence of a valid civil tax determination or collection purpose by the IRS. Pp. 313-317. (d) On the record here respondents have not shown sufficient justification to preclude enforcement of the summonses in question, absent any recommendation to the Justice Department for criminal prosecution and absent any showing that the special agent already possessed all of the evidence sought in the summonses or that the IRS in an institutional sense had abandoned pursuit of the taxpayer’s civil tax liability. Pp. 318-319. 554 F. 2d 302, reversed with directions to remand. Blackmun, J., delivered the opinion of the Court, in which Brennan, White, Marshall, and Powell, JJ., joined. Stewart, J., filed a dissenting opinion, in which Burger, C. J., and Rehnquist and Stevens, JJ., joined, post, p. 319. Deputy Solicitor General Wallace argued the cause for the United States et al. With him on the briefs were Solicitor General McCree, Assistant Attorney General Ferguson, Stuart A. Smith, Robert E. Lindsay, Charles E. Brookhart, and Carleton D. Powell. Matt P. Cushner argued the cause for respondents. With him on the brief was Gregory J. Perry. Mr. Justice Blackmun delivered the opinion of the Court. This case is a supplement to our decision in Donaldson v. United States, 400 U. S. 517 (1971). It presents the issue whether the District Court correctly refused to enforce Internal Revenue Service summonses when it specifically found that the special agent who issued them “was conducting his investigation solely for the purpose of unearthing evidence of criminal conduct.” 76-1 USTC U 9407, p. 84,073, 37 AFTR 2d fl 76-582, p. 76-1240 (ND Ill. 1976). 300 OCTOBER TERM, 1977 Opinion of the Court 437U.S. I In May 1975, John F. Olivero, a special agent with the Intelligence Division of the Chicago District of the Internal Revenue Service (hereinafter IRS or Service), received an assignment to investigate the tax liability of John Gattuso for his taxable years 1970-1972. App. 26—27, 33. Olivero testified that he had requested the assignment because of information he had received from a confidential informant and from an unrelated investigation. Id., at 35. The case was not referred to the IRS from another law enforcement agency, but the nature of the assignment, Olivero testified, was “[t]o investigate the possibility of any criminal violations of the Internal Revenue Code.” Id., at 33. Olivero pursued the case on his own, without the assistance of a revenue agent.1 He received information about Gattuso from the Federal Bureau of Investigation as a result of the previous investigation. Id., at 36. He solicited and received additional data from the United States Attorney for the Northern District of Illinois, the Secret Service, the Department of Housing and Urban Development, the IRS Collection Division, and the Cosmopolitan National Bank of Chicago. Id., at 37-40. Mr. Gattuso’s tax returns for the years in question disclosed rental income from real estate. That property was held in 1 Frequently, a revenue agent of the IRS Audit Division will refer a case on which he is working to the Intelligence Division for investigation of possible fraud. After such a referral, and at other times, the special agent and the revenue agent work together. Because of the importance and sensitivity of the criminal aspects of the joint investigation, the special agent assumes control of the inquiry. See, e. g., Internal Revenue Manual, ch. 4500, §§ 4563.431-4565.44 (CCH 1976 and 1978). As part of a planned reorganization, the IRS has announced its intention to redesignate the Audit Division and the Intelligence Division as the Examinations Division and the Criminal Enforcement Division, respectively. IRS News Release, Feb. 6, 1978. UNITED STATES v. LaSALLE NATIONAL BANK 301 298 Opinion of the Court Illinois land trusts2 by respondent LaSalle National Bank, as trustee, a fact revealed by land trust files collected by the IRS from banks. Id., at 27, 45. In order to determine the accuracy of Gattuso’s income reports, Olivero proceeded to issue two summonses, under the authority of § 7602 of the Internal Revenue Code of 1954, 26 U. S. C. § 7602,3 to respondent bank. Each summons related to a separate trust and requested, among other things, that the bank as trustee appear before Olivero at a designated time and place and produce its “files relating to Trust No. 31544 [or No. 35396] 2 Respondents describe an Illinois land trust as follows: “An Illinois land trust is a contract by which a trustee is vested with both legal and equitable title to real property and the interest of the beneficiary is considered personal property. Under this trust the beneficiary or any person designated in writing by the beneficiary has the exclusive power to direct or control the trustee in dealing with the title and the exclusive control of the management, operation, renting and selfing of the trust property together with the exclusive right to the earnings, avails and proceeds of said property. Ill. Rev. Stat. ch. 29, § 8.31 (1971)Brief for Respondents 1-2, n. 1. 3 Section 7602 reads: “For the purpose of ascertaining the correctness of any return, making a return where none has been made, determining the liability of any person for any internal revenue tax or the liability at law or in equity of any transferee or fiduciary of any person in respect of any internal revenue tax, or collecting any such liability, the Secretary or his delegate is authorized— “(1) To examine any books, papers, records, or other data which may be relevant or material to such inquiry; “(2) To summon the person liable for tax or required to perform the act, or any officer or employee of such person, or any person having possession, custody, or care of books of account containing entries relating to the business of the person liable for tax or required to perform the act, or any other person the Secretary or his delegate may deem proper, to appear before the Secretary or his delegate at a time and place named in the summons and to produce such books, papers, records, or other data, and to give such testimony, under oath, as may be relevant or material to such inquiry; and “(3) To take such testimony of the person concerned, under oath, as may be relevant or material to such inquiry.” 302 OCTOBER TERM, 1977 Opinion of the Court 437U.S. including the Trust Agreement” for the period 1970 through 1972 and also “all deeds, options, correspondence, closing statements and sellers statements, escrows, and tax bills pertaining to all property held in the trust at any time during” that period. App. 9-16. Respondent Joseph W. Lang, a vice president of the bank, appeared in response to the summonses but, on advice of counsel, refused to produce any of the materials requested. Brief for Respondents 2. The United States and Olivero, pursuant to §§ 7402 (b) and 7604 (a) of the Code, 26 U. S. C. §§ 7402 (b) and 7604 (a),4 then petitioned the United States District Court for the Northern District of Illinois for enforcement of the summonses. App. 5. This was on November 11, 1975. Olivero testified that when the petition was filed he had not determined whether criminal charges were justified and had not made any report or recommendation about the case to his superiors. Id., at 30. It was alleged in the petition and in an incorporated exhibit that the requested materials were necessary for the determination of the tax liability of Gattuso for the years in question and that the information contained in the documents was not in the possession of the petitioners. Id., at 7, 17-18. The District Court entered an order to show cause, id., at 19, and respondents answered through counsel, who also represented Gattuso. Id., at 20-22. 4 Section 7402 (b) states: “If any person is summoned under the internal revenue laws to appear, to testify, or to produce books, papers, or other data, the district court of the United States for the district in which such person resides or may be found shall have jurisdiction by appropriate process to compel such attendance, testimony, or production of books, papers, or other data.” Section 7604 (a) reads: “If any person is summoned under the internal revenue laws to appear, to testify, or to produce books, papers, records, or other data, the United States district court for the district in which such person resides or is found shall have jurisdiction by appropriate process to compel such attendance, testimony, or production of books, papers, records, or other data.” UNITED STATES v. LaSALLE NATIONAL BANK 303 298 Opinion of the Court At the ensuing hearing and in a post-hearing brief, respondents argued that Olivero’s investigation was “purely criminal” in nature. Id., at 82. Gregory J. Perry, a lawyer specializing in federal taxation and employed by the same law firm that filed the answer, testified that in June 1975 Olivero told him that the Gattuso investigation “was strictly related to criminal violations of the Internal Revenue Code.” Id., at 52. Respondents conceded that they bore the burden of proving that enforcement of the summonses would abuse the court’s process, but they contended that they did not have to show “that there is no civil purpose to the Summons.” Id., at 87. Instead, they urged that their burden was to show that the summonses were not issued in good faith because “the investigation is solely for the purpose of gathering evidence for use in a criminal prosecution.” Id., at 77. The District Court agreed with respondents’ contentions. Although at the hearing the court seemed to recognize “that in any criminal investigation there’s always a probability of civil tax liability,” id., at 61, it focused its attention on the purpose of Special Agent Olivero: “I’ll say now that I heard nothing in Agent Olivero’s testimony to suggest that the thought of a civil investigation ever crossed his mind. “Now, unless I find something in the in camera inspection [of the IRS case file] that gives more support to the Government position than the Agent’s testimony did, it would be my conclusion that he was at all times involved in a criminal investigation, at least in his own mind.” 5 Id., at 62. 5 The District Court was aware of and recognized the Government’s contention that the individual agent’s motive in the investigation was not dispositive: “The COURT: . . . [U]nder your theory any criminal investigation would 304 OCTOBER TERM, 1977 Opinion of the Court 437U.S. In its written memorandum, the District Court noted that Donaldson permitted the use of an IRS summons issued in good faith and prior to a recommendation for criminal prosecution. Relying on dictum in Reisman v. Caplin, 375 U. S. 440, 449 (1964), however, the court said that it was an improper use of the summons “to serve it solely for the purpose of obtaining evidence for use in a criminal prosecution.” 76-1 USTC, at 84,072, 37 AFTR 2d, at 76-1240. If, at the time of its issuance, the summons served this proscribed purpose, the court concluded, the absence of a formal criminal recommendation was irrelevant, the summons was not issued in good faith, and enforcement was precluded. The court then held: “It is apparent from the evidence that Special Agent John F. Olivero in his investigative activities had focused upon the possible criminal activities of John Gattuso, and was conducting his investigation solely for the purpose of unearthing evidence of criminal conduct by Mr. Gattuso.” Id., at 84,073, 37 AFTR 2d, at 76-1240. The United States Court of Appeals for the Seventh Circuit affirmed. 554 F. 2d 302 (1977). It concluded that the District Court correctly had included the issue of criminal purpose within the good-faith inquiry: “[T]he use of an administrative summons solely for not really be one until they closed it because there was always a possibility of a civil liability. “If that’s the law, you’re in trouble, Mr. Cushner [counsel for respondents]. “I think it boils down to an issue of law so it’s the cases really that I’m interested in plus any further clues I may find in the in camera inspection of the investigative file.” App. 61-62. The court agreed to. inspect the IRS investigative file in camera after it refused to permit respondents to inspect the file. Id., at 50-51, 61-62. UNITED STATES v. LaSALLE NATIONAL BANK 305 298 Opinion of the Court criminal purposes is a quintessential example of bad faith. . . . “We note that the district court formulated its factual finding by use of the expression ‘sole criminal purpose’ rather than by a label such as ‘bad faith.’ We find no basis for reversible error in that verbal formulation. The district court grasped the vital core of Donaldson and rendered its factual finding consistently therewith.” Id., at 309. The Court of Appeals further decided that the District Court had reached a factual, rather than a legal, conclusion when it found the summonses to have been issued solely for a criminal prosecution. Id., at 305. Appellate review, accordingly, was limited to application of the clearly-erroneous standard. Id., at 306. Although the Court of Appeals noted that Olivero had testified about the existence of a civil purpose for the investigation, the court said that “the record establishes that the district court did not believe him.” Id., at 309. The appellate court could not reverse the trial court’s judgment, it said, because it was “not left with a firm and definite conviction that a mistake [had] been made.” Id., at 306. Because of the importance of the issue in the enforcement of the internal revenue laws, and because of conflict among the Courts of Appeals concerning the scope of IRS summons authority under § 7602,° we granted certiorari. 434 U. S. 996 (1977). 6 Compare United States v. Hodge & Zweig, 548 F. 2d 1347, 1350-1351 (CAO 1977); United States v. Zack, 521 F. 2d 1366, 1368 (CA9 1975); United States v. McCarthy, 514 F. 2d 368, 374-375 (CA3 1975); United States v. Weingarden, 473 F. 2d 454, 460 (CA6 1973); United States v. Wall Corp., 154 U. S. App. D. C. 309, 311, 475 F. 2d 893, 895 (1972); and United States v. Billingsley, 469 F. 2d 1208, 1210 (CAIO 1972), with United States v. Morgan Guaranty Trust Co., 572 F. 2d 36, 41-42 (CA2 1978); and United States v. Troupe, 438 F. 2d 117, 119 (CA8 1971), 306 OCTOBER TERM, 1977 Opinion of the Court 437U.S. II In Donaldson v. United States, 400 U. S. 517 (1971), an IRS special agent issued summonses to a taxpayer’s putative former employer and its accountant for the production of the employer’s records of the taxpayer’s employment and compensation. When the records were not forthcoming, the IRS petitioned for the enforcement of the summonses. The taxpayer intervened and eventually appealed the enforcement order. This Court addressed the taxpayer’s contention that the summonses were unenforceable because they were issued in aid of an investigation that could have resulted in a criminal charge against the taxpayer. His argument there, see id., at 532, was based on the following dictum in Reisman v. Caplin, 375 U. S., at 449: “[T]he witness may challenge the summons on any appropriate ground. This would include, as the circuits have held, the defenses that the material is sought for the improper purpose of obtaining evidence for use in a criminal prosecution, Boren v. Tucker, 239 F. 2d 767, 772-773 . ..y In the light of the citation to Boren,7 the Court in Donaldson concluded that the dictum referred and was applicable to “the situation of a pending criminal charge or, at most, of an investigation solely for criminal purposes.” 400 U. S., at 533. regarding the conflict about whether the recommendation for criminal prosecution is dispositive of the so-called criminal purpose issue. Compare United States v. Hodge & Zweig, 548 F. 2d, at 1351; and United States v. Billingsley, 469 F. 2d, at 1210, with United States v. Lajko, 520 F. 2d 622, 625 (CA3 1975), regarding the conflict about whether the criminal recommendation from the IRS to the Department of Justice or the recommendation from the special agent to his superiors is important in the enforcement inquiry. 7 In Boren v. Tucker, 239 F. 2d 767, 772-773 (1956), the Ninth Circuit distinguished United States v. O’Connor, 118 F. Supp. 248 (Mass. 1953), which involved an investigation of a taxpayer already under indictment. UNITED STATES v. LaSALLE NATIONAL BANK 307 298 Opinion of the Court Discerning the meaning of the brief Reisman dictum, however, did not resolve for the Court the question posed by Donaldson. The validity of the summonses depended ultimately on whether they were among those authorized by Congress.8 Having reviewed the statutory scheme, 400 U. S., at 523-525, the Court concluded that Congress had authorized the use of summonses in investigating potentially criminal conduct. The statutory history, particularly the use of summonses under the Internal Revenue Code of 1939,9 supported this conclusion, as did consistent IRS practice and decisions concerning effective enforcement of other comparable federal statutes.10 The Court saw no reason to force the Service to choose either to forgo the use of congressionally authorized summonses or to abandon the option of recommending criminal prosecutions to the Department of Justice.11 As long as the summonses were issued in good-faith pursuit of the congressionally authorized purposes, and prior to any recommendation to the Department for prosecution, they were enforceable. Id., at 536. Ill The present case requires us to examine the limits of the good-faith use of an Internal Revenue summons issued under § 7602. As the preceding discussion demonstrates, Donaldson does not control the facts now before us. There, the taxpayer had argued that the mere potentiality of criminal prosecution should have precluded enforcement of the summons. 400 U. S., at 532. Here, on the other hand, the District Court 8 The Court had concluded earlier that the summoning of the employer’s and the accountant’s records for an investigation of the taxpayer did not violate the constitutional rights of any of them. 400 U. S., at 522. 9 See §§ 3614, 3615, 3616, and 3654 of the 1939 Code, 53 Stat. 438-440, 446. 10 See United States v. Kordel, 397 U. S. 1, 11 (1970) (Federal Food, Drug, and Cosmetic Act enforcement), citing Standard Sanitary Mjg. Co. v. United States, 226 U. S. 20, 51-52 (1912) (Sherman Act enforcement). 11 See Part III-B and n. 15, infra. 308 OCTOBER TERM, 1977 Opinion of the Court 437U.S. found that Special Agent Olivero was investigating Gattuso “solely for the purpose of unearthing evidence of criminal conduct.” 76-1 USTC, at 84,073, 37 AFTR 2d, at 70-1240. The question then becomes whether this finding necessarily leads to the conclusion that the summonses were not issued in good-faith pursuit of the congressionally authorized purposes of § 7602. A The Secretary of the Treasury and the Commissioner of Internal Revenue are charged with the responsibility of administering and enforcing the Internal Revenue Code. 26 U. S. C. §§ 7801 and 7802. Congress, by § 7601 (a), has required the Secretary to canvass revenue districts to “inquire after and concerning all persons therein who may be liable to pay any internal revenue tax.” With regard to suspected fraud, these duties encompass enforcement of both civil and criminal statutes. The willful submission of a false or fraudulent tax return may subject a taxpayer not only to criminal penalties under §§ 7206 and 7207 of the Code, but, as well, to a civil penalty, under § 6653 (b), of 50% of the underpayment. And § 6659 (a) provides that the civil penalty shall be considered as part of the tax liability of the taxpayer. Hence, when § 7602 permits the use of a summons “[f]or the purpose of ascertaining the correctness of any return, . . . determining the liability of any person for any internal revenue tax . . . , or collecting any such liability,” it necessarily permits the use of the summons for examination of suspected tax fraud and for the calculation of the 50% civil penalty. In Donaldson, 400 U. S., at 535, we clearly noted that § 7602 drew no distinction between the civil and the criminal aspects; that it “contains no restriction”; that the corresponding regulations were “positive”; and that there was no significance, “for civil as compared with criminal purposes, at the point of a special agent’s appearance.” The Court then upheld the use of the summonses even though fraudulent conduct carried the poten- UNITED STATES v. LaSALLE NATIONAL BANK 309 298 Opinion of the Court tial of criminal liability. The Court repeated this emphasis in Couch v. United States, 409 U. S. 322, 326 (1973): “It is now undisputed that a special agent is authorized, pursuant to 26 U. S. C. § 7602, to issue an Internal Revenue summons in aid of a tax investigation with civil and possible criminal consequences.” This result is inevitable because Congress has created a law enforcement system in which criminal and civil elements are inherently intertwined. When an investigation examines the possibility of criminal misconduct, it also necessarily inquires about the appropriateness of assessing the 50% civil tax penalty.12 12 The interrelated nature of the civil and criminal investigative functions is further demonstrated by the organization and functioning of the IRS. Pursuant to 26 CFR §601.107 (1977), each revenue district has an Intelligence Division, “whose mission is to encourage and achieve the highest possible degree of voluntary compliance with the internal revenue laws.” This purpose is implemented by “the investigation of possible criminal violations of such laws and the recommendation (when warranted) of prosecution and/or assertion of the 50 percent ad valorem addition to the tax.” Ibid. See generally Internal Revenue Service Organization and Functions §§ 1113.563, 1114.8, and 1118.6, 39 Fed. Reg. 11572, 11581, 11601, and 11607 (1974). In its Manual for employees, the IRS instructs that the jurisdiction of the Intelligence Division includes all civil penalties except those related to the estimated income tax. Internal Revenue Manual, ch. 4500, § 4561 (CCH 1976). The Manual adds: “Intelligence features are those activities of developing and presenting admissible evidence required to prove criminal violations and the ad valorem penalties for civil fraud, negligence and delinquency (except those concerning tax estimations) for all years involved in cases jointly investigated to completion.” Id., § 4565.31 (4). The Manual also contains detailed instructions for coordination between special agents and revenue agents during investigations of tax fraud. E. g., id., §4563.431 (1978), and §§4565.22, 4565.32, 4565.41^565.44 (1976). Statistics for the fiscal year 1976 show that the Intelligence Division has a substantially greater involvement with civil fraud than with criminal 310 OCTOBER TERM, 1977 Opinion of the Court 437 U. S. The legislative history of the Code supports the conclusion that Congress intended to design a system with interrelated criminal and civil elements. Section 7602 derives, assertedly without change in meaning,13 from corresponding and similar provisions in §§ 3614, 3615, and 3654 of the 1939 Code. By § 3614 (a) the Commissioner received the summons authority “for the purpose of ascertaining the correctness of any return or for the purpose of making a return where none has been made.” Section 3615 (b)(3) authorized the issuance of a summons “[w]henever any person who is required to deliver a monthly or other return of objects subject to tax delivers any return which, in the opinion of the collector, is erroneous, false, or fraudulent, or contains any undervaluation or understatement.” Section 3654 (a) stated the powers and duties of the collector: “Every collector within his collection district shall see that all laws and regulations relating to the collection of internal revenue taxes are faithfully executed and complied with, and shall aid in the prevention, detection, and punishment of any frauds in relation thereto. For such purposes, he shall have power to examine all persons, books, papers, accounts, and premises . . . and to summon any person to produce books and papers . . . and to compel compliance with such summons in the same manner as provided in section 3615.” Under § 3616 punishment for any fraud included both fine and imprisonment. The 1939 Code, therefore, contemplated the use of the summons in an investigation involving suspected fraud. Of 8,797 full-scale tax fraud investigations in that year, only 2,037 resulted in recommendations for prosecution. The 6,760 cases not recommended involved approximately $11 million in deficiencies and penalties. See 1976 Annual Report of the Commissioner of Internal Revenue 33, 61, 152. 13 See H. R. .Rep. No. 1337, 83d Cong., 2d Sess., A436 (1954); S. Rep. No. 1622, 83d Cong., 2d Sess., 617 (1954). UNITED STATES v. LaSALLE NATIONAL BANK 311 298 Opinion of the Court criminal conduct as well as behavior that could have been disciplined with a civil penalty.14 In short, Congress has not categorized tax fraud investigations into civil and criminal components. Any limitation on the good-faith use of an Internal Revenue summons must reflect this statutory premise. B The preceding discussion suggests why the primary limitation on the use of a summons occurs upon the recommendation of criminal prosecution to the Department of Justice. Only at that point do the criminal and civil aspects of a tax fraud case begin to diverge. See United States v. Hodge & Zweig, 548 F. 2d 1347, 1351 (CA9 1977); United States v. Billingsley, 469 F. 2d 1208, 1210 (CAIO 1972). We recognize, of course, that even upon recommendation to the Justice Department, the civil and criminal elements do not separate completely. The Government does not sacrifice its interest in unpaid taxes 14 Internal Revenue officials received similar summons authority in Revenue Acts prior to the 1939 Code. See, e. g., Revenue Act of 1918, § 1305, 40 Stat. 1142; Tariff Act of Oct. 3, 1913, § II H, 38 Stat. 178-179; Act of June 30, 1864, § 14,13 Stat. 226. The interrelated nature of fraud investigations thus was apparent as early as 1864. Section 14 of the 1864 Act permitted the issuance of a summons to investigate a suspected fraudulent return. It also prescribed a 100% increase in valuation as a civil penalty for falsehood. Section 15 established the criminal penalties for such conduct. Four years later, when Congress created the position of district supervisor, that official received similar summons authority. Act of July 20, 1868, § 49, 15 Stat. 144^145; see Cong. Globe, 40th Cong., 2d Sess., 3450 (1868). The federal courts enforced these summonses when they were issued in good faith and in compliance with instructions from the Commissioner. See In re Meador, 16 F. Cas. 1294, 1296 (No. 9,375) (ND Ga. 1869); Stanwood v. Green, 22 F. Cas. 1077, 1079 (No. 13,301) (SD Miss. 1870) (“it being understood that this right upon the part of the supervisor extends only to such books and papers as relate to their banking operations, and are connected with the internal revenue of the United States”). 312 OCTOBER TERM, 1977 Opinion of the Court 437U.S. just because a criminal prosecution begins. Logically, then, the IRS could use its summons authority under § 7602 to uncover information about the tax liability created by a fraud regardless of the status of the criminal case. But the rule forbidding such is a prophylactic intended to safeguard the following policy interests. A referral to the Justice Department permits criminal litigation to proceed. The IRS cannot try its own prosecutions. Such authority is reserved to the Department of Justice and, more particularly, to the United States Attorneys. 28 U. S. C. § 547 (1). Nothing in § 7602 or its legislative history suggests that Congress intended the summons authority to broaden the Justice Department’s right of criminal litigation discovery or to infringe on the role of the grand jury as a principal tool of criminal accusation. Accord, United States v. Morgan Guaranty Trust Co., 572 F. 2d 36 (CA2 1978); United States v. Weingarden, 473 F. 2d 454, 458-459 (CA6 1973); United States v. O’Connor, 118 F. Supp. 248, 250-251 (Mass. 1953); see Donaldson v. United States, 400 U. S., at 536; cf. Abel v. United States, 362 U. S. 217, 226 (1960). The likelihood that discovery would be broadened or the role of the grand jury infringed is substantial if post-referral use of the summons authority were permitted. For example, the IRS, upon referral, loses its ability to compromise both the criminal and the civil aspects of a fraud case. 26 U. S. C. § 7122 (a). After the referral, the authority to settle rests with the Department of Justice. Interagency cooperation on the calculation of the civil liability is then to be expected and probably encourages efficient settlement of the dispute. But such cooperation, when combined with the inherently intertwined nature of the criminal and civil elements of the case, suggests that it is unrealistic to attempt to build a partial information barrier between the two branches of the executive. Effective use of information to determine civil liability would inevitably result in criminal discovery. UNITED STATES v. LaSALLE NATIONAL BANK 313 298 Opinion of the Court The prophylactic restraint on the use of the summons effectively safeguards the two policy interests while encouraging maximum interagency cooperation.15 C Prior to a recommendation for prosecution to the Department of Justice, the IRS must use its summons authority in good faith. Donaldson v. United States, 400 U. S., at 536; United States v. Powell, 379 U. S. 48, 57-58 (1964). In Powell, the Court announced several elements of a good-faith exercise: “[The Service] must show that the investigation will be conducted pursuant to a legitimate purpose, that the inquiry may be relevant to the purpose, that the information sought is not already within the Commissioner’s 15 The Third Circuit has suggested that our reference in Donaldson to the recommendation for criminal prosecution (“We hold that under § 7602 an internal revenue summons may be issued in aid of an investigation if it is issued in good faith and prior to a recommendation for criminal prosecution,” 400 U. S., at 536) intended to draw a line at the recommendation to the Service’s district office from the special agent, rather than at the recommendation from the Service to the Justice Department. United States v. Lajko, 520 F. 2d, at 625. This misread our intent. Given the interrelated criminal/civil nature of tax fraud investigation whenever it remains within the jurisdiction of the Service, and given the utility of the summons to investigate civil tax liability, we decline to impose the prophylactic restraint on the summons authority any earlier than at the recommendation to the Department of Justice. We cannot deny that the potential for expanding the criminal discovery rights of the Justice Department or for usurping the role of the grand jury exists at the point of the recommendation by the special agent. But we think the possibilities for abuse of these policies are remote before the recommendation to Justice takes place and do not justify imposing an absolute ban on the use of the summons before that point. Earlier imposition of the ban, given the balance of policies and civil law enforcement interests, would unnecessarily hamstring the performance of the tax determination and collection functions by the Service. 314 OCTOBER TERM, 1977 Opinion of the Court 437U.S. possession, and that the administrative steps required by the Code have been followed .... [A] court may not permit its process to be abused. Such an abuse would take place if the summons had been issued for an improper purpose, such as to harass the taxpayer or to put pressure on him to settle a collateral dispute, or for any other purpose reflecting on the good faith of the particular investigation.” Ibid, (footnote omitted). A number of the Courts of Appeals, including the Seventh Circuit in this case, 554 F. 2d, at 309, have said that another improper purpose, which the Service may not pursue in good faith with a summons, is to gather evidence solely for a criminal investigation.16 The courts have based their conclusions in part on Donaldson’s explanation of the Reisman dictum. The language of Donaldson, however, must be read in the light of the recognition of the interrelated criminal/civil nature of a tax fraud inquiry. For a fraud investigation to be solely criminal in nature would require an extraordinary departure from the normally inseparable goals of examining whether the basis exists for criminal charges and for the assessment of civil penalties. In this case, respondents submit that such a departure did indeed occur because Special Agent Olivero was interested only in gathering evidence for a criminal prosecution. We disagree. The institutional responsibility of the Service to calculate and to collect civil fraud penalties and fraudulently reported or unreported taxes is not necessarily overturned by a single agent who attempts to build a criminal case. The 16 See, e. g., United States v. Hodge c& Zweig, 548 F. 2d, at 1350, 1351; United States v. Zack, 521 F. 2d, at 1368; United States v. Lajko, 520 F. 2d, at 625; United States v. McCarthy, 514 F. 2d, at 374-375; United States v. Theodore, 479 F. 2d 749, 753 (CA4 1973); United States v. Weingarden, 473 F. 2d, at 459; United States v. Wall Corp., 154 U. S. App. D. C., at 311, 475 F. 2d, at 895. UNITED STATES v. LaSALLE NATIONAL BANK 315 298 Opinion of the Court review process over and above his conclusions is multilayered and thorough. Apart from the control of his immediate supervisor, the agent’s final recommendation is reviewed by the district chief of the Intelligence Division, 26 CFR §§ 601.107 (b) and (c) (1977); Internal Revenue Manual, ch. 9600, §§ 9621.1, 9622.1, 9623 (CCH 1977) ; see Donaldson v. United States, 400 U. S., at 534. The Office of Regional Counsel also reviews the case before it is forwarded to the National Office of the Service or to the Justice Department. 26 CFR § 601.107 (c) (1977) ; Internal Revenue Service Organization and Functions §1116(3), 39 Fed. Reg. 11602 (1974); Internal Revenue Manual, ch. 9600, §§ 9624, 9631.2, 9631.4 (CCH 1977). If the Regional Counsel and the Assistant Regional Commissioner for Intelligence disagree about the disposition of a case, another complete review occurs at the national level centered in the Criminal Tax Division of the Office of General Counsel. Internal Revenue Service Organization and Functions § 1113.-(11) 22, 39 Fed. Reg. 11599 (1974) ; Internal Revenue Manual, ch. 9600, § 9651 (1) (CCH 1977). Only after the officials of at least two layers of review have concurred in the conclusion of the special agent does the referral to the Department of Justice take place. At any of the various stages, the Service can abandon the criminal prosecution, can decide instead to assert a civil penalty, or can pursue both goals. While the special agent is an important actor in the process, his motivation is hardly dispositive. It should also be noted that the layers of review provide the taxpayer with substantial protection against the hasty or overzealous judgment of the special agent. The taxpayer may obtain a conference with the district Intelligence Division officials upon request or whenever the chief of the Division determines that a conference would be in the best interests of the Government. 26 CFR § 601.107 (b)(2) (1977) ; Internal Revenue Manual, ch. 9300, § 9356.1 (CCH 1977). If prosecution has been recommended, the chief notifies the taxpayer of 316 OCTOBER TERM, 1977 Opinion of the Court 437U.S. the referral to the Regional Counsel. 26 CFR § 601.107 (c) (1977); Internal Revenue Manual, ch. 9300, §9355 (CCH 1977). As in Donaldson, then, where we refused to draw the line between permissible civil and impermissible criminal purposes at the entrance of the special agent into the investigation, 400 U. S., at 536, we cannot draw it on the basis of the agent’s personal intent. To do so would unnecessarily frustrate the enforcement of the tax laws by restricting the use of the summons according to the motivation of a single agent without regard to the enforcement policy of the Service as an institution. Furthermore, the inquiry into the criminal enforcement objectives of the agent would delay summons enforcement proceedings while parties clash over, and judges grapple with, the thought processes of each investigator.17 See United States v. Morgan Guaranty Trust Co., 572 F. 2d 36 (CA2 1978). This obviously is undesirable and unrewarding. As a result, the question whether an investigation has solely criminal purposes must be answered only by an examination of the institutional posture of the IRS. Contrary to the assertion of respondents, this means that those opposing enforcement of a summons do bear the burden to disprove the actual existence of a valid civil tax determination or collection purpose by the Service. After all, the purpose of the good-faith inquiry is to determine whether the agency is honestly pursuing the goals of § 7602 by issuing the summons. Without doubt, this burden is a heavy one. Because criminal and civil fraud liabilities are coterminous, the Service rarely will be found to have acted in bad faith by pursuing the former. On the other hand, we cannot abandon this aspect of the good-faith inquiry altogether.18 We shall not countenance 17 We recognize, of course, that examination of agent motive may be necessary to evaluate the good-faith factors of Powell, for example, to consider whether a summons was issued to harass a taxpayer. 18 The dissent would abandon this aspect of the good-faith inquiry. It would permit the IRS to use the summons authority solely for criminal UNITED STATES v. LaSALLE NATIONAL BANK 317 298 Opinion of the Court delay in submitting a recommendation to the Justice Department when there is an institutional commitment to make the referral and the Service merely would like to gather additional evidence for the prosecution. Such a delay would be tantamount to the use of the summons authority after the recommendation and would permit the Government to expand its criminal discovery rights. Similarly, the good-faith standard will not permit the IRS to become an information-gathering agency for other departments, including the Department of Justice, regardless of the status of criminal cases.19 investigation. It reaches this conclusion because it says the Code contains no limitation to prevent sueh use. Its argument reveals a fundamental misunderstanding about the authority of the IRS. The Service does not enjoy inherent authority to summon production of the private papers of citizens. It may exercise only that authority granted by Congress. In § 7602 Congress has bestowed upon the Service the authority to summon production for four purposes only: for “ascertaining the correctness of any return, making a return where none has been made, determining the liability of any person for any internal revenue tax ... or collecting any such liability.” Congress therefore intended the summons authority to be used to aid the determination and collection of taxes. These purposes do not include the goal of filing criminal charges against citizens. Consequently, summons authority does not exist to aid criminal investigations solely. The error of the dissent is that it seeks a limit on the face of the statute when it should seek an affirmative grant of summons authority for purely criminal investigations. We have made that search and could uncover nothing in the Code or its legislative history to suggest that Congress intended to permit exclusively criminal use of summonses. As a result, the IRS employs its authority in good faith when it pursues the four purposes of § 7602, which do not include aiding criminal investigations solely. 19 To the limited extent that the institutional good faith of the Service with regard to criminal purpose may be questioned before any renommen-dation to the Department of Justice, our position on this issue necessarily rejects the Government’s argument that prerecommendation enforcement of summonses must meet only the Powell elements of good faith. We have concluded that the Government’s contention fails to recognize the essence of the good-faith inquiry. The Powell elements were not intended as an exclusive statement about the meaning of good faith. They were examples 318 OCTOBER TERM, 1977 Opinion of the Court 437U.S. D In summary, then, several requirements emerge for the enforcement of an IRS summons.20 First, the summons must be issued before the Service recommends to the Department of Justice that a criminal prosecution, which reasonably would relate to the subject matter of the summons, be undertaken. Second, the Service at all times must use the summons authority in good-faith pursuit of the congressionally authorized purposes of § 7602. This second prerequisite requires the Service to meet the Powell standards of good faith. It also requires that the Service not abandon in an institutional sense, as explained in Parts III-A and III-C above, the pursuit of civil tax determination or collection. IV * On the record before us, respondents have not demonstrated sufficient justification to preclude enforcement of the IRS summonses. No recommendation to the Justice Department for criminal prosecution has been made. Of the Powell criteria, respondents challenge only one aspect of the Service’s showing: They suggest that Olivero already may possess the evidence requested in the summonses. Brief for Respondents 16-19. Although the record shows that Olivero had uncovered the names and identities of the LaSalle National Bank land trusts, it does not show that the Service knows the value of the trusts or their income or the allocation of interests therein. Because production of the bank’s complete records on the trusts reasonably could be expected to reveal part or all of this information, which would be material to the computation of agency action not in good-faith pursuit of the congressionally authorized purposes of § 7602. The dispositive question in each case, then, is whether the Service is pursuing the authorized purposes in good faith. 20 These requirements are not intended to be exclusive. Future cases may well reveal the need to prevent other forms of agency abuse of congressional authority and judicial process. UNITED STATES v. LaSALLE NATIONAL BANK 319 298 Stewart, J., dissenting of Gattuso’s tax liability, the Powell criteria do not preclude enforcement. Finally, the District Court refused enforcement because it found that Olivero’s personal motivation was to gather evidence solely for a criminal prosecution. The court, however, failed to consider whether the Service in an institutional sense had abandoned its pursuit of Gattuso’s civil tax liability.21 The Court of Appeals did not require that inquiry. On the record presently developed, we cannot conclude that such an abandonment has occurred. The judgment of the Court of Appeals is therefore reversed with instructions to that court to remand the case to the District Court for further proceedings consistent with this opinion. It is so ordered. Mr. Justice Stewart, with whom The Chief Justice, Mr. Justice Rehnquist, and Mr. Justice Stevens join, dissenting. This case is here only because of judicial misreadings of a passage in the Court’s opinion in Donaldson v. United States, 400 U. S. 517, 533. That passage has been read by the federal courts, in this case and in others, to mean that a sum 21 Respondents argue that the District Court made a factual finding when it concluded that the summonses were issued solely to gather evidence for a criminal prosecution. They then submit that the District Court’s decision may be overturned only if this Court holds this finding to be clearly erroneous. Several Courts of Appeals have discussed the factual and legal issues that lurk in summons enforcement proceedings. Compare United States v. Zack, 521 F. 2d, at 1367-1368; United States v. National State Bank, 454 F. 2d 1249, 1252 (CA7 1972); Boren v. Tucker, 239 F. 2d, at 773, with United States v. Weingarden, 473 F. 2d, at 460. Whether the issue of the Service’s good faith generally poses a factual question, or a legal and factual one, or a legal question, is not necessarily presented in the case now before the Court, and we do not reach it. The lower courts employed an incorrect legal standard to measure g6od faith when they limited their consideration to the personal motivation of Special Agent Olivero. In this case, then, a legal error compels reversal. 320 OCTOBER TERM, 1977 Stewart, J., dissenting 437 U. S. mons under § 7602 of the Internal Revenue Code, 26 U. S. C. § 7602, is improper if issued in aid of an investigation solely for criminal purposes.1 Yet the statute itself contains no such limitation, and the Donaldson opinion in fact clearly stated that there are but two limits upon enforcement of such a summons: It must be “issued in good faith and prior to a recommendation for criminal prosecution.” 400 U. S., at 536. I adhere to that view. The Court concedes that the task of establishing the “purpose” of an individual agent is “undesirable and unrewarding.” Ante, at 316. Yet the burden it imposes today—to discover the “institutional good faith” of the entire Internal Revenue Service—is, in my view, even less desirable and less rewarding. The elusiveness of “institutional good faith” as described by the Court can produce little but endless discovery proceedings and ultimate frustration of the fair administration of the Internal Revenue Code. In short, I fear that the Court’s new criteria will prove wholly unworkable. Earlier this year the Court of Appeals for the Second Circuit had occasion to deal with the issue now before us in the case of United States n. Morgan Guaranty Trust Co., 572 F. 2d 36. Judge Friendly’s perceptive opinion for his court in that case read the Donaldson opinion correctly: This Court was there “laying down an objective test, ‘prior to a recommendation for criminal prosecution,’ that would avoid a need for determining the thought processes of special agents; and . . . the ‘good faith’ requirement of the holding related to such wholly different matters as those mentioned in” the case of United States v. Powell, 379 U. S. 48.2 “Such a view would ... be 1 See ante, at 305-306, n. 6. 2 As Judge Friendly pointed out, this Court’s Powell opinion simply declared that a court may not permit its process in enforcing a summons to be abused, and its examples of “abuse” were: “ 'Such an abuse would take place if the summons had been issued for an improper purpose, such as to harass the taxpayer or to put pressure on him UNITED STATES v. LaSALLE NATIONAL BANK 321 298 Stewart, J., dissenting consistent with the only rationale that has ever been offered for preventing an otherwise legitimate use of an Internal Revenue Service third party summons, namely that Congress could not have intended the statute to trench on the power of the grand jury or to broaden the Government’s right to discovery in a criminal case . ...” 572 F. 2d, at 41—42. Instead of standing by the objective and comparatively bright-line test of Donaldson, as now clarified, the Court today further muddies the waters. It does not even attempt to identify the source of the requirements it now adds to enforcement proceedings under §§ 7402 (b) and 7604 (a) of the Code. These requirements are not suggested by anything in the statutes themselves, and nobody suggests that they derive from the Constitution. They are simply imposed by the Court from out of nowhere, and they seem to me unjustified, unworkable, and unwise. I would reverse the judgment, not for further hearings in the District Court, but with instructions to order enforcement of the summons. to settle a collateral dispute, or for any other purpose reflecting on the good faith of the particular investigation.’ [379 U. S., at 58.] “Nothing was said to indicate that an intention by the Commissioner to uncover criminal tax liability would reflect 'on the good faith’ of the inquiry, and the rule of ejusdem generis would dictate the contrary.” 572 F. 2d, at 40. 322 OCTOBER TERM, 1977 437 U. S. Syllabus GREYHOUND CORP, et al. v. MT. HOOD STAGES, INC., dba PACIFIC TRAILWAYS CERTIORARI TO THE UNITED STATES COURT OF APPEALS FOR THE NINTH CIRCUIT No. 77-598. Argued April 24, 1978—Decided June 19, 1978 On October 7, 1964, respondent motor carrier instituted a proceeding before the Interstate Commerce Commission in which it asked the ICC to reopen proceedings in which the ICC, over respondent’s opposition, had approved petitioner’s acquisition of several bus companies, alleging that petitioner had not lived up to representations that the acquisitions would not adversely affect respondent. On December 14, 1964, the United States petitioned for leave (and later was allowed) to intervene in the ICC proceeding, stating that respondent’s allegations made “a serious charge” but that it did not know whether they were “true or false.” After extensive hearings, the ICC decided against petitioner. In the meantime on July 5, 1968, respondent filed an action in District Court alleging, inter alia, violations of the federal antitrust laws, and the jury found violations of the Sherman Act and fraudulent concealment of such violations. The court held that the Government’s petition to intervene in the ICC proceeding served to toll the statute of limitations under §5 (i) of the Clayton Act (which provides that “[w]henever any civil or criminal proceeding is instituted by the United States to prevent, restrain, or punish violations of any of the antitrust laws, . . . the running of the statute of limitations in respect of every private . . . right of action arising under said laws and based in whole or in part on any matter complained of in said proceeding shall be suspended during the pendency thereof and for one year thereafter”), with the result that the Act’s four-year period of limitations extended back to December 14, 1960, when it was combined with fraudulent concealment to create a 20-year damages period. The Court of Appeals affirmed, holding that the literal wording of § 5 (i) was not controlling and that §5(i)’s purpose in furthering effective enforcement of the antitrust laws by permitting private litigants to benefit from governmental antitrust enforcement efforts would be advanced by treating the United States’ petition to intervene as the “functional equivalent of a direct action” by the United States. Held: The Clayton Act’s statute of limitations was GREYHOUND CORP. v. MT. HOOD STAGES, INC. 323 322 Syllabus not tolled under § 5 (i) by the filing of the Government’s petition to intervene in the ICC proceeding. Pp. 330-337. (a) The ICC proceeding was plainly not “instituted by the United States” within the meaning of § 5 (i). It strains accepted usage to argue that a party who intervenes in a proceeding instituted by someone else has also “instituted that proceeding.” In fact, the United States not only did not institute the ICC proceeding but was not in a position to do so, since in view of its statement that it did not know whether respondent’s allegations were “true or false,” it could not in good faith have made the charging allegations necessary to institute the proceeding. Pp. 330-331. (b) Neither had the United States, within the meaning of § 5 (i), “complained of” anything on which the District Court action was based, since in the ICC proceeding its petition to intervene charged petitioner with no wrongdoing, took no position on the merits, sought no relief, and disclaimed any knowledge of the relevant facts, seeking only an opportunity for respondent to establish its allegations. Pp. 331-332. (c) What is now § 5 (i) was enacted to ensure that private litigants would have the benefit of prior Government antitrust efforts, and this purpose would not be served by construing § 5 (i) as applicable to the facts of this case, where respondent is seeking to benefit not from a Government antitrust action, but from an ICC proceeding respondent itself instituted. Pp. 332-334. (d) Application of § 5 (i) to this case would also fail to give weight to Congress’ purpose in amending the Clayton Act to provide a uniform four-year period of limitations and thus eliminate the prior confusion caused by determining the period of limitations by state law. P. 334. 555 F. 2d 687, vacated and remanded. Blackmun, J., delivered the opinion for a unanimous Court. Burger, C. J., filed a concurring opinion, post, p. 337. John R. Reese argued the cause for petitioners. With him on the briefs were Richard C. Brautigam, James H. Clarke, and Keith A. Jenkins. Eugene C. Crew argued the cause for respondent. With him on the brief were Michael N. Khourie, Bruce M. Hall, and Donald A. Schafer. 324 OCTOBER TERM, 1977 Opinion of the Court 437 U. S. Mr. Justice Blackmun delivered the opinion of the Court. This case presents the issue whether § 5 (i) of the Clayton Act, as amended, 88 Stat. 1706, 90 Stat. 1396, 15 U. S. C. § 16 (i) (1976 ed.),1 operates to toll the running of the Act’s statute of limitations2 from the date on which the United States filed a petition for leave to intervene in an Interstate Commerce Commission proceeding previously instituted by the plaintiff. I Petitioner Greyhound3 and respondent Mt. Hood Stages, Inc. (doing business as Pacific Trailways), are motor common 1 Section 5 (i), as set forth in 15 U. S. C. § 16 (i) (1976 ed.), provides: "Whenever any civil or criminal proceeding is instituted by the United States to prevent, restrain, or punish violations of any of the antitrust laws, but not including an action under section 15a of this title, the running of the statute of limitations in respect to every private or State right of action arising under said laws and based in whole or in part on any matter complained of in said proceeding shall be suspended during the pendency thereof and for one year thereafter: Provided, however, That whenever the running of the statute of limitations in respect of a cause of action arising under section 15 or 15c of this title is suspended hereunder, any action to enforce such cause of action shall be forever barred unless commenced either within the period of suspension or within four years after the cause of action accrued.” 2 Section 4B, 69 Stat. 283, as amended, 15 U. S. C. § 15b (1976 ed.). It provides: “Any action to enforce any cause of action under sections 15, 15a, or 15c of this title shall be forever barred unless commenced within four years after the cause of action accrued. No cause of action barred under existing law on the effective date of this Act shall be revived by this Act.” 3 Petitioner The Greyhound Corporation is a Delaware corporation that now is a diversified holding company owning, among other assets, all the issued and outstanding capital stock of petitioner Greyhound Lines, Inc., a California corporation. On December 31, 1963, The Greyhound Corporation discontinued its operation of scheduled common carrier bus service and transferred its motor carrier operating rights and properties to Greyhound Lines, Inc., App. 68; cf. Mt. Hood Stages, Inc., 104 M. C. C. 449, 465 (1968). For convenience, we refer to the two corporations collectively as GREYHOUND CORP. v. MT. HOOD STAGES, INC. 325 322 Opinion of the Court carriers of passengers and package express and are subject to regulation by the Interstate Commerce Commission (ICC). Greyhound is the largest common carrier by bus in the United States. Mt. Hood is one of Greyhound’s comparatively small competitors; it operates over routes in Oregon, Idaho, and Utah. Its principal routes are between Portland, Eugene, and Albany, Ore., in the west, and Salt Lake City, in the east, and between Klamath Falls, Ore., in the south, and Biggs and The Dalles, Ore., in the north. Greyhound’s route authority surrounds that of Mt. Hood. During the period from 1947 to 1956, Greyhound acquired control of eight bus companies operating in the Western United States. See Mt. Hood Stages, Inc., 104 M. C. C. 449, 450, and n. 1 (1968). In the proceedings before the ICC, Mt. Hood opposed four of those acquisitions,4 alleging that, if the acquisitions were approved, Greyhound could route traffic around Mt. Hood’s operations and thereby deprive the public of the most convenient service and jeopardize Mt. Hood’s continued existence.5 Greyhound successfully contended, however, that the acquisitions were not intended to, and would not, have such consequences. Greyhound represented to the ICC that the acquisitions “would not adversely affect connecting carriers; that arrangements with such carriers, including interchange of traffic and open gateways, would be maintained; that it was not the policy of Greyhound to route passengers over circuitous routes; that its agents were instructed to quote the direct route as well as the Greyhound route and give passengers their choice; and that Greyhound had always “Greyhound.” The formal transfer of rights and properties at the end of 1963 has no significance for purposes of this litigation. 4 Mt. Hood, however, withdrew its opposition to one of these. See id., at 452. 5 See 555 F. 2d 687, 689 (CA9 1977). 326 OCTOBER TERM, 1977 Opinion of the Court 437U.S. carried MH’s schedules in its folders and cooperated in every way to acquaint the public with its service and thus promote additional traffic and business for their Unes.” 6 Greyhound also represented to the Commission that it would continue the joint through-bus arrangement with Mt. Hood.7 As Greyhound had anticipated, the ICC relied on these representations in determining that the proposed acquisitions were in the public interest. Id., at 454r-457,461. In July 1964, Greyhound terminated the through-bus arrangement with Mt. Hood. On October 7 of that year, Mt. Hood filed a petition with the Commission, pursuant to § 5 (10) (formerly § 5 (9)) of the Interstate Commerce Act,8 alleging that Greyhound had not lived up to various representations it had made to the ICC and asking the Commission to reopen the acquisition proceedings “for further hearing to consider the necessity of attaching certain terms, conditions and limitations to the privileges therein granted” or, in the alternative, to order Greyhound to divest itself of operations acquired in those proceedings. App. 4. The allegations in Mt. Hood’s petition to the Commission were essentially the same as those Mt. Hood made later in this antitrust suit, that 6 This quoted material is from the opinion in the subsequent ICC proceeding instituted by Mt. Hood to reopen the eight acquisition proceedings. Mt. Hood Stages, Inc., 104 M. C. C., at 452. Greyhound’s representations in those eight proceedings were so summarized. 7 This arrangement, initiated in 1949, provided for a through bus from San Francisco to Spokane, using Mt. Hood’s bridge route between Klamath Falls and Biggs. The route was shorter by 110 miles and several hours than the all-Greyhound route via Portland. It provided better service to travelers and was profitable for both companies. 555 F. 2d, at 689 n. 3. 8 Section 5 (10) of the Interstate Commerce Act, as amended, 90 Stat. 63, 66, 49 U. S. C. § 5 (10) (1976 ed.), provides: “The Commission may from time to time, for good cause shown, make such orders, supplemental to any order made under paragraph (1), (2), or (8), of this section, as it may deem necessary or appropriate.” GREYHOUND CORP. v. MT. HOOD STAGES, INC. 327 322 Opinion of the Court is, that Greyhound had canceled the through-bus connection, had scheduled connecting service so as to preclude reasonable connections with Mt. Hood, had directed Greyhound’s agents and independent joint ticket agents to send traffic around Mt. Hood’s routes through use of longer all-Greyhound routes, and had interfered in various ways with the distribution of Mt. Hood’s schedules and the quotation of Mt. Hood’s rates and services, all with the intent of injuring Mt. Hood. Id., at 10-11. Slightly more than two months later, on December 14, 1964, the United States petitioned for leave to intervene in the ICC proceeding. Id., at 36. In its petition, the United States stated it had an interest in the proceeding and it urged that the Commission hold a hearing on Mt. Hood’s allegations. The Government’s petition observed that Mt. Hood’s allegations “make a serious charge,” id., at 37, but added: “We have no way of knowing whether those of Mt. Hood’s allegations which Greyhound denies are true or false; resolution of such controversies is a typical function of a hearing.”® Id., at 37-38. On May 27, 1965, the United States and others were granted permission to intervene in the ICC proceeding. Id., at 43. Such permission, however, was on condition that it “shall not be construed to allow intervenors to introduce evidence which will unduly broaden the issues raised in this proceeding.” Ibid. After an extensive evidentiary hearing, the examiner resolved all factual issues against Greyhound and recommended entry of an order requiring Greyhound to abide by the representations it had made in the acquisition proceedings. Mt. Hood 9 Reiterating this point, the United States’ petition, stated: “Mt. Hood’s grave allegations, whether true or false, as well as Greyhound’s answer raise issues too serious and important to be disposed of summarily without a full adversary hearing in which allegation and denial can be put to the test of proof and cross-examination.” App. 38. 328 OCTOBER TERM, 1977 Opinion of the Court 437U.S. Stages, Inc., 104 M. C. C., at 464-496. On April 5, 1968, Division 3 of the ICC sustained the examiner’s findings but deferred entry of a supplemental order to allow voluntary negotiations between the parties. Id., at 462-463. On July 5, 1968, Mt. Hood filed this action in the United States District Court for the District of Oregon for damages and injunctive relief, alleging violations of the antitrust laws and common-law and statutory unfair competition. App. 46. Mt. Hood’s complaint alleged, as to the antitrust violations, that, beginning before 1947 and continuing to the date of the complaint, Greyhound had restrained and monopolized commerce in the carriage by motorcoach of passengers and their luggage between points in the Western United States, including Oregon, Idaho, and Utah, by means essentially the same as those that were the subject of the ICC proceeding. Id., at 49-52. In the Commission proceeding, meanwhile, the efforts of the parties to agree upon an order failed. The entire Commission therefore entered an order requiring Greyhound to restore the practices and traffic patterns existing when the acquisitions at issue were authorized and, specifically, to eliminate the anticompetitive practices of which Mt. Hood had complained. See Greyhound Lines, Inc. v. United States, 308 F. Supp. 1033, 1037 (ND Ill. 1970). A three-judge United States District Court denied Greyhound’s motion to set aside the Commission’s order and granted the counterclaim of the United States and the ICC by the issuance of its own order in similar terms, thus granting injunctive relief. Id., at 1040-1041. Following entry of the District Court’s order enforcing the ICC decision, Mt. Hood amended its complaint in this antitrust suit to eliminate its prayer for injunctive relief.10 App. 57. 10 Greyhound thereafter disobeyed the three-judge District Court’s order and was adjudged in criminal contempt. Certain of its officers were adjudged in civil contempt. Fines aggregating $600,000 were imposed. United States n. Greyhound Corp., 363 F. Supp. 525 (ND Ill. 1973), and 370 F. Supp. 881 (ND Ill. 1974), aff’d, 508 F. 2d 529 (CA7 1974). GREYHOUND CORP. v. MT. HOOD STAGES, INC. 329 322 Opinion of the Court In the present action, interrogatories were submitted to the jury. By its special verdict returned in May 1973, the jury found that, as alleged by Mt. Hood, Greyhound had violated both §§ 1 and 2 of the Sherman Act; that Greyhound had fraudulently concealed these antitrust violations during the period from January 1, 1953, to July 4, 1964; but that Mt. Hood knew or should have known of the violation on December 14, 1960. App. 82. The trial court held that the Government’s petition to intervene in the ICC modification proceeding on December 14, 1964, served to toll the statute of limitations under § 5 (i) of the Clayton Act. App. 80. The result was that the Act’s four-year period of limitations extended back to December 14, 1960, where it was combined with the fraudulent concealment to create a 20-year damages period.11 Damages of $13,146,090 (after trebling) were awarded Mt. Hood, plus attorneys’ fees of $1,250,000 and costs. Id., at 83, 104, 106. On appeal, the United States Court of Appeals for the Ninth Circuit affirmed. 555 F. 2d 687 (1977). We granted certiorari limited to the issue of the correctness of the interpretation of § 5 (i) by the District Court and the Court of Appeals.12 434 U.S. 1008(1978). 11 The four-year period of limitations, as already noted, n. 2, supra, is contained in § 4B of the Clayton Act, 15 U. S. C. § 15b (1976 ed.). Tolling of the statute was essential to the award of all damages beyond the normal four-year period, that is, back beyond July 5, 1964, the date four years prior to the date of filing of the antitrust complaint. The sum of $5,194,617, after trebling, is involved in the tolling issue. 12 Other issues advanced by Greyhound in its petition for certiorari, review of which was not granted, were (a) whether § 5 (12) of the Interstate Commerce Act, 49 U. S. C. §5 (12), and applicable antitrust principles permitted the treble-damages award by the jury’s application of antitrust standards to acquisitions approved by the ICC and to the manner of operation of the acquired companies which is subject to the Commission’s “exclusive and plenary” regulatory authority; (b) whether § 5 (a) of the Clayton Act, as amended, 15 U. S. C. § 16 (a) (1976 ed.), permitted the jury to base a finding of violation of the Sherman Act on consent decrees that expressly denied any antitrust violation and were 330 OCTOBER TERM, 1977 Opinion of the Court 437 U. 8. II In holding that the United States’ intervention in the ICC proceeding served to toll, by reason of § 5 (i), the Clayton Act’s period of limitations, the Court of Appeals stated that “[t]he literal wording of section [5 (i)] is not controlling.” 555 F. 2d, at 699. The court, therefore, sought to identify the congressional purpose behind § 5 (i) and to effectuate that purpose. 555 F. 2d, at 699. In the court’s view, the purpose of § 5 (i) “is to further effective enforcement of the antitrust laws by permitting private litigants to have the benefits that may flow from governmental antitrust enforcement efforts.” 555 F. 2d, at 699. The Court of Appeals, quoting the District Court (App. 80), declared that this purpose would be advanced by “ ‘treating intervention by Antitrust Division lawyers as the functional equivalent of a direct action by them.’ ” 555 F. 2d, at 700. We find this reasoning unpersuasive. In particular, we are unable to agree that the language of § 5 (i) is so unhelpful. Neither do we agree that the congressional purpose behind § 5 (i) is advanced by the holdings of the District Court and the Court of Appeals. A Logic and precedent dictate that “ ‘[t]he starting point in every case involving construction of a statute is the language itself.’ ” Santa Fe Industries, Inc. v. Green, 430 U. S. 462, 472 (1977), and Ernst & Ernst v. Hochjelder, 425 U. S. 185, 197 (1976), each quoting Blue Chip Stamps v. Manor Drug Stores, 421 U. S. 723, 756 (1975) (Powell, J., concurring). Examination of the language of § 5 (i) prevents acceptance of respondent’s position. Section 5 (i) begins: “Whenever any civil or criminal proceeding is instituted by the United States . . . .” (Emphasis entered before any testimony was taken; and (c) whether § 4B of the Clayton Act was tolled by fraudulent concealment. GREYHOUND CORP. v. MT. HOOD STAGES, INC. 331 322 Opinion of the Court added.) The ICC proceeding at issue here plainly was not one instituted by the United States. As the foregoing statement of facts demonstrates, and as the Court of Appeals acknowledged, “Mt. Hood rather than the United States instituted the proceedings.” 555 F. 2d, at 699. It strains accepted usage to argue that a party who intervenes in a proceeding instituted by someone else has also “instituted” that proceeding. This Court has observed: “When the term [to intervene] is used in reference to legal proceedings, it covers the right of one to interpose in, or become a party to, a proceeding already instituted....” Rocca v. Thompson, 223 U. S. 317, 330 (1912) (emphasis added). In truth, the United States not only did not institute the proceeding, but also was not in a position to do so. As its petition to intervene stated, the Government had “no way of knowing” whether Mt. Hood’s allegations, which Greyhound denied, were “true or false,” and thus it could not in good faith have made the charging allegations necessary to institute the proceeding. At least in this case, therefore, the question is not primarily one of form, that is, who reached the ICC first; it is one of substance, that is, who investigated the facts enabling it to make charging allegations and seek relief and thereby to “institute” the proceeding. Just as the United States cannot be said to have “instituted” the ICC proceeding, neither had it “complained of,” within the meaning of § 5 (i), anything on which the present action is based. The cases in which the applicability of § 5 (i) has been considered establish that the determination of whether a private action is based on matters “complained of” in a prior Government action “[i]n general . . . must be limited to a comparison of the two complaints on their face.” Leh v. General Petroleum Corp., 382 U. S. 54, 65 (1965); accord, Luria Steel <& Trading Corp. v. Ogden Corp., 484 F. 2d 1016, 1022 (CA3 1973), cert, denied, 414 U. S. 1158 (1974); Rader v. 332 OCTOBER TERM, 1977 Opinion of the Court 437 U. S. Balfour, 440 F. 2d 469, 473 (CA7), cert, denied sub nom. Alpha Chi Omega v. Rader, 404 U. S. 983 (1971). In the ICC proceeding here in question, the United States’ petition for leave to intervene charged Greyhound with no wrongdoing, took no position on the merits, sought no relief, and, indeed, disclaimed any knowledge of the relevant facts. It sought only an opportunity for Mt. Hood to establish its allegations. This case, therefore, simply cannot be viewed as one based on any matter “complained of” by the United States.13 B Moreover, the language of § 5 (i) that we rely upon accurately manifests Congress’ intent in enacting the section. As the Court previously has noted, the original § 5 of the Clayton 13 The Government’s petition to intervene is clearly distinguishable from the Federal Trade Commission’s complaint that this Court, in Minnesota Mining & Mjg. Co. n. New Jersey Wood Finishing Co., 381 U. S.311 (1965), held to have tolled the Clayton Act’s period of limitations under the predecessor'of §5 (i), 15 U. S. C. § 16 (b) (1964 ed.). There the FTC had filed a proceeding against the subsequent antitrust defendant under § 7 of the Clayton Act, 15 U. S. C. § 18 (1964 ed.). It was clear that the Government had actually charged the defendant with violations of the antitrust laws. The subsequent private antitrust action was directly based on the Government’s allegations (which had resulted in a consent order). 381 U. 8., at 313, 322-323. The petition to intervene in question here is also distinguishable from cases (the correctness of which we do not address) holding the Clayton Act’s period of limitations to have been tolled by § 5 of the Federal Trade Commission Act, 15 U. 8. C. §45 (1976 ed.). See, e. g., Luria Steel & Trading Corp. v. Ogden Corp., 484 F. 2d 1016 (CA3 1973), cert denied, 414 U. S. 1158 (1974); Rader v. Balfour, 440 F. 2d 469 (CA7), cert, denied sub nom. Alpha Chi Omega v. Rader, 404 U. S. 983 (1971); Lippa’s, Inc. v. Lenox, Inc., 305 F. Supp. 182 (Vt. 1969). In each of these cases the Government actually had charged the defendant with, and sought the prevention or punishment of, specific anticompetitive conduct or antitrust violations, and a comparison of the Government’s charges with the private litigant’s complaint showed that the private action was based on the matter complained of by the Government. GREYHOUND CORP. v. MT. HOOD STAGES, INC. 333 322 Opinion of the Court Act, 38 Stat. 731, was adopted in response to the request of President Wilson and consisted of material that now constitutes §§ 5 (a)14 and 5 (i).15 In a speech to Congress on January 20,1914, the President urged that a statute be enacted that would permit victims of antitrust violations to have “redress upon the facts and judgments proved and entered in suits by the Government” and that “the statute of limitations shall be suffered to run against such litigants only from the date of the conclusion of the Government’s action. It is not fair that the private litigant should be obliged to set up and establish again the facts which the Government has proved.”16 51 Cong. Rec. 1964 (1914). This very language of the President was quoted in part in Minnesota Mining & Mfg. Co. v. New Jersey Wood Finishing Co., 381 U. S. 311, 318 (1965). Congress acceded to the President’s request. What is now 14 15 U. S. C. § 16 (a) (1976 ed.). This section provides: “A final judgment or decree heretofore or hereafter rendered in any civil or c[r]iminal proceeding brought by or on behalf of the United States under the antitrust laws to the effect that a defendant has violated said laws shall be prima facie evidence against such defendant in any action or proceeding brought by any other party against such defendant under said laws or by the United States under section 15a of this title, as to all matters respecting which said judgment or decree would be an estoppel as between the parties thereto: Provided, That this section shall not apply to consent judgments or decrees entered before any testimony has been taken or to judgments or decrees entered in actions under section 15a of this title.” 15 The original version of what is now § 5 (i) provided: “Whenever any suit or proceeding in equity or criminal prosecution is instituted by the United States to prevent, restrain or punish violations of any of the antitrust laws, the running of the statute of limitations in respect of each and every private right of action arising under said laws and based in whole or in part on any matter complained of in said suit or proceeding shall be suspended during the pendency thereof.” 38 Stat. 731. 16 President Wilson’s message was quoted frequently during the course of the congressional debates to explain the purpose of the amendments. See, e. g., 51 Cong. Rec. 9090, 9488 (1914). 334 OCTOBER TERM, 1977 Opinion of the Court 437 U. S. § 5 (i) was enacted to ensure that private litigants would have the benefit of prior Government antitrust enforcement efforts. 381 U. 8., at 317. Here, however, as already has been pointed out, Mt. Hood is seeking to benefit not from a Government antitrust action but from an ICC proceeding that Mt. Hood itself initiated. Accordingly, construing § 5 (i) as applicable to the facts of this case would not serve Congress’ most obvious purpose. It would also fail to give any weight to another related and important congressional purpose. A Ninth Circuit panel very recently emphasized: “Although the plaintiff is correct in asserting that [§ 5 (i) ] serves the broad and beneficent purpose of aiding private antitrust litigants ... it is also true that it is a statute of repose.” Dungan v. Morgan Drive-Away, Inc., 570 F. 2d 867, 869 (1978). This is clear upon examination of the 1955 amendments to the Clayton Act. 69 Stat. 282. Before these amendments, the period of limitations under the Clayton Act was determined by state law. This bred confusion in the computation of the period within which a private suit was required to be brought, especially when the Act’s tolling provision (what is now § 5 (i)) came into play. In order to eliminate this confusion, the amendments established a uniform period of limitations of four years17 and declared that the suspension of the statute would extend “during the pendency” of the federal proceeding and “for one year thereafter.” Finally, the amendments mandated, in what is now the proviso to § 5 (i), that, in the event the statute of limitations is tolled, any private right of action based on the matter complained of in the action by the Government “shall be forever barred unless commenced . . . within four years after the cause of action accrued.”18 17 69 Stat. 283, now § 4B of the Clayton Act, as amended, 15 U. S. C. § 15b (1976 ed.). 18 69 Stat. 283. GREYHOUND CORP. v. MT. HOOD STAGES, INC. 335 322 Opinion of the Court The Senate Report accompanying the 1955 amendments reflects congressional policy against “undue prolongation of [antitrust] proceedings” by extending the limitations period. It noted: “While the committee believes it important to safeguard the rights of plaintiffs by tolling the statute during the pendency of Government antitrust actions, it recognizes that in many instances the long duration of such proceedings taken in conjunction with a lengthy statute of limitations may tend to prolong stale claims, unduly impair efficient business operations, and overburden the calendars of courts. The committee believes the provision of this bill will tend to shorten the period over which private treble-damage actions will extend by requiring that the plaintiff bring his suit within 4 years after it accrued or within 1 year after the Government’s case has been concluded. “While the committee considers it highly desirable to toll the statute of limitations during a Government antitrust action and to grant plaintiff a reasonable time thereafter in which to bring suit, it does not believe that the undue prolongation of proceedings is conducive to effective and efficient enforcement of the antitrust laws.” S. Rep. No. 619, 84th Cong., 1st Sess., 6 (1955).19 In view of the congressional emphasis on certainty and predictability in the application of § 5 (i), the Court of Appeals’ conclusion that the United States’ petition to intervene should be treated as the “functional equivalent of a direct action” by the United States, 555 F. 2d, at 700, is unacceptable. A functional-equivalence standard, applied this loosely, resurrects the very confusion and uncertainty concerning the application of the statute of limitations that Congress sought to eliminate in the 1955 amendments. In a case such as this, 19 See also H. R. Rep. No. 422, 84th Cong., 1st Sess., 8-9 (1955). 336 OCTOBER TERM, 1977 Opinion of the Court 437U.S. in which the Government took no position in its initial petition, a functional-equivalence test would require a detailed review of the record in each proceeding to see what position the Government ultimately took and whether its participation was or was not the “functional equivalent of a direct action.” The Government, of course, may well change its position. For example, in Denver & R. G. W. R. Co. v. United States, 387 U. S. 485 (1967), the Government intervened in an ICC proceeding with one position, adopted another in the District Court, and then “completely reversed” itself in this Court. Id., at 490-492. Thus, endorsement of the suggested functional-equivalence test would mean that it might be impossible to determine whether Government proceedings would toll the statute until those proceedings were finally resolved. As the Court of Appeals seems to have acknowledged, such an approach would lead to serious problems. 555 F. 2d, at 699 n. 31. See also Dungan v. Morgan Drive-Away, Inc., 570 F. 2d, at 870-871; cf. Leh v. General Petroleum Corp., 382 U. S., at 65. To be sure, one way around these problems would be to say that the statute is tolled anytime the United States participates in any regulatory proceeding, regardless of what it contends or does in that proceeding. Even respondent, however, appears to recognize the undesirability of this result, and that such an interpretation has no support in the language or history of the statute.20 Ill We conclude, in sum, that the Clayton Act’s statute of limitations was not tolled, under § 5 (i), by the filing of the 20 We do not mean to suggest that no rational distinctions concerning the Government’s participation in regulatory proceedings can be drawn. It may be appropriate, in a given case, to apply § 5 (i) where the Government’s petition to intervene in fact charged a violation of the antitrust laws and demanded relief to prevent, restrain, or enjoin that violation. That, however, is not this case, and we expressly decline to offer any view as to the applicability vel non of § 5 (i) in such a context. GREYHOUND CORP. v. MT. HOOD STAGES, INC. 337 322 Burger, C. J., concurring Government’s petition to intervene in the ICC proceeding. The judgment of the Court of Appeals is therefore vacated, and the case is remanded for further proceedings consistent with this opinion.21 It is so ordered. Mr. Chief Justice Burger, concurring. I concur fully in the Court’s opinion, but with great reluctance ; in my view respondent is entitled to the award of treble damages ordered by the District Court. Given the Court’s analysis of the legal issues involved here, the opinion today has no occasion to focus on Greyhound’s egregious behavior toward Mt. Hood Stages—aimed at total destruction of a competitor. In the present case the jury found Greyhound not only to be in violation of the Sherman Act, but that it had fraudulently concealed its antitrust violations for more than a decade. Moreover, the Interstate Commerce Commission found that petitioner’s actions were “inspired by a desire to stifle competition,” in particular an intent to “injure or destroy” respondent. Mount Hood Stages, Inc., 104 M. C. C. 449, 461 (1968). Beyond its unlawful conduct, Greyhound took the added step of willfully disobeying the enforcement order of the United States District Court. In assessing criminal fines of $600,000 against Greyhound, the District Court, in a careful and detailed opinion, observed that Greyhound had “displayed a contemptuous reluctance to even commence compliance” with the court’s order. United States v. Greyhound 21 As already stated, supra, at 329, we limited our grant of certiorari to the issue of the applicability of § 5 (i). Respondent nevertheless argues that even if § 5 (i) is not applicable, the Clayton Act’s statute of limitations was tolled under equitable principles. Pursuant to the terms of our grant of certiorari, we see no compulsion—indeed, no justification—for our reaching this distinct issue. It will be for the Court of Appeals, on remand, to determine whether respondent may argue this point and, if so, its merits. Similarly, we express no view on what other issues may be raised on remand. 338 OCTOBER TERM, 1977 Burger, C. J., concurring 437U.S. Corp., 370 F. Supp. 881, 884 (ND Ill. 1974). The District Court went on to note: “In determining the extent of Greyhound’s willful defiance of the order, the court recognizes Greyhound’s record of purposeful non-action, protracted resistance, and emasculating interpretations of the order. The court also notes Greyhound’s ‘paper compliance’ program and the reluctance with which Greyhound’s top management became actively involved in securing compliance with the order. All of this suggests that Greyhound’s failure to comply with certain parts of the order was deliberate.” Ibid. These determinations by the District Court were upheld in every respect by the Court of Appeals. United States v. Greyhound Corp., 508 F. 2d 529 (CA7 1974). There is no question that Mount Hood has been injured substantially by Greyhound. Moreover, were it not for the statute of limitations in the Clayton Act, respondent would clearly receive the full measure of treble damages. However, I am bound to agree with the Court’s opinion that the explicit language of § 5 (i) of the Clayton Act, as amended, 15 U. S. C. § 16 (i) (1976 ed.), precludes a statutory tolling of the statute of limitations. But as the Court carefully stresses, ante, at 337 n. 21, we expressly do not reach respondent’s claim that the limitations period should be tolled on equitable grounds. The Court, of Appeals explicitly left this question open, 555 F. 2d 687, 701 n. 34, and the Court’s opinion today leaves it free to re-examine the issue on remand.* *The authority of a federal court, sitting as a chancellor, to toll a statute of limitations on equitable grounds is a well-established part of our jurisprudence. See, e. g., American Pipe & Constr. Co. v. Utah, 414 U. S. 538 (1974); Burnett v. New York Central R. Co., 380 U. S. 424 (1965); Telegraphers v. Railway Express Agency, 321 U. S. 342, 347-349 (1944). With respect to the limitations period of the Clayton Act, equitable tolling is particularly appropriate since the addition of a federal GREYHOUND CORP. v. MT. HOOD STAGES, INC. 339 322 Burger, C. J., concurring Since the Court’s remand allows for an inquiry into the issue of equitable tolling, the Court of Appeals may apply traditional equitable principles in reaching its decision. See, e. g., 2 J. Pomeroy, Equity Jurisprudence 90-143 (5th ed. 1941). limitations period in the Act was essentially a “procedural” change in the statute. American Pipe, supra, at 558 n. 29. 340 OCTOBER TERM, 1977 Syllabus 437 U. S. OPPENHEIMER FUND, INC., et al. v. SANDERS et al. CERTIORARI TO THE UNITED STATES COURT OF APPEALS FOR THE SECOND CIRCUIT No. 77-335. Argued February 28-March 1, 1978—Decided June 19, 1978 Respondents brought a class action under Fed. Rule Civ. Proc. 23 (b) (3) on behalf of themselves and a class of purchasers against petitioners (including an open-end investment fund, its management corporation, and a brokerage firm), seeking to recover the amount by which the allegedly artificially inflated price respondents paid for fund shares exceeded their value. Respondents sought to require petitioners to help compile a list of the names and addresses of the members of the plaintiff class from records kept by the fund’s transfer agent so that the individual notice required by Rule 23 (c) (2) could be sent. The class proposed by respondents numbered about 121,000 persons, of whom about 103,000 still held shares, and, since 171,000 persons currently held shares, approximately 68,000 were not members of the class. To compile a list of the class members’ names and addresses, the transfer agent’s employees would have had to sort manually through many records, keypunch 150,000 to 300,000 computer cards, and create several new computer programs, all for an estimated cost of over $16,000. Respondents’ proposed redefinition of the plaintiff class, opposed by petitioners, to include only those persons who bought fund shares during a specified period and who still held shares was rejected by the District Court as involving an arbitrary reduction in the class, but the court held that the cost of sorting out the list of class members was the petitioners’ responsibility, while also rejecting respondents’ proposal, opposed by petitioners, that the class notice be included in a regular fund mailing, because it would reach the 68,000 shareholders who were not class members. On petitioners’ appeal, the Court of Appeals affirmed, holding that the federal discovery rules authorized the District Court to order petitioners to assist in compiling the class fist and to bear the $16,000 expense incident thereto. Held: 1. Federal Rule Civ. Proc. 23 (d), which empowers district courts to enter appropriate orders in the handling of class actions, not the discovery rules, is the appropriate source of authority for the District Court’s order directing petitioners to help compile the list of class members. The information as to such members is sought to facilitate the sending of notice rather than to define or clarify issues in the case, OPPENHEIMER FUND, INC. v. SANDERS 341 340 Syllabus as is the function of the discovery rules, and thus cannot be forced into the concept of relevancy reflected in Fed. Rule Civ. Proc. 26(b)(1), which permits discovery “regarding any matter, not privileged, which is relevant to the subject matter involved in the pending action.” Pp. 350-356. 2. Where a defendant in a class action can perform one of the tasks necessary to send notice, such as identification, more efficiently than the representative plaintiff, the district court has discretion to order him to perform the task under Rule 23 (d), and also has some discretion in allocating the cost of complying with such an order, although as a general rule the representative plaintiff should bear all costs relating to the sending of notice because it is he who seeks to maintain the suit as a class action. See Eisen v. Carlisle & Jac quelin, 417 U. S. 156. Pp. 356-359. 3. Here, however, the District Court abused its discretion in requiring petitioners to bear the expense of identifying class members and in not requiring respondents to pay the transfer agent, where respondents can obtain the information sought by paying the transfer agent the same amount that petitioners would have to pay, the information must be obtained to comply with respondents’ obligation to provide notice to their class, and no special circumstances have been shown to warrant requiring petitioners to bear the expense. Pp. 359-364. (a) Petitioners’ opposition to respondents’ proposed redefinition of the class and to the method of sending notice is an insufficient reason for requiring petitioners to pay the transfer agent, because it is neither fair nor good policy to penalize a defendant for prevailing on an argument against a representative plaintiff’s proposals. Pp. 360-361. (b) Nor is the fact that $16,000 is a “relatively modest” sum in comparison to the fund’s assets a sufficient reason for requiring petitioners to bear the expenses, since the proper test is normally whether the cost is substantial, not whether it is “modest” in relation to ability to pay. Pp. 361-362. (c) The District Court’s order cannot be justified on the ground that part of the records in question were kept on computer tapes rather than in less modern forms. P. 362. (d) And petitioners should not be required to bear the identification expense simply because they are alleged to have breached a fiduciary duty to respondents and their class, since a bare allegation of wrongdoing, whether by breach of fiduciary duty or otherwise, is not a fair reason for requiring a defendant to undertake financial burdens and risks to further a plaintiff’s case. P. 363. 558 F. 2d 636, reversed and remanded. 342 OCTOBER TERM, 1977 Opinion of the Court 437 U. S. Powell, J., delivered the opinion for a unanimous Court. Alfred, Berman argued the cause for petitioners. With him on the briefs were Norman L. Greene, Gerald Gordon, John F. Davidson, and Daniel E. Kirsch. Donald N. Ruby argued the cause and filed a brief for respondents. Mr. Justice Powell delivered the opinion of the Court. Respondents are the representative plaintiffs in a class action brought under Fed. Rule Civ. Proc. 23 (b)(3). They sought to require petitioners, the defendants below, to help compile a list of the names and addresses of the members of the plaintiff class from records kept by the transfer agent for one of petitioners so that the individual notice required by Rule 23 (c) (2) could be sent. The Court of Appeals for the Second Circuit held that the federal discovery rules, Fed. Rules Civ. Proc. 26-37, authorize the District Court to order petitioners to assist in compiling the list and to bear the $16,000 expense incident thereto. We hold that Rule 23 (d), which concerns the conduct of class actions, not the discovery rules, empowers the District Court to direct petitioners to help compile such a list. We further hold that, although the District Court has some discretion in allocating the cost of complying with such an order, that discretion was abused in this case. We therefore reverse and remand. I Petitioner Oppenheimer Fund, Inc. (Fund), is an open-end diversified investment fund registered under the Investment Company Act of 1940, 15 U. S. C. § 80a-l et seq. (1976 ed.). The Fund and its agents sell shares to the public at their net asset value plus a sales charge. Petitioner Oppenheimer Management Corp. (Management Corp.) manages the Fund’s investment portfolio. Pursuant to an investment advisory OPPENHEIMER FUND, INC. v. SANDERS 343 340 Opinion of the Court agreement, the Fund pays Management Corp, a fee which is computed in part as a percentage of the Fund’s net asset value. Petitioner Oppenheimer & Co. is a brokerage firm that owns 82% of the stock of Management Corp., including all of its voting stock. The individual petitioners are directors or officers of the Fund or Management Corp., or partners in Oppenheimer & Co. Respondents bought shares in the Fund at various times in 1968 and 1969. On March 26, May 12, and June 18, 1969, they filed three separate complaints, later consolidated, which alleged that the petitioners, other than the Fund, had violated federal securities laws in 1968 and 1969 by issuing or causing to be issued misleading prospectuses and annual reports about the Fund.1 In particular, respondents alleged that the prospectuses and reports failed to disclose the fact that the Fund invested in “restricted” securities,2 the risks involved in such investments, and the method used to value the restricted securities on the Fund’s books. They also alleged that the restricted securities had been overvalued on the Fund’s books, causing the Fund’s net asset value, and thus the price of shares in the Fund, to be inflated artificially. On behalf of themselves and a class of purchasers, respondents sought to recover from petitioners, other than the Fund, the amount by 1 The complaints alleged violations of the Securities Act of 1933, 15 U. S. C. § 77a et seq. (1976 ed.), the Securities Exchange Act of 1934, 15 U. S. C. § 78a et seq. (1976 ed.), the Investment Company Act of 1940, 15 U. S. C. § 80a-l et seq. (1976 ed.), and rules promulgated under these Acts. They also alleged pendent state-law claims of fraud and breach of fiduciary duty. 2 “Restricted” securities are “securities acquired directly or indirectly from the issuer thereof, or from an affiliate of such issuer, in a transaction or chain of transactions not involving any public offering . . . .” 17 CFR § 230.144 (a) (3) (1977). The public sale or distribution of such securities is restricted under the Securities Act of 1933 until the securities are registered or an exemption from registration becomes available. See 15 U. S. C. §§ 77d, 77e (1976 ed.). 344 OCTOBER TERM, 1977 Opinion of the Court 437 U. S. which the price they paid for Fund shares exceeded the shares’ value.3 In April 1973, respondents moved pursuant to Fed. Rule Civ. Proc. 23 (b)(3) for an order allowing them to represent a class of plaintiffs consisting of all persons who bought shares in the Fund between March 28, 1968, and April 24, 1970.4 Relying on Eisen v. Carlisle & Jacquelin, 54 F. R. D. 565 (SDNY 1972), respondents also sought an order directing petitioners to pay for the notice to absent class members required by Fed. Rule Civ. Proc. 23(c)(2). On May 1, 1973, however, the Court of Appeals for the Second Circuit held that the District Court in Eisen erred in ordering the defendants to pay 90% of the cost of notifying members of a Rule 23 (b)(3) plaintiff class. Eisen v. Carlisle & Jacquelin (Eisen III), 479 F. 2d 1005. Respondents thereupon deposed employees of the Fund’s transfer agent, which kept records from which the class members’ names and addresses could be derived, in order to develop information relevant to issues of manageability, identification, and methods of notice upon which the District Court would have to pass. These employees’, statements, together with information supplied by the Fund, established that the class proposed by respondents numbered about 3 Later in the proceedings respondents’ counsel estimated that the average recovery per class member would be about $15, and that the aggregate recovery might be $1% million. In a separate count of their complaints, respondents also sought derivative relief on behalf of the Fund to recover excessive management fees paid by the Fund to Management Corp, as a result of the Fund’s allegedly inflated net asset value. 4 Petitioners denied the material allegations of the complaints. In addition, they alleged a setoff against respondents and their class to the extent that the price paid by the Fund to redeem shares had exceeded their value. The non-Fund petitioners also alleged that if they were liable to respondents and their class for overvaluation of Fund shares, then the Fund would be liable to them for excess amounts received by the Fund as a result of the overvaluation. OPPENHEIMER FUND, INC. v. SANDERS 345 340 Opinion of the Court 121,000 persons. About 103,000 still held shares in the Fund, while some 18,000 had sold their shares after the end of the class period. Since about 171,000 persons currently held shares in the Fund, it appeared that approximately 68,000 current Fund shareholders were not members of the class. The transfer agent’s employees also testified that in order to compile a list of the class members’ names and addresses, they would have to sort manually through a considerable volume of paper records, keypunch between 150,000 and 300,000 computer cards, and create eight new computer programs for use with records kept on computer tapes that either are in existence or would have to be created from the paper records. See App. 163-212. The cost of these operations was estimated in 1973 to exceed $16,000. Having learned all this, and in the face of Eisen III, respondents moved to redefine the class to include only those persons who had bought Fund shares between March 28, 1968, and April 24, 1970, and who still held shares in the Fund. Respondents also proposed that the class notice be inserted in one of the Fund’s periodic mailings to its current shareholders, and they offered to pay the cost of printing and inserting the notices, which was about $5,000. App. 146. These proposals would have made it unnecessary to compile a separate list of the members of the redefined class in order to notify them. Petitioners opposed redefinition of the class on the ground that it arbitrarily would exclude about 18,000 former Fund shareholders who had bought shares during the relevant period, possibly to their prejudice. They also opposed including the class notice in a Fund mailing which would reach the 68,000 current shareholders who were not class members. This, petitioners feared, could set off a wave of selling to the detriment of the Fund.5 5 Petitioners submitted the sworn affidavit of Robert Galli, Secretary of the Fund and Administrative Vice President and Secretary of Management Corp., which stated that this was a real possibility in light of '“the 346 OCTOBER TERM, 1977 Opinion of the Court 437U.S. On May 15, 1975, more than six years after the litigation began, the District Court ruled on the motions then pending. Sanders v. Levy, 20 Fed. Rules Serv. 2d 1218 (SDNY 1975). The court first held that the suit met the requirements for classaction treatment under Rule 23(b)(3). Id., at 1220-1221. It then rejected respondents’ proposed redefinition of the class because it “would involve an arbitrary reduction in the class.” Id., at 1221.6 At the same time, however, the court held that “the cost of culling out the list of class members ... is the responsibility of defendants.” Ibid. The only explanation given was that “the expense is relatively modest and it is defendants who are seeking to have the class defined in a manner which appears to require the additional expense.” Ibid. Finally, the court rejected respondents’ proposal that the class notice be included in a regular Fund mailing. Noting that the mailing would reach many current Fund shareholders who were not members of the class, the District Judge said that his “solution to this problem starts with my earlier ruling that it is the responsibility of defendants to cull out from their records a list of all class members and provide this list to plaintiffs. Plaintiffs will then have the responsibility to prepare the necessary notice and mail it at their expense.” Id., at 1222.7 current loss of investor confidence in the stock market and the uncertain conditions under which that market exists at this time.” App. 130-131. G The District Court also rejected a proposal by petitioners to set April 25, 1969, as the closing date of the class period, holding that respondents had raised triable claims of misrepresentations after that date. 20 Fed. Rules Serv. 2d, at 1221-1222. 7 The court subsequently modified this order to allow the notice to class members who still were Fund shareholders to be inserted in the envelopes of a periodic Fund mailing, “provided that the notices are sent only to class members and that plaintiffs pay in full the Fund’s extra costs of mailing, including the costs of segregating the envelopes going to the class members from the envelopes going to' other Fund shareholders.” At the same time,, the court held that the Fund should bear the identification OPPENHEIMER FUND, INC. v. SANDERS 347 340 Opinion of the Court On petitioners’ appeal, a divided panel of the Court of Appeals reversed the District Court’s order insofar as it required petitioners to bear the cost required for the transfer agent to compile a list of the class members’ names and addresses. Sanders v. Levy, 558 F. 2d 636 (CA2 1976).8 The majority thought that Eisen IV, which had affirmed Eisen III in pertinent part, required respondents to pay this cost because the identification of class members is an integral step in the process of notifying them. 558 F. 2d, at 642.9 On rehearing en banc, however, the Court of Appeals reversed the panel’s decision and affirmed the District Court’s order by a vote of seven to three. Id., at 646?° It thought that Eisen IV did not control this case because respondents might obtain the class members’ names and addresses under the costs in the first instance, “without prejudice to the right of this defendant, at the conclusion of the action, to make whatever claim it would be legally entitled to make regarding reimbursement by another party.” The court denied the Fund’s request that respondents be required to post bond for the identification costs. 8 All three members of the panel agreed that the order allocating the expense of identification was appealable under the collateral-order doctrine of Cohen v. Beneficial Loan Corp., 337 U. S. 541 (1949). 558 F. 2d, at 638-639; id., at 643 (Hays, J., dissenting in part). We agree. See Eisen v. Carlisle & Jacquelin (Eisen IV), 417 U. S. 156, 171-172 (1974). The panel also unanimously affirmed the District Court’s ruling that the suit could proceed as a class action. 558 F. 2d, at 642-643; id., at 643 (Hays, J., dissenting in part). This issue is not before us. 9 The panel majority also suggested that the Fund should not be required to bear this expense because it, unlike the other petitioners, was not named as a defendant in the class-action portion of this suit. See id., at 640. The Fund itself, which is in the position of a defendant because it ultimately may be liable for any damages that respondents and their class recover, see n. 4, supra, does not argue in this Court that it should not bear the expense because it is not a formal defendant. We therefore do not rely on any distinction that might be drawn between the Fund and the other petitioners in this respect. 10 District Judge Palmieri, the author of the panel majority opinion, did not participate in the rehearing en banc. 348 OCTOBER TERM, 1977 Opinion of the Court 437U.S. federal discovery rules, Fed. Rules Civ. Proc. 26-37. The en banc court further held that although Rule 26 (c) protects parties from “undue burden or expense” in complying with discovery requests, the District Court did not abuse its discretion under that Rule in requiring petitioners to bear this expense. 558 F. 2d, at 649-650. By holding that the discovery rules apply to this case, the en banc court brought itself into conflict with the Court of Appeals for the Fifth Circuit, which recently had held: “The time and expense of gathering [class members’] names and addresses is a necessary predicate to providing each with notice of the action’s pendency without which the action may not proceed [citing Eisen IV]. Viewed in this context, it becomes strikingly clear that rather than being controlled by the federal civil discovery rules, identification of absentee class members’ names and addresses is part and parcel of rule 23 (c)(2)’s mandate that the class members receive ‘the best notice practicable under the circumstances, including individual notice to all members who can be identified through reasonable effort.’ ” In re Nissan Motor Corp. Antitrust Litigation, 552 F. 2d 1088, 1102 (1977). In the Fifth Circuit’s view, Rule 23 (d), which empowers district courts to enter appropriate orders in the handling of class actions, is the procedural device by which a district court may enlist the aid of a defendant in identifying class members to whom notice must be sent. The Nissan court found it unnecessary to decide whether Eisen IV requires a representative plaintiff always to bear the cost of identifying class members. Since the representative plaintiffs could perform the required search through the defendants’ records as readily as the defendants themselves■, and since the search had to be performed in order to advance the representative plaintiffs’ case, they were required to perform it and thus to bear its cost. See 552 F. 2d, at 1102-1103. OPPENHEIMER FUND, INC. v. SANDERS 349 340 Opinion of the Court We granted certiorari in the instant case to resolve the conflict that thus has arisen and to consider the underlying cost-allocation problems. 434 U. S. 919 (1977). II The issues in this case arise because of the notice requirement of Fed. Rule Civ. Proc. 23(c)(2), which provides in part: “In any class action maintained under subdivision (b) (3), the court shall direct to the members of the class the best notice practicable under the circumstances, including individual notice to all members who can be identified through reasonable effort.” In Eisen IV, the Court held that the plain language of this Rule “requires that individual notice be sent to all class members who can be identified with reasonable effort.” 417 U. S., at 177. The Court also found no authority for a district court to hold a preliminary hearing on the merits of a suit in order to decide which party should bear the cost required to prepare and mail the class notice. Id., at 177-178. Instead, it held: “In the absence of any support under Rule 23, [the representative plaintiff’s] effort to impose the cost of notice on [defendants] must fail. The usual rule is that a plaintiff must initially bear the cost of notice to the class. . . . Where, as here, the relationship between the parties is truly adversary, the plaintiff must pay for the cost of notice as part of the ordinary burden of financing his own suit.” Id., at 178-179. In Eisen IV, the defendants had offered to provide a list of many of the class members’ names and addresses at their own expense in the first instance, if the representative plaintiff would prepare and mail individual notice to these class members.11 Eisen IV therefore did not present issues concerning 11 See App. in Eisen v. Carlisle & Jacquelin, 0. T. 1973, No. 73-203, pp. 184-185. 350 OCTOBER TERM, 1977 Opinion of the Court 437U.S. either the procedure by which a representative plaintiff might require a defendant to help identify class members, or whether costs may be allocated to the defendant in such a case. The specific holding of Eisen IV is that where a representative plaintiff prepares and mails the class notice himself, he must bear the cost of doing so. The parties in the instant case center much of their argument on the questions whether the discovery rules authorize a district court to order a defendant to help identify the members of a plaintiff class so that individual notice can be sent and, if so, which rule applies in this case. For the reasons stated in Part A below, we hold that Rule 23 (d), not the discovery rules, is the appropriate source of authority for such an order. This conclusion, however, is not dispositive of the cost-allocation question. As we explain in Part B, we think that where a defendant can perform one of the tasks necessary to send notice, such as identification, more efficiently than the representative plaintiff, the district court has discretion to order him to perform the task under Rule 23 (d). In such cases, the district court also has some discretion in allocating the cost of complying with its order. In Part C, however, we conclude that the District Court abused its discretion in this case. A Although respondents’ request resembles discovery in that it seeks to obtain information, we are convinced that it more properly is handled under Rule 23(d). The critical point is that the information is sought to facilitate the sending of notice rather than to define or clarify issues in the case. The general scope of discovery is defined by Fed. Rule Civ. Proc. 26(b)(1) as follows: “Parties may obtain discovery regarding any matter, not privileged, which is relevant to the subject matter involved in the pending action, whether it relates to the claim or defense of the party seeking discovery or to the OPPENHEIMER FUND, INC. v. SANDERS 351 340 Opinion of the Court claim or defense of any other party, including the existence, description, nature, custody, condition and location of any books, documents, or other tangible things and the identity and location of persons having knowledge of any discoverable matter. It is not ground for objection that the information sought will be inadmissible at the trial if the information sought appears reasonably calculated to lead to the discovery of admissible evidence.” The key phrase in this definition—“relevant to the subject matter involved in the pending action”—has been construed broadly to encompass any matter that bears on, or that reasonably could lead to other matter that could bear on, any issue that is or may be in the case. See Hickman n. Taylor, 329 U. S. 495, 501 (1947).12 Consistently with the notice-pleading system established by the Rules, discovery is not limited to issues raised by the pleadings, for discovery itself is designed to help define and clarify the issues. Id., at 500-501. Nor is discovery limited to the merits of a case, for a variety of fact-oriented issues may arise during litigation that are not related to the merits.13 At the same time, “discovery, like all matters of procedure, has ultimate and necessary boundaries.” Id., at 507. Dis 12“[T]he court should and ordinarily does interpret 'relevant’ very broadly to mean matter that is relevant to anything that is or may become an issue in the litigation.” 4 J. Moore, Federal Practice If 26.56 [1], p. 26-131 n. 34 (2d ed. 1976). 13 For example, where issues arise as to jurisdiction or venue, discovery is available to ascertain the facts bearing on such issues. See id., If 26.56 [6]; Note, The Use of Discovery to Obtain Jurisdictional Facts, 59 Va. L. Rev. 533 (1973). Similarly, discovery often has been used to illuminate issues upon which a district court must pass in deciding whether a suit should proceed as a class action under Rule 23, such as numerosity, common questions, and adequacy of representation. See Annot., Discovery for Purposes of Determining Whether Class Action Requirements Under Rule 23 (a) and (b) of Federal Rules of Civil Procedure Are Satisfied, 24 A. L. R. Fed. 872 (1975). 352 OCTOBER TERM, 1977 Opinion of the Court 437U.S. covery of matter not “reasonably calculated to lead to the discovery of admissible evidence” is not within the scope of Rule 26 (b) (1). Thus, it is proper to deny discovery of matter that is relevant only to claims or defenses that have been stricken,14 or to events that occurred before an applicable limitations period, unless the information sought is otherwise relevant to issues in the case.15 For the same reason, an amendment to Rule 26 (b) was required to bring within the scope of discovery the existence and contents of insurance agreements under which an insurer may be liable to satisfy a judgment against a defendant, for that information ordinarily cannot be considered, and would not lead to information that could be considered, by a court or jury in deciding any issues.16 Respondents’ attempt to obtain the class members’ names and addresses cannot be forced into the concept of “relevancy” described above. The difficulty is that respondents do not seek this information for any bearing that it might have on issues in the case. See 558 F. 2d, at 653 (en banc dissent).17 14 See, e. g., United States v. ^16.81 Acres of Land, 514 F. 2d 627, 632 (CA7 1975); Bourget v. Government Employees Ins. Co., 313 F. Supp. 367, 372-373 (Conn. 1970), reversed on other grounds, 456 F. 2d 282 (CA2 1972). 15 See 4 J. Moore, Federal Practice 126.56 [1], pp. 26-126 to 26-128 (2d ed. 1976), and cases there cited. 16 Before Rule 26 (b) (2) was added in 1970, many courts held that such agreements were not within the scope of discovery, although other courts, swayed by the fact that revelation of such agreements tends to encourage settlements, held otherwise. See Advisory Committee’s Notes on 1970 Amendment to Fed. Rule Civ. Proc. 26, 28 U. S. C. App., p. 7777; 4 J. Moore, Federal Practice 126.62 [1] (2d ed. 1976). The Advisory Committee appears to have viewed this amendment as changing rather than clarifying the Rules, for it stated: “[T]he provision makes no change in existing law on discovery of indemnity agreements other than insurance agreements by persons carrying on an insurance business.” 28 U. S. C. App., p. 7778 (emphasis supplied). 17 This difficulty may explain why the District Court, after calling for briefs on the question whether the discovery rules applied, see Brief for OPPENHEIMER, FUND, INC. v. SANDERS 353 340 Opinion of the Court If respondents had sought the information because of its relevance to the issues, they would not have been willing, as they were, to abandon their request if the District Court would accept their proposed redefinition of the class and method of sending notice. Respondents argued to the District Court that they desired this information to enable them to send the class notice, and not for any other purpose. Taking them at their word, it would appear that respondents’ request is not within the scope of Rule 26 (b)(1).18 The en banc majority avoided holding that the class members’ names and addresses are “relevant to the subject matter involved in the pending action” within the meaning of Rule 26 (b)(1) simply because respondents need this information in Respondents 10 n. 4, did not expressly rely on those rules. See also Note, Allocation of Identification Costs in Class Actions: Sanders v. Levy, 91 Harv. L. Rev. 703, 708-709 (1978) (distinguishing between “information . . . sought solely to provide adequate notice” and “valid discovery”). In deciding whether a request comes within the discovery rules, a court is not required to blind itself to the purpose for which a party seeks information. Thus, when the purpose of a discovery request is to gather information for use in proceedings other than the pending suit, discovery properly is denied. See Mississippi Power Co. v. Pedbody Coal Co., 69 F. R. D. 558, 565-568 (SD Miss. 1976) ; Econo-Car International, Inc. v. Antilles Car Rentals, Inc., 61 F. R. D. 8, 10 (V. I. 1973), rev’d on other grounds, 499 F. 2d 1391 (CA3 1974). Likewise, discovery should be denied when a party’s aim is to delay bringing a case to trial, or embarrass or harass the person from whom he seeks discovery. See United States v. Howard, 360 F. 2d 373, 381 (CA3 1966) ; Bcdistrieri v. Holtzman, 52 F. R. D. 23, 24-25 (ED Wis. 1971). See also n. 20, infra. 18 Respondents contend that they should be able to obtain the class members’ names and addresses under the discovery rules because it is “well settled that [a] plaintiff is entitled to conduct discovery with respect to a broad range of matters which pertain to the maintenance of a class action under Rule 23.” Brief for Respondents 25 n. 17 ; see n. 13, supra. The difference between the cases relied on by respondents and this case is that respondents do not seek information because it may bear on some issue which the District Court must decide, but only for the purpose of sending notice. 354 OCTOBER TERM, 1977 Opinion of the Court 437U.S. order to send the class notice. Tacitly acknowledging that discovery must be aimed at illuminating issues in the case, the court instead hypothesized that there is “a potential issue in all [Rule 23(b)(3) class-action] litigation whether the required notice has properly been sent. A list of the names and addresses of the class members would of course be essential to the resolution of that issue.” 558 F. 2d, at 648. But aside from the fact that respondents themselves never pretended to be anticipating this “potential issue,” it is apparent that the “potential issue” cannot arise until respondents already have obtained the very information they seek.19 Nor do we perceive any other “potential issues” that could bring respondents’ request within the scope of legitimate discovery. In short, we do not think that the discovery rules are the right tool for this job.20 Rule 23, on the other hand, deals comprehensively with class actions, and thus is the natural place to look for authority for orders regulating the sending of notice. It is clear that Rule 23 (d) vests power in the district court to order one of the parties to perform the tasks necessary to send notice.21 19 Until respondents obtain the information and send the class notice, no issue can arise as to whether it was sent “properly.” 29 We do not hold that class members’ names and addresses never can be obtained under the discovery rules. There may be instances where this information could be relevant to issues that arise under Rule 23, see n. 13, supra, or where a party has reason to believe that communication with some members of the class could yield information bearing on these or other issues. Respondents make no such claims of relevance, however, and none is apparent here. Moreover, it may be doubted whether any of these purposes would require compilation of the names and addresses of all members of a large class. See Berland v. Mack, 48 F. R. D. 121, 126 (SDNY 1969). There is a distinction in principle between requests for identification of class members that are made to enable a party to send notice, and requests that are made for true discovery purposes. See n. 17, supra. 21 Although Rule 23 (c) (2) states that “the court shall direct” notice to class members, it commonly is agreed that the court should order one of OPPENHEIMER FUND, INC. v. SANDERS 355 340 Opinion of the Court Moreover, district courts sometimes have found it appropriate to order a defendant, rather than a representative plaintiff, to perform tasks other than identification that are necessary to the sending of notice.22 Since identification simply is another task that must be performed in order to send notice, we agree with the Court of Appeals for the Fifth Circuit that Rule 23 (d) also authorizes a district court in appropriate circumstances to require a defendant’s cooperation in identifying the class members to whom notice must be sent.23 We therefore turn to a consideration of the circumstances in which the parties to perform the necessary tasks. See Frankel, Some Preliminary Observations Concerning Civil Rule 23, 43 F. R. D. 39, 44 (1968); Kaplan, Continuing Work of the Civil Committee: 1966 Amendments of the Federal Rules of Civil Procedure (I), 81 Harv. L. Rev. 356, 398 n. 157 (1967). Rule 23 (d) provides that in the conduct of a class action, “the court may make appropriate orders: ... (2) requiring, for the protection of the members of the class or otherwise for the fair conduct of the action, that notice be given in such manner as the court may direct . . . ; [and] (5) dealing with similar procedural matters.” The Advisory Committee apparently contemplated that the court would make orders drawing on the authority of either Rule 23 (d) (2) or 23 (d) (5) in order to provide the notice required by Rule 23 (c)(2), for its note to Rule 23 (d)(2) states that “under subdivision (c)(2), notice must be ordered . . . Advisory Committee’s Notes to Fed. Rule Civ. Proc. 23, 28 U. S. C. App., p. 7768 (emphasis supplied). 22 Thus, a number of courts have required defendants in Rule 23 (b) (3) class actions to enclose class notices in their own periodic mailings to class members in order to reduce the expense of sending the notice, as respondents asked the District Court in this case to do. See, e. g., Ste. Marie v. Eastern R. Assn., 72 F. R. D. 443, 450 n. 2 (SDNY 1976); Gates v. Dalton, 67 F. R. D. 621, 633 (EDNY 1975); Popkin n. Wheelabrator-Frye, Inc., 20 Fed. Rules Serv. 2d 125, 130 (SDNY 1975). See also Eisen IV, 417 U. S., at 180 n. 1 (Douglas, J., dissenting in part). 23 Our conclusion that Rule 23 (d), not the discovery rules, is the appropriate source of authority is supported by the fact that, although a number of courts have ordered defendants to help identify class members in the course of ordering notice, few have relied on the discovery rules. See In re Nissan Motor Corp. Antitrust Litigation, 552 F. 2d 1088, 1101-1102 (CA5 1977) (collecting cases). 356 OCTOBER TERM, 1977 Opinion of the Court 437 U. S. such an order is appropriate and of how the cost of the defendant’s complying with such an order should be allocated. B Although the Fifth Circuit held that Rule 23 (d), not the discovery rules, authorizes a district court to order a defendant to provide information needed to identify class members to whom notice must be sent, it also suggested that principles embodied in the discovery rules for allocating the performance of tasks and payment of costs might be relevant to a district court’s exercise of discretion under Rule 23 (d). See Nissan, 552 F. 2d, at 1102. Petitioners and the en banc dissent, on the other hand, argue that Eisen IV always requires a representative plaintiff to pay all costs incident to sending notice, whether he or the defendant performs the required tasks. Eisen IV does not compel this latter conclusion, for it did not involve a situation where a defendant properly was ordered under Rule 23 (d) to perform any of the tasks necessary to sending the notice. The first question that a district court must consider under Rule 23 (d) is which party should perform particular tasks necessary to send the class notice. The general rule must be that the representative plaintiff should perform the tasks, for it is he who seeks to maintain the suit as a class action and to represent other members of his class. In Eisen IV we noted the general principle that a party must bear the “burden of financing his own suit,” 417 U. S., at 179. Thus ordinarily there is no warrant for shifting the cost of the representative plaintiff’s performance of these tasks to the defendant. In some instances, however, the defendant may be able to perform a necessary task with less difficulty or expense than could the representative plaintiff. In such cases, we think that the district court properly may exercise its discretion under Rule 23 (d) to order the defendant to perform the task in question. As the Nissan court recognized, in identify- OPPENHEIMER FUND, INC. v. SANDERS 357 340 Opinion of the Court ing the instances in which such an order may be appropriate, a rough analogy might usefully be drawn to practice under Rule 33 (c) of the discovery rules.24 Under that Rule, when one party directs an interrogatory to another party which can be answered by examination of the responding party’s business records, “it is a sufficient answer to such interrogatory to specify the records from which the answer may be derived or ascertained and to afford to the party serving the interrogatory reasonable opportunity to” examine and copy the records, if the burden of deriving the answer would be “substantially the same” for either party. Not unlike Eisen IV, this provision is intended to place the “burden of discovery upon its potential benefitee.” 25 The holding of Nissan represents application of a similar principle, for when the court concluded that the representative plaintiffs could derive the names and addresses of the class members from the defendants’ records with substantially the same effort as the defendants, it required the representative plaintiffs to perform this task and hence to bear the cost. See supra, at 348. But where the burden of deriving the answer would not be “substantially the same,” and the task could be performed more efficiently by the responding party, the discovery rules normally require the responding party to derive the answer itself.26 24 The analogy to the discovery rules is not perfect, for those rules contemplate that discovery will proceed without judicial intervention unless a party moves for a protective order under Rule 26 (c) or an order compelling discovery under Rule 37 (a). Rule 23, on the other hand, contemplates that the district court routinely must approve the form of the class notice and order how it should be sent and who should perform the necessary tasks. 25 Advisory Committee’s Notes on 1970 Amendment to Fed. Rule Civ. Proc. 33 (c), 28 U. S. C. App., p. 7793, quoting D. Louisell, Modem California Discovery 125 (1963). 26 See Foster v. Boise-Cascade, Inc., 20 Fed. Rules Serv. 2d 466, 470 (SD Tex. 1975); Chrapliwy n. Uniroyal, Inc., 17 Fed. Rules Serv. 2d 719, 722 (ND Ind. 1973); Advisory Committee’s Notes, supra, at 7793. 358 OCTOBER TERM, 1977 Opinion of the Court 437U.S. In those cases where a district court properly decides under Rule 23 (d) that a defendant rather than the representative plaintiff should perform a task necessary to send the class notice, the question that then will arise is which party should bear the expense. On one hand, it may be argued that this should be borne by the defendant because a party ordinarily must bear the expense of complying with orders properly issued by the district court; but Eisen IV strongly suggests that the representative plaintiff should bear this expense because it is he who seeks to maintain the suit as a class action. In this situation, the district court must exercise its discretion in deciding whether to leave the cost of complying with its order where it falls, on the defendant, or place it on the party that benefits, the representative plaintiff. Once again, a rough analogy might usefully be drawn to practice under the discovery rules. Under those rules, the presumption is that the responding party must bear the expense of complying with discovery requests, but he may invoke the district court’s discretion under Rule 26 (c) to grant orders protecting him from “undue burden or expense” in doing so, including orders conditioning discovery on the requesting .party’s payment of the costs of discovery. The analogy necessarily is imperfect, however, because in the Rule 23 (d) context, the defendant’s own case rarely will be advanced by his having performed the tasks. Cf. n. 30, infra. Thus, one of the reasons for declining to shift costs under Rule 26 (c) usually will be absent in the Rule 23 (d) context.27 For this reason, a district court exercising its discretion under Rule 23 (d) should be considerably more ready to place the cost of the defendant’s performing an ordered task on the representative plaintiff, who derives the benefit, than under Rule 26 (c). In 27 Cf., e. g., Hodgson v. Adams Drug Co., 15 Fed. Rules Serv. 2d 828, 830 (RI 1971); Adelman n. Nordberg Mjg. Co., 6 F. R. D. 383, 384 (ED Wis. 1947); 4A J. Moore, Federal Practice 33.20, pp. 33-113 to 33-114 (2d ed. 1975). OPPENHEIMER FUND, INC. v. SANDERS 359 340 Opinion of the Court the usual case, the test should be whether the expense is substantial, rather than, as under Rule 26 (c), whether it is “undue.” Nevertheless, in some instances, the expense involved may be so insubstantial as not to warrant the effort required to calculate it and shift it to the representative plaintiff. In Nissan, for example, the court did not find it necessary to direct the representative plaintiffs to reimburse the defendants for the expense of producing their files for inspection. In other cases, it may be appropriate to leave the cost where it falls because the task ordered is one that the defendant must perform in any event in the ordinary course of its business.28 Although we do not attempt to catalogue the instances in which a district court might be justified in placing the expense on the defendant, we caution that courts must not stray too far from the principle underlying Eisen IV that the representative plaintiff should bear all costs relating to the sending of notice because it is he who seeks to maintain the suit as a class action. C In this case, we think the District Court abused its discretion in requiring petitioners to bear the expense of identifying class members. The records containing the needed information are kept by the transfer agent, not petitioners. Since petitioners apparently have the right to control these records, and since the class members can be identified only by reference to them, the District Court acted within its authority under Rule 23 (d) in ordering petitioners to direct the transfer agent to make the records available to respondents. The preparation of the desired list requires, as indicated above, the manual sorting out of names and addresses from old 28 Thus, where defendants have been directed to enclose class notices in their own periodic mailings and the additional expense has not been substantial, representative plaintiffs have not been required to reimburse the defendants for envelopes or postage. See cases cited in n. 22, supra. 360 OCTOBER TERM, 1977 Opinion of the Court 437U.S. records maintained on paper, the keypunching of up to 300,000 computer cards, and the creation of new computer programs for use with extant tapes and tapes that would have to be created from the paper records. It appears that neither petitioners nor respondents can perform these tasks, for both sides assume that the list can be generated only by hiring the services of a third party, the transfer agent, for a sum exceeding $16,000. As the expense of hiring the transfer agent would be no greater for respondents, who seek the information, than for petitioners, respondents should bear the expense. See Nissan, 552 F. 2d, at 1102-1103.29 The District Court offered two reasons why petitioners should pay the transfer agent, but neither is persuasive. First, the court thought that petitioners should bear this cost because it was their opposition to respondents’ proposed redefinition of the class and method of sending notice that made it necessary to incur the cost. A district court necessarily has some discretion in deciding the composition of a proper class and how notice should be sent. Nor is it improper for the court to consider the potential impact that rulings on these issues may have on the expense that the representative plaintiff must bear in order to send the notice. See Eisen IV, 417 U. S., at 179 n. 16; id., at 179-181 (Douglas, J., dissenting in part). But it is neither fair nor good policy to penalize a defendant for prevailing on an argument against a representative plaintiff’s proposals. If a defendant’s argument has merit, it should be accepted regardless of his willingness to bear the extra expense that its acceptance would require. Otherwise, a defendant may be discouraged from advancing arguments entirely appropriate to the protection of his rights' or the rights of absent class members. The potential for inequity appears to have been realized 29 See also Note, Allocation of Identification Costs in Class Actions, 66 Calif. L. Rev. 105, 115 (1978). OPPENHEIMER FUND, INC. v. SANDERS Opinion of the Court 361 340 in this case. The District Court seems to have agreed with petitioners that respondents’ proposed redefinition of the class was improper.30 Otherwise its actions would be difficult to fathom, for its rejection of the proposed redefinition increased the cost to respondents as well as petitioners.31 By the same token, if the District Court believed that sending the notice to current Fund shareholders who were not class members might harm the Fund, it should not have required the Fund to buy protection from this threat. Yet it must have believed that the Fund would be harmed, for otherwise there was no reason to reject respondents’ proposal and thus increase the cost that respondents themselves would have to bear. For these reasons, we hold that the District Court erred in linking the questions of class definition and method of notice to the cost-allocation question. The second reason advanced by the District Court was that $16,000 is a “relatively modest” sum, presumably in comparison to the Fund’s total assets, which exceed $500 million. Although in some circumstances the ability of a party to bear a burden may be a consideration, the test in this respect normally should be whether the cost is substantial; not whether 30 The District Court characterized the proposal as “arbitrary,” Sanders v. Levy, 20 Fed. Rules Serv. 2d 1218, 1221 (SDNY 1975), and stated that it ruled “in favor of” petitioners on this issue, id., at 1222. Although the court also suggested that petitioners opposed the redefinition because it would reduce the res judicata effect of the judgment, id., at 1221, petitioners themselves never made this argument. We also note that the representative plaintiff in Eisen IV argued, without success, that the defendants should pay part of the cost of notice because of the supposed res judicata benefits to them from class-action treatment. Reply Brief for Petitioner in Eisen v. Carlisle & Jacquelin, O. T. 1973, No. 73-203, pp. 25-26. We did not think then, nor do we now, that an unwilling defendant should be forced to purchase these “benefits.” 31 Respondents were required to bear the additional expense at least of envelopes and postage for notice to class members who no longer held shares in the Fund. See n. 7, supra. 362 OCTOBER TERM, 1977 Opinion of the Court 437 U. S. it is “modest” in relation to ability to pay. In the context of a lawsuit in which the defendants deny all liability, the imposition on them of a threshold expense of $16,000 to enable the plaintiffs to identify their own class hardly can be viewed as an insubstantial burden. Cf. Eisen IV, supra, at 176. As the expenditure would benefit only respondents, we think that the amount of money involved here would cut strongly against the District Court’s holding, even if the principle of Nissan did not control. The panel dissent and the en banc majority suggested several additional reasons to justify the District Court’s order, none of which we find persuasive. Both opinions suggest that the fact that part of these records are kept on computer tapes justifies imposing a greater burden on petitioners than might be imposed on a party whose records are kept in another form. Thus, the panel dissent warned that potential defendants may be tempted to use computers “irretrievably [to bury] information to immunize business activity from later scrutiny,” 558 F. 2d, at 645 n. 1, and the en banc majority argued that even where no bad motive is present, “complex electronic processes may be required to extract information which might have been obtainable through a minimum of effort had different systems been used.” Id., at 649. We do not think these reasons justify the order in this case. There is no indication or contention that these petitioners have acted in bad faith to conceal information from respondents. In addition, although it may be expensive to retrieve information stored in computers when no program yet exists for the particular job, there is no reason to think that the same information could be extracted any less expensively if the records were kept in less modern forms. Indeed, one might expect the reverse to be true, for otherwise computers would not have gained such widespread use in the storing and handling of information. Finally, the suggestion that petitioners should have used “different systems” to keep their rec- OPPENHEIMER FUND, INC. v. SANDERS 363 340 Opinion of the Court ords borders on the frivolous. Apart from the fact that no one has suggested what “different systems” petitioners should have used, we do not think a defendant should be penalized for not maintaining his records in the form most convenient to some potential future litigants whose identity and perceived needs could not have been anticipated. See id., at 654 (en banc dissent). Respondents also contend that petitioners should be required to bear the identification expense because they are alleged to have breached a fiduciary duty to respondents and their class. See also id., at 645-646 (panel dissent). Although we had no occasion in Eisen IV to consider this argument, see 417 U. S., at 178, and n. 15, suggestions to this effect have met with trenchant criticism elsewhere.32 A bare allegation of wrongdoing, whether by breach of fiduciary duty or otherwise, is not a fair reason for requiring a defendant to undertake financial burdens and risks to further a plaintiff’s case. Nor would it be in the interests of the class of persons to whom a fiduciary duty is owed to require them, through the fiduciary, to help finance every suit by one of their number that alleges a breach of fiduciary duty, without regard to whether the suit has any merit. Ill Given that respondents can obtain the information sought here by paying the transfer agent the same amount that petitioners would have to pay, that the information must be obtained to comply with respondents’ obligation to provide notice to their class, and that no special circumstances have been shown to warrant requiring petitioners to bear the ex- 32 See, e. g., 558 F. 2d, at 640-641 (panel majority); Popkin n. Wheel-abrator-Frye, Inc., 20 Fed. Rules Serv. 2d, at 129s—130; Berland v. Mack, 48 F. R. D. 121, 131-132 (SDNY 1969); Note, 23 Kan. L. Rev. 309, 318-319 (1975). 364 OCTOBER TERM, 1977 Opinion of the Court 437 U. S. pense, we hold that the District Court abused its discretion in not requiring respondents to pay the transfer agent to identify the members of their own class. The judgment of the Court of Appeals is reversed, and the case is remanded for further proceedings consistent with this opinion. It is so ordered. OWEN EQUIPMENT & ERECTION CO. v. KROGER 365 Syllabus OWEN EQUIPMENT & ERECTION CO. v. KROGER, ADMINISTRATRIX CERTIORARI TO THE UNITED STATES COURT OF APPEALS FOR THE EIGHTH CIRCUIT No. 77-677. Argued April 18, 1978—Decided June 21, 1978 Respondent, a citizen of Iowa, sued for damages based on the wrongful death of her husband, who was electrocuted when the boom of a steel crane next to which he was walking came too close to a high-tension electric power line. The action was brought in federal court on the basis of diversity of citizenship against a Nebraska corporation (OPPD), whose negligent operation of the power line was alleged to have caused decedent’s death. OPPD then filed a third-party complaint against petitioner company which owned and operated the crane, alleging that petitioner’s negligence proximately caused the death. Respondent was thereafter granted leave to amend her complaint by naming petitioner, which she alleged to be a Nebraska corporation with its principal place of business in Nebraska, as an additional defendant. OPPD successfully moved for summary judgment, leaving petitioner as the sole defendant. Though in its answer petitioner admitted that it was a corporation organized and existing under the laws of Nebraska, during trial it was disclosed that petitioner’s principal place of business was in Iowa. Since both parties were thus Iowa citizens, petitioner moved to dismiss on the basis of lack of federal jurisdiction. After the jury had returned a verdict for respondent, the District Court denied petitioner’s motion to dismiss. The Court of Appeals affirmed, holding that under Mine Workers n. Gibbs, 383 U. S. 715, the District Court had jurisdictional power, in its discretion, to adjudicate the claim, which arose from the “core of 'operative facts’ giving rise to both [respondent’s] claim against OPPD and OPPD’s claim against [petitioner],” and that the District Court had properly exercised its discretion because petitioner had concealed its Iowa citizenship from respondent. Held: The District Court had no power to entertain respondent’s lawsuit against petitioner as a third-party defendant since diversity jurisdiction was lacking. Gibbs, supra, distinguished. Pp. 370-377. (a) A finding that federal and nonfederal claims arise from a “common nucleus of operative fact,” the Gibbs test, does not suffice to establish that a federal court has power to hear nonfederal as well as 366 OCTOBER TERM, 1977 Syllabus 437 U. S. federal claims, since, though the constitutional power to adjudicate the nonfederal claim may exist, it does not follow that statutory authorization has been granted. Aiding er v. Howard, 427 U. S. 1; Zahn n. International Paper Co., 414 U. S. 291. Pp. 370-373. (b) Here the relevant statute, 28 U. S. C. § 1332 (a)(1), which confers upon federal courts jurisdiction over civil actions where the amount in controversy exceeds $10,000 and is between citizens of different States, requires complete diversity of citizenship, and it is thus congressionally mandated that diversity jurisdiction is not available when any plaintiff is a citizen of the same State as any defendant, a situation that developed in this case when respondent amended her complaint. Pp. 373-374. (c) Under the Court of Appeals’ ancillary-jurisdiction theory a plaintiff could defeat the statutory requirement of complete diversity simply by suing only those defendants of diverse citizenship and waiting for them to implead nondiverse defendants. Pp. 374-375. (d) In determining whether jurisdiction over a nonfederal claim exists, the context in which that claim is asserted is crucial. Here the nonfederal claim was simply not ancillary to the federal one, as respondent’s claim against petitioner was entirely separate from her original claim against OPPD, and petitioner’s liability to her did not depend at all upon whether or not OPPD was also liable. Moreover, the nonfederal claim here was asserted by the plaintiff, who voluntarily chose to sue upon a state-law claim in federal court, whereas ancillary jurisdiction typically involves claims by a defending party haled into court against, his will, or by another person whose rights might be irretrievably lost unless he could assert them in an ongoing action in federal court. Pp. 375-376. 558 F. 2d 417, reversed. Stewart, J., delivered the opinion of the Court, in which Burger, C. J., and Marshall, Blackmun, Powell, Rehnquist, and Stevens, JJ., joined. White, J., filed a dissenting opinion, in which Brennan, J., joined, post, p. 377. Emil F. Sodoro argued the cause for petitioner. With him on the briefs were David A. Johnson and Ronald H. Stave. Warren C. Schrempp argued the cause for respondent. With him on the brief were John J. Hanley and Thomas G. McQuade. OWEN EQUIPMENT & ERECTION CO. v. KROGER 367 365 Opinion of the Court Mr. Justice Stewart delivered the opinion of the Court. In an action in which federal jurisdiction is based on diversity of citizenship, may the plaintiff assert a claim against a third-party defendant when there is no independent basis for federal jurisdiction over that claim? The Court of Appeals for the Eighth Circuit held in this case that such a claim is within the ancillary jurisdiction of the federal courts. We granted certiorari, 434 U. S. 1008, because this decision conflicts with several recent decisions of other Courts of Appeals.1 I On January 18, 1972, James Kroger was electrocuted when the boom of a steel crane next to which he was walking came too close to a high-tension electric power line. The respondent (his widow, who is the administratrix of his estate) filed a wrongful-death action in the United States District Court for the District of Nebraska against the Omaha Public Power District (OPPD). Her complaint alleged that OPPD’s negligent construction, maintenance, and operation of the power line had caused Kroger’s death. Federal jurisdiction was based on diversity of citizenship, since the respondent was a citizen of Iowa and OPPD was a Nebraska corporation. OPPD then filed a third-party complaint pursuant to Fed. Rule Civ. Proc. 14 (a)2 against the petitioner, Owen Equip 1 Fawvor n. Texaco, Inc., 546 F. 2d 636 (CA5); Saaljrank n. O’Daniel, 533 F. 2d 325 (CA6); Parker N. W. W. Moore & Sons, 528 F. 2d 764 (CA4); Joseph v. Chrysler Corp., 513 F. 2d 626 (CA3), aff’g 61 F. R. D. 347 (WD Pa.); Kenrose Mfg. Co. v. Fred Whitaker Co., 512 F. 2d 890 (CA4). 2 Rule 14 (a) provides in relevant part: “At any time after commencement of the action a defending party, as a third-party plaintiff, may cause a summons and complaint to be served upon a person not a party to the action who is or may be liable to him for all or part of the plaintiff’s claim against him. . . . The person served with the summons and third-party complaint, hereinafter called the third-party defendant, shall make his defenses to the third-party plaintiff’s 368 OCTOBER TERM, 1977 Opinion of the Court 437U.S. ment and Erection Co. (Owen), alleging that the crane was owned and operated by Owen, and that Owen’s negligence had been the proximate cause of Kroger’s death.3 OPPD later moved for summary judgment on the respondent’s complaint against it. While this motion was pending, the respondent was granted leave to file an amended complaint naming Owen as an additional defendant. Thereafter, the District Court granted OPPD’s motion for summary judgment in an unreported opinion.4 The case thus went to trial between the respondent and the petitioner alone. The respondent’s amended complaint alleged that Owen was “a Nebraska corporation with its principal place of busi- claim as provided in. Rule 12 and his counterclaims against the third-party plaintiff and cross-claims against other third-party defendants as provided in Rule 13. The third-party defendant may assert against the plaintiff any defenses which the third-party plaintiff has to the plaintiff’s claim. The third-party defendant may also assert any claim against the plaintiff arising out of the transaction or occurrence that is the subject matter of the plaintiff’s claim against the third-party plaintiff. The plaintiff may assert any claim against the third-party defendant arising out of the transaction or occurrence that is the subject matter of the plaintiff’s claim against the third-party plaintiff, and the third-party defendant thereupon shall assert his defenses as provided in Rule 12 and his counter-claims and cross-claims as provided in Rule 13.” 3 Under Rule 14 (a), a third-party defendant may not be impleaded merely because he may be liable to the plaintiff. See n. 2, supra; see also Advisory Committee’s Notes on 1946 Amendment to Fed. Rule Civ. Proc. 14, 28 U. S. C. App., pp. 7752-7753. While the third-party complaint in this case alleged merely that Owen’s negligence caused Kroger’s death, and the basis of Owen’s alleged liability to OPPD is nowhere spelled out, OPPD evidently relied upon the state common-law right of contribution among joint tortfeasors. See Dairyland Ins. Co. v. Mumert, 212 N. W. 2d 436, 438 (Iowa); Best v. Yerkes, 247 Iowa 800, 77 N. W. 2d 23. The petitioner has never challenged the propriety of the third-party complaint as such. 4 Judgment was entered pursuant to Fed. Rule Civ. Proc. 54 (b), and the Court of Appeals affirmed. Kroger v. Omaha Public Power Dist., 523 F. 2d 161 (CA8). OWEN EQUIPMENT & ERECTION CO. v. KROGER 369 365 Opinion of the Court ness in Nebraska.” Owen’s answer admitted that it was “a corporation organized and existing under the laws of the State of Nebraska,” and denied every other allegation of the complaint. On the third day of trial, however, it was disclosed that the petitioner’s principal place of business was in Iowa, not Nebraska,5 and that the petitioner and the respondent were thus both citizens of Iowa.® The petitioner then moved to dismiss the complaint for lack of jurisdiction. The District Court reserved decision on the motion, and the jury thereafter returned a verdict in favor of the respondent. In an unreported opinion issued after the trial, the District Court denied the petitioner’s motion to dismiss the complaint. The judgment was affirmed on appeal. 558 F. 2d 417. The Court of Appeals held that under this Court’s decision in Mine Workers v. Gibbs, 383 U. S. 715, the District Court had jurisdictional power, in its discretion, to adjudicate the respondent’s claim against the petitioner because that claim arose from the “core of ‘operative facts’ giving rise to both [respondent’s] claim against OPPD and OPPD’s claim against Owen.” 558 F. 2d, at 424. It further held that the District Court had properly exercised its discretion in proceeding to decide the case even after summary judgment had been granted to OPPD, because the petitioner had concealed its Iowa citizenship from the respondent. Rehearing en banc was denied by an equally divided court. 558 F. 2d 417. 5 The problem apparently was one of geography. Although the Missouri River generally marks the boundary between Iowa and Nebraska, Carter Lake, Iowa, where the accident occurred and where Owen had its main office, lies west of the river, adjacent to Omaha, Neb. Apparently the river once avulsed at one of its bends, cutting Carter Lake off from the rest of Iowa. 6 Title 28 U. S. C. § 1332 (c) provides that “[f]or the purposes of [diversity jurisdiction] . . . , a corporation shall be deemed a citizen of any State by which it has been incorporated and of the State where it has its principal place of business.” 370 OCTOBER TERM, 1977 Opinion of the Court 437 U. S. II It is undisputed that there was no independent basis of federal jurisdiction over the respondent’s state-law tort action against the petitioner, since both are citizens of Iowa. And although Fed. Rule Civ. Proc. 14 (a) permits a plaintiff to assert a claim against a third-party defendant, see n. 2, supra, it does not purport to say whether or not such a claim requires an independent basis of federal jurisdiction. Indeed, it could not determine that question, since it is axiomatic that the Federal Rules of Civil Procedure do not create or withdraw federal jurisdiction.7 In affirming the District Court’s judgment, the Court of Appeals relied upon the doctrine of ancillary jurisdiction, whose contours it believed were defined by this Court’s holding in Mine Workers v. Gibbs, supra. The Gibbs case differed from this one in that it involved pendent jurisdiction, which concerns the resolution of a plaintiff’s federal- and statelaw claims against a single defendant in one action. By contrast, in this case there was no claim based upon substantive federal law, but rather state-law tort claims against two different defendants. Nonetheless, the Court of Appeals was correct in perceiving that Gibbs and this case are two species of the same generic problem: Under what circumstances may a federal court hear and decide a state-law claim arising between citizens of the same State?8 But we believe that the Court of Appeals failed to understand the scope of the doctrine of the Gibbs case. The plaintiff in Gibbs alleged that the defendant union had violated the common law of Tennessee as well as the federal 7 Fed. Rule Civ. Proc. 82; see Snyder v. Harris, 394 U. S. 332; Sibbach n. Wilson & Co., 312 U. S. 1, 10. 8 No more than in Aldinger v. Howard, 427 U. S. 1, is it necessary to determine here “whether there are any ‘principled’ differences between pendent and ancillary jurisdiction; or, if there are, what effect Gibbs had on such differences.” Id., at 13. OWEN EQUIPMENT & ERECTION CO. v. KROGER 371 365 Opinion of the Court prohibition of secondary boycotts. This Court held that, although the parties were not of diverse citizenship, the District Court properly entertained the state-law claim as pendent to the federal claim. The crucial holding was stated as follows: “Pendent jurisdiction, in the sense of judicial power, exists whenever there is a claim ‘arising under [the] Constitution, the Laws of the United States, and Treaties made, or which shall be made, under their Authority . . . ,’ U. S. Const., Art. Ill, § 2, and the relationship between that claim and the state claim permits the conclusion that the entire action before the court comprises but one constitutional ‘case.’ . . . The state and federal claims must derive from a common nucleus of operative fact. But if, considered without regard to their federal or state character, a plaintiff’s claims are such that he would ordinarily be expected to try them all in one judicial proceeding, then, assuming substantiality of the federal issues, there is power in federal courts to hear the whole.” 383 U. S., at 725 (emphasis in original).9 It is apparent that Gibbs delineated the constitutional limits of federal judicial power. But even if it be assumed that the District Court in the present case had constitutional power to decide the respondent’s lawsuit against the petitioner,10 it does not follow that the decision of the Court of Appeals 9 The Court further noted that even when such power exists, its exercise remains a matter of discretion based upon “considerations of judicial economy, convenience and fairness to litigants,” 383 U. 8., at 726, and held that the District Court had not abused its discretion in retaining jurisdiction of the state-law claim, 10 Federal jurisdiction in Gibbs was based upon the existence of a question of federal law. The Court of Appeals in the present case believed that the “common nucleus of operative fact” test also determines the outer boundaries of constitutionally permissible federal jurisdiction when that jurisdiction is based upon diversity of citizenship. We may assume without deciding that the Court of Appeals was correct in this regard. See also n. 13, infra. 372 OCTOBER TERM, 1977 Opinion of the Court 437U.S. was correct. Constitutional power is merely the first hurdle that must be overcome in determining that a federal court has jurisdiction over a particular controversy. For the jurisdiction of the federal courts is limited not only by the provisions of Art. Ill of the Constitution, but also by Acts of Congress. Palmore v. United States, 411 U. S. 389, 401; Lockerty v. Phillips, 319 U. S. 182,187; Kline v. Burke Constr. Co., 260 IL S. 226, 234; Cary v. Curtis, 3 How. 236, 245. That statutory law as well as the Constitution may limit a federal court’s jurisdiction over nonfederal claims11 is well illustrated by two recent decisions of this Court, Aiding er v. Howard, 427 U. S. 1, and Zahn v. International Paper Co., 414 IT. S. 291. In Aldinger the Court held that a Federal District Court lacked jurisdiction over a state-law claim against a county, even if that claim was alleged to be pendent to one against county officials under 42 IL S. C. § 1983. In Zahn the Court held that in a diversity class action under Fed. Rule Civ. Proc. 23 (b)(3), the claim of each member of the plaintiff class must independently satisfy the minimum jurisdictional amount set by 28 IL S. C. § 1332 (a), and rejected the argument that jurisdiction existed over those claims that involved $10,000 or less as ancillary to those that involved more. In each case, despite the fact that federal and nonfederal claims arose from a “common nucleus of operative fact,” the Court held that the statute conferring jurisdiction over the federal claim did not allow the exercise of jurisdiction over the nonfederal claims.12 11 As used in this opinion, the term “nonfederal claim” means one as to which there is no independent basis for federal jurisdiction. Conversely, a “federal claim” means one as to which an independent basis for federal jurisdiction exists. 12 In Monell v. New York City Dept, of Social Services, 436 U. S. 658, we have overruled Monroe n. Pape, 365 U. S. 167, insofar as it held that political subdivisions are never amenable to suit under 42 U. S. C. § 1983—the basis of the holding in Aldinger that 28 U. S. C. § 1343 (3) OWEN EQUIPMENT & ERECTION CO. v. KROGER 373 365 Opinion of the Court The Aiding er and Zahn cases thus make clear that a finding that federal and nonfederal claims arise from a “common nucleus of operative fact/’ the test of Gibbs, does not end the inquiry into whether a federal court has power to hear the nonfederal claims along with the federal ones. Beyond this constitutional minimum, there must be an examination of the posture in which the nonfederal claim is asserted and of the specific statute that confers jurisdiction over the federal claim, in order to determine whether “Congress in [that statute] has . . . expressly or by implication negated” the exercise of jurisdiction over the particular nonfederal claim. Aiding er v. Howard, supra, at 18. Ill The relevant statute in this case, 28 U. S. C. § 1332 (a)(1), confers upon federal courts jurisdiction over “civil actions where the matter in controversy exceeds the sum or value of $10,000 . . . and is between . . . citizens of different States.” This statute and its predecessors have consistently been held to require complete diversity of citizenship.13 That is, diversity jurisdiction does not exist unless each defendant is a citizen of a different State from each plaintiff. Over the years Congress has repeatedly re-enacted or amended the statute conferring diversity jurisdiction, leaving intact this rule of complete diversity.14 Whatever may have been the original does not allow pendent jurisdiction of a state-law claim against a county. But Monell in no way qualifies the holding of Aldinger that the jurisdictional questions presented in a case such as this one are statutory as well as constitutional, a point on which the dissenters in Aldinger agreed with the Court. See 427 U. S., at 22 n. 3 (Brennan, J., joined by Marshall and Blackmun, JJ., dissenting). 13 E. g., Strawbridge v. Curtiss, 3 Cranch 267; Coal Co. v. Blatchford, 11 Wall. 172; Indianapolis v. Chase Nat. Bank, 314 U. S. 63, 69; American Fire & Cas. Co. v. Finn, 341 U. S. 6, 17. It is settled that complete diversity is not a constitutional requirement. State Farm Fire & Cas. Co. v. Tashire, 386 U. S. 523, 530-531. 14 The various Acts are enumerated and described in 1 J. Moore, Federal Practice IT 0.71 [4] (2d ed. 1977). 374 OCTOBER TERM, 1977 Opinion of the Court 437U.S. purposes of diversity-of-citizenship jurisdiction,15 this subsequent history clearly demonstrates a congressional mandate that diversity jurisdiction is not to be available when any plaintiff is a citizen of the same State as any defendant. Cf. Snyder v. Harris, 394 U. S. 332, 338-339.16 Thus it is clear that the respondent could not originally have brought suit in federal court naming Owen and OPPD as codefendants, since citizens of Iowa would have been on both sides of the litigation. Yet the identical lawsuit resulted when she amended her complaint. Complete diversity was destroyed just as surely as if she had sued Owen initially. In either situation, in the plain language of the statute, the “matter in controversy” could not be “between . . . citizens of different States.” It is a fundamental precept that federal courts are courts of limited jurisdiction. The limits upon federal jurisdiction, whether imposed by the Constitution or by Congress, must be neither disregarded nor evaded. Yet under the reasoning of the Court of Appeals in this case, a plaintiff could defeat the statutory requirement of complete diversity by the simple expedient of suing only those defendants who were of diverse citizenship and waiting for them to implead nondiverse defendants.17 If, as the Court of Appeals thought, a “common 15 See C. Wright, Law of Federal Courts §23 (3d ed. 1976), for a discussion of the various theories that have been advanced to explain the constitutional grant of diversity-of-citizenship jurisdiction. 16 Notably, Congress enacted § 1332 as part of the Judicial Code of 1948, 62 Stat. 930, shortly after Rule 14 was amended in 1946. When the Rule was amended, the Advisory Committee noted that “in any case where the plaintiff could not have joined the third party originally because of jurisdictional limitations such as lack of diversity of citizenship, the majority view is that any attempt by the plaintiff to amend his complaint and assert a claim against the impleaded third party would be unavailing.” 28 U. S. C. App., p. 7752. The subsequent re-enactment without relevant change of the diversity statute may thus be seen as evidence of congressional approval of that “majority view.” 17 This is not an unlikely hypothesis, since a defendant in a tort suit OWEN EQUIPMENT & ERECTION CO. v. KROGER 375 365 Opinion of the Court nucleus of operative fact” were the only requirement for ancillary jurisdiction in a diversity case, there would be no principled reason why the respondent in this case could not have joined her cause of action against Owen in her original complaint as ancillary to her claim against OPPD. Congress’ requirement of complete diversity would thus have been evaded completely. It is true, as the Court of Appeals noted, that the exercise of ancillary jurisdiction over nonfederal claims has often been upheld in situations involving impleader, cross-claims or counterclaims.18 But in determining whether jurisdiction such as this one would surely try to limit his liability by impleading any joint tortfeasors for indemnity or contribution. Some commentators have suggested that the possible abuse of third-party practice could be dealt with under 28 U. S. C. § 1359, which forbids collusive attempts to create federal jurisdiction. See, e. g., 3 J. Moore, Federal Practice f 14.27 [1], p. 14-571 (2d ed. 1974); 6 C. Wright & A. Miller, Federal Practice and Procedure §1444, pp. 231-232 (1971); Note, Rule 14 Claims and Ancillary Jurisdiction, 57 Va. L. Rev. 265, 274-275 (1971). The dissenting opinion today also expresses this view. Post, at 383. But there is nothing necessarily collusive about a plaintiff’s selectively suing only those tortfeasors of diverse citizenship, or about the named defendants’ desire to implead joint tortfeasors. Nonetheless, the requirement of complete diversity would be eviscerated by such a course of events. 18 The ancillary jurisdiction of the federal courts derives originally from cases such as Freeman v. Howe, 24 How. 450, which held that when federal jurisdiction “effectively controls the property or fund under dispute, other claimants thereto should be allowed to intervene in order to protect their interests, without regard to jurisdiction.” Aldinger n. Howard, ^7 U. S., at 11. More recently, it has been said to include cases that involve multiparty practice, such as compulsory counterclaims, e. g., Moore v. New York Cotton Exchange, 270 U. S. 593; impleader, e. g., H. L. Peterson Co. v. Applewhite, 383 F. 2d 430, 433 (CA5); Dery v. Wyer, 265 F. 2d 804 (CA2); cross-claims, e. g., LASA Per L’Industria Del Marmo Soc. Per Azioni v. Alexander, 414 F. 2d 143 (CA6); Scott v. Fancher, 369 F. 2d 842, 844 (CA5); Glen Falls Indemnity Co. n. United States ex rel. Westinghouse Electric Supply Co., 229 F. 2d 370, 373-374 (CA9); or intervention as of right, e. g., Phelps v. Oaks, 117 U. S. 236, 241; Smith Petroleum Service, Inc. v. Monsanto Chemical Co., 420 F. 2d 1103, 1113-1115 (CA5). 376 OCTOBER TERM, 1977 Opinion of the Court 437U.S. over a nonfederal claim exists, the context in which the non-federal claim is asserted is crucial. See A Idinger v. Howard, 427 U. S., at 14. And the claim here arises in a setting quite different from the kinds of nonfederal claims that have been viewed in other cases as falling within the ancillary jurisdiction of the federal courts. First, the nonfederal claim in this case was simply not ancillary to the federal one in the same sense that, for example, the impleader by a defendant of a third-party defendant always is. A third-party complaint depends at least in part upon the resolution of the primary lawsuit. See n. 3, supra. Its relation to the original complaint is thus not mere factual similarity but logical dependence. Cf. Moore n. New York Cotton Exchange, 270 U. S. 593, 610. The respondent’s claim against the petitioner, however, was entirely separate from her original claim against OPPD, since the petitioner’s liability to her depended not at all upon whether or not OPPD was also liable. Far from being an ancillary and dependent claim, it was a new and independent one. Second, the nonfederal claim here was asserted by the plaintiff, who voluntarily chose to bring suit upon a state-law claim in a federal court. By contrast, ancillary jurisdiction typically involves claims by a defending party haled into court against his will, or by another person whose rights might be irretrievably lost unless he could assert them in an ongoing action in a federal court.19 A plaintiff cannot complain if ancillary jurisdiction does not encompass all of his possible claims in a case such as this one, since it is he who has chosen the federal rather than the state forum and must thus accept its limitations. “[T]he efficiency plaintiff seeks so avidly is available without question in the state courts.” Kenrose Mjg. Co. v. Fred Whitaker Co., 512 F. 2d 890, 894 (CA4).20 19 See n. 18, supra. 20 Whether Iowa’s statute of limitations would now bar an action by the respondent in an Iowa court is, of course, entirely a matter of state OWEN EQUIPMENT & ERECTION CO. v. KROGER 377 365 White, J., dissenting It is not unreasonable to assume that, in generally requiring complete diversity, Congress did not intend to confine the jurisdiction of federal courts so inflexibly that they are unable to protect legal rights or effectively to resolve an entire, logically entwined lawsuit. Those practical needs are the basis of the doctrine of ancillary jurisdiction. But neither the convenience of litigants nor considerations of judicial economy can suffice to justify extension of the doctrine of ancillary jurisdiction to a plaintiff’s cause of action against a citizen of the same State in a diversity case. Congress has established the basic rule that diversity jurisdiction exists under 28 U. S. C. § 1332 only when there is complete diversity of citizenship. “The policy of the statute calls for its strict construction.” Healy v. Ratta, 292 U. S. 263, 270; Indianapolis v. Chase Nat. Bank, 314 U. S. 63, 76; Thomson v. Gaskill, 315 U. S. 442, 446; Snyder v. Harris, 394 U. S., at 340. To allow the requirement of complete diversity to be circumvented as it was in this case would simply flout the congressional command.21 Accordingly, the judgment of the Court of Appeals is reversed. It is so ordered. Mr. Justice White, with whom Mr. Justice Brennan joins, dissenting. The Court today states that “[i]t is not unreasonable to assume that, in generally requiring complete diversity, Congress did not intend to confine the jurisdiction of federal courts so law. See Iowa Code §614.10 (1977). Compare 558 F. 2d, at 420, with id., at 432 n. 42 (Bright, J., dissenting); cf. Burnett n. New York Centred R. Co., 380 U. S. 424, 431-432, and n. 9. 21 Our holding is that the District Court lacked power to entertain the respondent’s lawsuit against the petitioner. Thus, the asserted inequity in the respondent’s alleged concealment of its citizenship is irrelevant. Federal judicial power does not depend upon “prior action or consent of the parties.” American Fire & Cas. Co. v. Finn, 341 U. S., at 17-18. 378 OCTOBER TERM, 1977 White, J., dissenting 437U.S. inflexibly that they are unable . . . effectively to resolve an entire, logically entwined lawsuit.” Ante, at 377. In spite of this recognition, the majority goes on to hold that in diversity suits federal courts do not have the jurisdictional power to entertain a claim asserted by a plaintiff against a third-party defendant, no matter how entwined it is with the matter already before the court, unless there is an independent basis for jurisdiction over that claim. Because I find no support for such a requirement in either Art. Ill of the Constitution or in any statutory law, I dissent from the Court’s “unnecessarily grudging” 1 approach. The plaintiff below, Mrs. Kroger, chose to bring her lawsuit against the Omaha Public Power District (OPPD) in Federal District Court. No one questions the power of the District Court to entertain this claim, for Mrs. Kroger at the time was a citizen of Iowa, OPPD was a citizen of Nebraska, and the amount in controversy was greater than $10,000; jurisdiction therefore existed under 28 U. S. C. § 1332 (a). As permitted by Fed. Rule Civ. Proc. 14 (a), OPPD impleaded petitioner Owen Equipment & Erection Co. (Owen). Although OPPD’s claim against Owen did not raise a federal question and although it was alleged that Owen was a citizen of the same State as OPPD, the parties and the court apparently believed that the District Court’s ancillary jurisdiction encompassed this claim. Subsequently, Mrs. Kroger asserted a claim against Owen, everyone believing at the time that these two parties were citizens of different States. Because it later came to light that Mrs. Kroger and Owen were in fact both citizens of Iowa, the Court concludes that the District Court lacked jurisdiction over the claim. In Mine Workers v. Gibbs, 383 U. S. 715, 725 (1966), we held that once a claim has been stated that is of sufficient substance to confer subject-matter jurisdiction on the federal dis- 1 See Mine Workers v. Gibbs, 383 U. S. 715, 725 (1966). OWEN EQUIPMENT & ERECTION CO. v. KROGER 379 365 White, J., dissenting trict court, the court has judicial power to consider a non-federal claim if it and the federal claim2 are derived from “a common nucleus of operative fact.” Although the specific facts of that case concerned a state claim that was said to be pendent to a federal-question claim, the Court’s language and reasoning were broad enough to cover the instant factual situation: “ [I]f, considered without regard to their federal or state character, a plaintiff’s claims are such that he would ordinarily be expected to try them all in one judicial proceeding, then, assuming substantiality of the federal issues, there is power in federal courts to hear the whole.” Ibid, (footnote omitted). In the present case, Mrs. Kroger’s claim against Owen and her claim against OPPD derived from a common nucleus of fact; this is necessarily so because in order for a plaintiff to assert a claim against a third-party defendant, Fed. Rule Civ. Proc. 14 (a) requires that it “arisfe] out of the transaction or occurrence that is the subject matter of the plaintiff’s claim against the third-party plaintiff . . . .” Furthermore, the substantiality of the claim Mrs. Kroger asserted against OPPD is unquestioned. Accordingly, as far as Art. Ill of the Constitution is concerned, the District Court had power to entertain Mrs. Kroger’s claim against Owen. The majority correctly points out, however, that the analysis cannot stop here. As Aldinger v. Howard, 427 U. S. 1 (1976), teaches, the jurisdictional power of the federal courts may be limited by Congress, as well as by the Constitution. In Aldinger, although the plaintiff’s state claim against Spokane County was closely connected with her 42 U. S. C. § 1983 claim against the county treasurer, the Court held that the District Court did not have pendent jurisdiction over the state claim, for, under the Court’s precedents at that time, it was thought that Congress had specifically determined not to confer on the federal courts jurisdiction over civil rights 21 use the terms “federal claim” and “nonfederal claim” in the same sense that the majority uses them. See ante, at 372 n. 11. 380 OCTOBER TERM, 1977 White, J., dissenting 437 U.S. claims against cities and counties. That being so, the Court refused to allow “the federal courts to fashion a jurisdictional doctrine under the general language of Art. Ill enabling them to circumvent this exclusion . . . .” 427 U. 8., at 16.3 In the present case, the only indication of congressional intent that the Court can find is that contained in the diversity jurisdictional statute, 28 U. S. C. § 1332 (a), which states that “district courts shall have original jurisdiction of all civil actions where the matter in controversy exceeds the sum or value of $10,000 . . . and is between . . . citizens of different States .... Because this statute has been interpreted as requiring complete diversity of citizenship between each plaintiff and each defendant, Strawbridge v. Curtiss, 3 Cranch 267 (1806), the Court holds that the District Court did not have ancillary jurisdiction over Mrs. Kroger’s claim against Owen. In so holding, the Court unnecessarily expands the scope of the complete-diversity requirement while substantially limiting the doctrine of ancillary jurisdiction. The complete-diversity requirement, of course, could be viewed as meaning that in a diversity case, a federal district court may adjudicate only those claims that are between parties of different States. Thus, in order for a defendant to implead a third-party defendant, there would have to be diversity of citizenship; the same would also be true for crossclaims between defendants and for a third-party defendant’s claim against a plaintiff. Even the majority, however, refuses to read the complete-diversity requirement so broadly; it 3 We were careful in Aldinger to point out the limited nature of our holding: “There are, of course, many variations in the language which Congress has employed to confer jurisdiction upon the federal courts, and we decide here only the issue of so-called ‘pendent party’ jurisdiction with respect to a claim brought under §§ 1343 (3) and 1983. Other statutory grants and other alignments of parties and claims might call for a different result ” 427 U. 8., at 18. OWEN EQUIPMENT & ERECTION CO. v. KROGER 381 365 White, J., dissenting recognizes with seeming approval the exercise of ancillary jurisdiction over nonfederal claims in situations involving impleader, cross-claims, and counterclaims. See ante, at 375. Given the Court’s willingness to recognize ancillary jurisdiction in these contexts, despite the requirements of § 1332 (a), I see no justification for the Court’s refusal to approve the District Court’s exercise of ancillary jurisdiction in the present case. It is significant that a plaintiff who asserts a claim against a third-party defendant is not seeking to add a new party to the lawsuit. In the present case, for example, Owen had already been brought into the suit by OPPD, and, that having been done, Mrs. Kroger merely sought to assert against Owen a claim arising out of the same transaction that was already before the court. Thus the situation presented here is unlike that in Aiding er, supra, wherein the Court noted: “[I]t is one thing to authorize two parties, already present in federal court by virtue of a case over which the court has jurisdiction, to litigate in addition to their federal claim a state-law claim over which there is no independent basis of federal jurisdiction. But it is quite another thing to permit a plaintiff, who has asserted a claim against one defendant with respect to which there is federal jurisdiction, to join an entirely different defendant on the basis of a state-law claim over which there is no independent basis of federal jurisdiction, simply because his claim against the first defendant and his claim against the second defendant ‘derive from a common nucleus of operative fact.’ . . . True, the same considerations of judicial economy would be served insofar as plaintiff’s claims ‘are such that he would ordinarily be expected to try them all in one judicial proceeding . . ..’ [Gibbs, 383 U. S., at 725.] But the addition of a completely new party would run counter to the well-established principle that federal courts, as opposed to state trial courts of 382 437 U.S. OCTOBER TERM, 1977 White, J., dissenting general jurisdiction, are courts of limited jurisdiction marked out by Congress.” 427 U. S., at 14-15. Because in the instant case Mrs. Kroger merely sought to assert a claim against someone already a party to the suit, considerations of judicial economy, convenience, and fairness to the litigants—the factors relied upon in Gibbs—support the recognition of ancillary jurisdiction here. Already before the court was the whole question of the cause of Mr. Kroger’s death. Mrs. Kroger initially contended that OPPD was responsible; OPPD in turn contended that Owen’s negligence had been the proximate cause of Mr. Kroger’s death. In spite of the fact that the question of Owen’s negligence was already before the District Court, the majority requires Mrs. Kroger to bring a separate action in state court in order to assert that very claim. Even if the Iowa statute of limitations will still permit such a suit, see ante, at 376-377, n. 20, considerations of judicial economy are certainly not served by requiring such duplicative litigation.4 The majority, however, brushes aside such considerations of convenience, judicial economy, and fairness because it concludes that recognizing ancillary jurisdiction over a plaintiff’s claim against a third-party defendant would permit the plaintiff to circumvent the complete-diversity requirement and thereby “flout the congressional command.” Since the plain- 4 It is true that prior to trial OPPD was dismissed as a party to the suit and that, as we indicated in Gibbs, the dismissal prior to trial of the federal claim will generally require the dismissal of the nonfederal claim as well. See 383 U. S., at 726. Given the unusual facts of the present case, however—in particular, the fact that the actual location of Owen’s principal place of business was not revealed until the third day of trial—fairness to the parties would lead me to conclude that the District Court did not abuse its discretion in retaining jurisdiction over Mrs. Kroger’s claim against Owen. Under the Court’s disposition, of course, it would not matter whether or not the federal claim is tried, for in either situation the court would have no jurisdiction over the plaintiff’s nonfederal claim against the third-party defendant. OWEN EQUIPMENT & ERECTION CO. v. KROGER 383 365 White, J., dissenting tiff in such a case does not bring the third-party defendant into the suit, however, there is no occasion for deliberate circumvention of the diversity requirement, absent collusion with the defendant. In the case of such collusion, of which there is absolutely no indication here,5 the court can dismiss the action under the authority of 28 U. S. C. § 1359.6 In the absence of such collusion, there is no reason to adopt an absolute rule prohibiting the plaintiff from asserting those claims that he may properly assert against the third-party defendant pursuant to Fed. Rule Civ. Proc. 14 (a). The plaintiff in such a situation brings suit against the defendant only, with absolutely no assurance that the defendant will decide or be able to implead a particular third-party defendant. Since the plaintiff has no control over the defendant’s decision to implead a third party, the fact that he could not have originally sued that party in federal court should be irrelevant. Moreover, the fact that a plaintiff in some cases may be able to foresee the subsequent chain of events leading to the impleader does not seem to me to be a sufficient reason to declare that a district court does not have the power to exercise ancillary jurisdiction over the plaintiff’s claims against the third-party defendant.7 5 When Mrs. Kroger brought suit, it was believed that Owen was a citizen of Nebraska, not Iowa. Therefore, had she desired at that time to make Owen a party to the suit, she would have done so directly by naming Owen as a defendant. 6 Section 1359 states: “A district court shall not have jurisdiction of a civil action in which any party, by assignment or otherwise, has been improperly or collusively made or joined to invoke the jurisdiction of such court.” 7 Under the Gibbs analysis, recognition of the district court’s power to hear a plaintiff’s nonfederal claim against a third-party defendant in a diversity suit would not mean that the court would be required to entertain such claims in all cases. The district court would have the discretion to dismiss the nonfederal claim if it concluded that the interests of judicial economy, convenience, and fairness would not be served by the retention of the claim in the federal lawsuit. See Gibbs, 383 U. S., at 726. Ac 384 OCTOBER TERM, 1977 White, J., dissenting 437 U. S. We have previously noted that “ [subsequent decisions of this Court indicate that Strawbridge is not to be given an expansive reading.” State Farm Fire & Cas. Co. v. Tashire, 386 U. S. 523, 531 n. 6 (1967). In light of this teaching, it seems to me appropriate to view § 1332 as requiring complete diversity only between the plaintiff and those parties he actually brings into the suit. Beyond that, I would hold that in a diversity case the District Court has power, both constitutional and statutory, to entertain all claims among the parties arising from the same nucleus of operative fact as the plaintiff’s original, jurisdiction-conferring claim against the defendant. Accordingly, I dissent from the Court’s disposition of the present case. cordingly, the majority’s concerns that lead it to conclude that ancillary jurisdiction should not be recognized in the present situation could be met on a case-by-case basis, rather than by the absolute rule it adopts. MINCEY v. ARIZONA 385 Syllabus MINCEY v. ARIZONA CERTIORARI TO THE SUPREME COURT OF ARIZONA No. 77-5353. Argued February 21, 1978—Decided June 21, 1978 During a narcotics raid on petitioner’s apartment by an undercover police officer and several plainclothes policemen, the undercover officer was shot and killed, and petitioner was wounded, as were two other persons in the apartment. Other than looking for victims of the shooting and arranging for medical assistance, the narcotics agents, pursuant to a police department directive that police officers should not investigate incidents in which they are involved, made no further investigation. Shortly thereafter, however, homicide detectives arrived on the scene to take charge of the investigation, and they proceeded to conduct an exhaustive four-day warrantless search of the apartment, which included the opening of dresser drawers, the ripping up of carpets, and the seizure of 200 to 300 objects. In the evening of the same day as the raid, one of the detectives went to the hospital where petitioner was confined in the intensive-care unit, and, after giving him Miranda warnings, persisted in interrogating him while he was lying in bed barely conscious, encumbered by tubes, needles, and a breathing apparatus, and despite the fact that he repeatedly asked that the interrogation stop until he could get a lawyer. Subsequently, petitioner was indicted for, and convicted of, murder, assault, and narcotics offenses. At his trial in an Arizona court, during which much of the evidence introduced against him was the product of the four-day search, and on appeal, petitioner contended that the evidence used against him had been unlawfully seized from his apartment without a warrant and that statements obtained from him at the hospital, used to impeach his credibility, were inadmissible because they had not been made voluntarily. The Arizona Supreme Court reversed the murder and assault convictions on state-law grounds, but affirmed the narcotics convictions, holding that the warrantless search of a homicide scene is permissible under the Fourth and Fourteenth Amendments and that petitioner’s statements in the hospital were voluntary. Held: 1. The “murder scene exception” created by the Arizona Supreme Court to the warrant requirement is inconsistent with the Fourth and Fourteenth Amendments, and the warrantless search of petitioner’s apartment was not constitutionally permissible simply because a homicide had occurred there. Pp. 388-395. 386 OCTOBER TERM, 1977 Syllabus 437 U. S. (a) The search cannot be justified on the ground that no constitutionally protected right of privacy was invaded, it being one thing to say that one who is legally taken into police custody has a lessened right of privacy in his person, and quite another to argue that he also has a lessened right of privacy in his entire house. Pp. 391-392. (b) Nor can the search be justified on the ground that a possible homicide inevitably presents an emergency situation, especially since there was no emergency threatening life or limb, all persons in the apartment having been located before the search began. Pp. 392-393. (c) The seriousness of the offense under investigation did not itself create exigent circumstances of the kind that under the Fourth Amendment justify a warrantless search, where there is no indication that evidence would be lost, destroyed, or removed during the time required to obtain a search warrant and there is no suggestion that a warrant could not easily and conveniently have been obtained. Pp. 393-394. (d) The Arizona Supreme Court’s guidelines for the “murder scene exception” did not afford sufficient protection to a person in whose home a homicide or assault occurs, where they conferred unbridled discretion upon the individual officer to interpret such terms as “reasonable . . . search,” “serious personal injury with likelihood of death where there is reason to suspect foul play,” and “reasonable period,” it being this kind of judgmental assessment of the reasonableness and scope of a proposed search that the Fourth Amendment requires be made by a neutral and objective magistrate, not a police officer. Pp. 394r-395. 2. Due process requires that the statements obtained from petitioner in the hospital not be used in any way against him at his trial, where it is apparent from the record that they were not “the product of his free and rational choice,” Greenwald v. Wisconsin, 390 U. 8. 519, 521, but to the contrary that he wanted not to answer his interrogator, and that while he was weakened by pain and shock, isolated from family, friends, and legal counsel, and barely conscious, his will was simply overbome. While statements made by a defendant in circumstances violating the strictures of Miranda v. Arizona, 384 U. 8. 436, are admissible for impeachment if their “trustworthiness . . . satisfies legal standards,” Harris v. New York, 401 U. S. 222, 224; Oregon v. Hass, 420 U. 8. 714, 722, any criminal trial use against a defendant of his involuntary statement is a denial of due process of law. Pp. 396-402. 115 Ariz. 472, 566 P. 2d 273, reversed and remanded. Stewart, J., delivered the opinion of the Court, in which Burger, C. J., and Brennan, White, Marshall, Blackmun, Powell, and Stevens, JJ., joined, and in Part I of which Rehnquist, J., joined. Marshall, J., MINCEY v. KKLZßMK 387 385 Opinion of the Court filed a concurring opinion, in which Brennan, J., joined, post, p. 402. Rehnquist, J., filed an opinion concurring in part and dissenting in part, post, p. 405. Richard Oseran argued the cause for petitioner. With him on the brief was Frederick S. Klein. Galen H. Wilkes, Assistant Attorney General of Arizona, argued the cause for respondent. With him on the brief were Bruce E. Babbitt, Attorney General, Philip G. Urry, Assistant Attorney General, and William J. Schafer III. Mr. Justice Stewart delivered the opinion of the Court. On the afternoon of October 28, 1974, undercover police officer Barry Headricks of the Metropolitan Area Narcotics Squad knocked on the door of an apartment in Tucson, Ariz., occupied by the petitioner, Rufus Mincey. Earlier in the day, Officer Headricks had allegedly arranged to purchase a quantity of heroin from Mincey and had left, ostensibly to obtain money. On his return he was accompanied by nine other plainclothes policemen and a deputy county attorney. The door was opened by John Hodgman, one of three acquaintances of Mincey who were in the living room of the apartment. Officer Headricks slipped inside and moved quickly into the bedroom. Hodgman attempted to slam the door in order to keep the other officers from entering, but was pushed back against the wall. As the police entered the apartment, a rapid volley of shots was heard from the bedroom. Officer Headricks emerged and collapsed on the floor. When other officers entered the bedroom they found Mincey lying on the floor, wounded and semiconscious. Officer Headricks died a few hours later in the hospital. The petitioner was indicted for murder, assault,1 and three 1 The assault charge was based on the wounding of a person in the living room who was hit by a bullet that came through the wall. 388 OCTOBER TERM, 1977 Opinion of the Court 437 U. S. counts of narcotics offenses. He was tried at a single trial and convicted on all the charges. At his trial and on appeal, he contended that evidence used against him had been unlawfully seized from his apartment without a warrant and that statements used to impeach his credibility were inadmissible because they had not been made voluntarily. The Arizona Supreme Court reversed the murder and assault convictions on state-law grounds,2 but affirmed the narcotics convictions. 115 Ariz. 472, 566 P. 2d 273. It held that the warrantless search of a homicide scene is permissible under the Fourth and Fourteenth Amendments and that Mincey’s statements were voluntary. We granted certiorari to consider these substantial constitutional questions. 434 U. S. 902. I The first question presented is whether the search of Mincey’s apartment was constitutionally permissible. After the shooting, the narcotics agents, thinking that other persons in the apartment might have been injured, looked about quickly for other victims. They found a young woman wounded in the bedroom closet and Mincey apparently unconscious in the bedroom, as well as Mincey’s three acquaintances (one of whom had been wounded in the head) in the living room. Emergency assistance was requested, and some medical aid was administered to Officer Headricks. But the agents refrained from further investigation, pursuant to a Tucson Police Department directive that police officers should not investigate incidents in which they are involved. They neither searched further nor seized any evidence; they merely guarded the suspects and the premises. Within 10 minutes, however, homicide detectives who had 2 The state appellate court held that the jury had been improperly instructed on criminal intent. It appears from the record in this case that the retrial of the petitioner on the murder and assault charges was stayed by the trial court after certiorari was granted by this Court. MINCEY v. ARIZONA 389 385 Opinion of the Court heard a radio report of the shooting arrived and took charge of the investigation. They supervised the removal of Officer Headricks and the suspects, trying to make sure that the scene was disturbed as little as possible, and then proceeded to gather evidence. Their search lasted four days,3 during which period the entire apartment was searched, photographed, and diagrammed. The officers opened drawers, closets, and cupboards, and inspected their contents; they emptied clothing pockets; they dug bullet fragments out of the walls and floors; they pulled up sections of the carpet and removed them for examination. Every item in the apartment was closely examined and inventoried, and 200 to 300 objects were seized. In short, Mincey’s apartment was subjected to an exhaustive and intrusive search. No warrant was ever obtained. The petitioner’s pretrial motion to suppress the fruits of this search was denied after a hearing. Much of the evidence introduced against him at trial (including photographs and diagrams, bullets and shell casings, guns, narcotics, and narcotics paraphernalia) was the product of the four-day search of his apartment. On appeal, the Arizona Supreme Court reaffirmed previous decisions in which it had held that the warrantless search of the scene of a homicide is constitutionally permissible.4 It stated its ruling as follows: “We hold a reasonable, warrantless search of the scene of a homicide—or of a serious personal injury with likelihood of death where there is reason to suspect foul play— 3 The police also returned to the apartment in November 1974, at the request of the petitioner’s landlord, to remove property of the petitioner that remained in the apartment after his lease had expired on October 31. 4 State v. Sample, 107 Ariz. 407, 489 P. 2d 44; State ex rel. Berger v. Superior Court, 110 Ariz. 281, 517 P. 2d 1277; State n. Duke, 110 Ariz. 320, 518 P. 2d 570. The Court of Appeals for the Ninth Circuit reversed the denial of a petition for a writ of habeas corpus filed by the defendant whose conviction was upheld in State v. Sample, supra, on the ground, inter alia, that the warrantless search of the homicide scene violated the Fourth and Fourteenth Amendments. Sample n. Eyman, 469 F. 2d 819. 390 OCTOBER TERM, 1977 Opinion of the Court 437 U. S. does not violate the Fourth Amendment to the United States Constitution where the law enforcement officers were legally on the premises in the first instance. . . . For the search to be reasonable, the purpose must be limited to determining the circumstances of death and the scope must not exceed that purpose. The search must also begin within a reasonable period following the time when the officials first learn of the murder (or potential murder).” 115 Ariz., at 482, 566 P. 2d, at 283. Since the investigating homicide detectives knew that Officer Headricks was seriously injured, began the search promptly upon their arrival at the apartment, and searched only for evidence either establishing the circumstances of death or “relevant to motive and intent or knowledge (narcotics, e. g.),” id., at 483, 566 P. 2d, at 284, the court found that the warrantless search of the petitioner’s apartment had not violated the Fourth and Fourteenth Amendments. We cannot agree. The Fourth Amendment proscribes all unreasonable searches and seizures, and it is a cardinal principle that “searches conducted outside the judicial process, without prior approval by judge or magistrate, are per se unreasonable under the Fourth Amendment—subject only to a few specifically established and well-delineated exceptions.” Katz v. United States, 389 U. S. 347, 357 (footnotes omitted); see also South Dakota v. Opperman, 428 U. S. 364, 381 (Powell, J., concurring); Coolidge n. New Hampshire, 403 U. S. 443, 481; Vale v. Louisiana, 399 U. S. 30, 34; Terry n. Ohio, 392 U. S. 1, 20; Trupiano n. United States, 334 U. S. 699, 705. The Arizona Supreme Court did not hold that the search of the petitioner’s apartment fell within any of the exceptions to the warrant requirement previously recognized by this Court, but gather that the search of a homicide scene should be recognized as an additional exception. Several reasons are advanced by the State to meet its “bur MINCEY v. ARIZONA 391 385 Opinion of the Court den ... to show the existence of such an exceptional situation” as to justify creating a new exception to the warrant requirement. See Vale v. Louisiana, supra, at 34; United States v. Jeffers, 342 U. S. 48. 51. None of these reasons, however, persuades us of the validity of the generic exception delineated by the Arizona Supreme Court. The first contention is that the search of the petitioner’s apartment did not invade any constitutionally protected right of privacy. See Katz n. United States, supra. This argument appears to have two prongs. On the one hand, the State urges that by shooting Officer Headricks, Mincey forfeited any reasonable expectation of privacy in his apartment. We have recently rejected a similar waiver argument in Michigan v. Tyler, 436 U. S. 499, 505-506; it suffices here to say that this reasoning would impermissibly convict the suspect even before the evidence against him was gathered.5 On the other hand, the State contends that the police entry to arrest Mincey was so great an invasion of his privacy that the additional intrusion caused by the search was constitutionally irrelevant. But this claim is hardly tenable in light of the extensive nature of this search. It is one thing to say that one who is legally taken into police custody has a lessened right of privacy in his person. See United States v. Edwards, 415 U. S. 800, 808-809; United States v. Robinson, 414 U. S. 218. It is quite another to argue that he also has a lessened right of privacy in his entire house. Indeed this very argument was rejected when it was advanced to support the warrantless search of a dwelling where a search occurred as “incident” to the arrest of its occupant. Chimel v. California, 395 U. S. 752, 766 n. 12. 5 Moreover, this rationale would be inapplicable if a homicide occurred at the home of the victim or of a stranger, yet the Arizona cases indicate that a warrantless search in such a case would also be permissible under the “murder scene exception.” Cf. State v. Sample, supra, at 409, 489 P. 2d, at 46. 392 OCTOBER TERM, 1977 Opinion of the Court 437U.S. Thus, this search cannot be justified on the ground that no constitutionally protected right of privacy was invaded. The State’s second argument in support of its categorical exception to the warrant requirement is that a possible homicide presents an emergency situation demanding immediate action. We do not question the right of the police to respond to emergency situations. Numerous state6 and federal7 cases have recognized that the Fourth Amendment does not bar police officers from making warrantless entries and searches when they reasonably believe that a person within is in need of immediate aid. Similarly, when the police come upon the scene of a homicide they may make a prompt warrantless search of the area to see if there are other victims or if a killer is still on the premises. Cf. Michigan v. Tyler, supra, at 509-510. “The need to protect or preserve life or avoid serious injury is justification for what would be otherwise illegal absent an exigency or emergency.” Wayne v. 6K g., People v. HUI, 12 Cal. 3d 731, 753-757, 528 P. 2d 1, 18-21; Patrick v. State, 227 A. 2d 486, 488-490 (Del.); People v. Brooks, 1 Ill. App. 3d 767, 775-777, 289 N. E. 2d 207, 212-214; Maxey v. State, 251 Ind. 645, 649-650, 244 N. E. 2d 650, 653-654; Davis v. State, 236 Md. 389, 395-397, 204 A. 2d 76, 80-82; State v. Hardin, 90 Nev. 10, 518 P. 2d 151; State v. Gosser, 50 N. J. 438, 446-448, 236 A. 2d 377, 381-382; People v. Mitchell, 39 N. Y. 2d 173, 347 N. E. 2d 607; State v. Pires, 55 Wis. 2d 597, 603-605, 201 N. W. 2d 153, 156-158. Other cases are collected in Note, The Emergency Doctrine, Civil Search and Seizure, and the Fourth Amendment, 43 Ford. L. Rev. 571, 584 n. 102 (1975). See also ALI Model Code of Pre-Arraignment Procedure § SS 260.5 (Prop. Off. Draft 1975). By citing these cases and those in the note following, of course, we do not mean to approve the specific holding of each case. 7 E. g., Root n. Gauper, 438 F. 2d 361, 364-365 (CA8); United States v. Barone, 330 F. 2d 543 (CA2); Wayne v. United States, 115 U. S. App. D. C. 234, 238-243, 318 F. 2d 205, 209-214 (opinion of Burger, J.); United States n. James, 408 F. Supp. 527, 533 (SD Miss.); United States ex rel. Parson v. Anderson, 354 F. Supp. 1060, 1086-1087 (Del.), aff’d, 481 F. 2d 94 (CA3); see Warden v. Hayden, 387 U. S. 294, 298-299; McDonald v. United States, 335 U. S. 451, 45A-456; Johnson v. United States, 333 U. S. 10, 14r-15. MINCEY v. 393 385 Opinion of the Court United States, 115 U. S. App. D. C. 234, 241, 318 F. 2d 205, 212 (opinion of Burger, J.). And the police may seize any evidence that is in plain view during the course of their legitimate emergency activities. Michigan v. Tyler, supra, at 509-510; Coolidge v. New Hampshire, 403 U. S., at 465-466. But a warrantless search must be “strictly circumscribed by the exigencies which justify its initiation,” Terry v. Ohio, 392 U. S., at 25-26, and it simply cannot be contended that this search was justified by any emergency threatening life or limb. All the persons in Mincey’s apartment had been located before the investigating homicide officers arrived there and began their search. And a four-day search that included opening dresser drawers and ripping up carpets can hardly be rationalized in terms of the legitimate concerns that justify an emergency search. Third, the State points to the vital public interest in the prompt investigation of the extremely serious crime of murder. No one can doubt the importance of this goal. But the public interest in the investigation of other serious crimes is comparable. If the warrantless search of a homicide scene is reasonable, why not the warrantless search of the scene of a rape, a robbery, or a burglary? “No consideration relevant to the Fourth Amendment suggests any point of rational limitation” of such a doctrine. Chimel v. California, supra, at 766. Moreover, the mere fact that law enforcement may be made more efficient can never by itself justify disregard of the Fourth Amendment. Cf. Coolidge v. New Hampshire, supra, at 481. The investigation of crime would always be simplified if warrants were unnecessary. But the Fourth Amendment reflects the view of those who wrote the Bill of Rights that the privacy of a person’s home and property may not be totally sacrificed in the name of maximum simplicity in enforcement of the criminal law. See United States v. Chadwick, 433 U. S. 1, 6-11. For this reason, warrants are 394 OCTOBER TERM, 1977 Opinion of the Court 437U.S. generally required to search a person’s home or his person unless “the exigencies of the situation” make the needs of law enforcement so compelling that the warrantless search is objectively reasonable under the Fourth Amendment. McDonald v. United States, 335 U. S. 451, 456; Johnson v. United States, 333 U. S. 10, 14^15. See, e. g., Chimel v. California, supra (search of arrested suspect and area within his control for weapons or evidence); Warden v. Hayden, 387 U. S. 294, 298-300 (“hot pursuit” of fleeing suspect); Schmerber v. California, 384 U. S. 757, 770-771 (imminent destruction of evidence) ; see also supra, at 392-393. Except for the fact that the offense under investigation was a homicide, there were no exigent circumstances in this case, as, indeed, the Arizona Supreme Court recognized. 115 Ariz., at 482, 566 P. 2d, at 283. There was no indication that evidence would be lost, destroyed, or removed during the time required to obtain a search warrant. Indeed, the police guard at the apartment minimized that possibility. And there is no suggestion that a search warrant could not easily and conveniently have been obtained. We decline to hold that the seriousness of the offense under investigation itself creates exigent circumstances of the kind that under the Fourth Amendment justify a warrantless search. Finally, the State argues that the “murder scene exception” is constitutionally permissible because it is narrowly confined by the guidelines set forth in the decision of the Arizona Supreme Court, see supra, at 389—390.8 In light of the extensive search that took place in this case it may be questioned what protection the guidelines afford a person in whose home a homicide or assault occurs. Indeed, these so-called guidelines 8 The State also relies on the fact that observance of these guidelines can be enforced by a motion to suppress evidence. But the Fourth Amendment “is designed to prevent, not simply to redress, unlawful police action.” Chimel n. California, 395 U. S. 752, 766 n. 12. MINCEY v. ARIZONA 395 385 Opinion of the Court are hardly so rigidly confining as the State seems to assert. They confer unbridled discretion upon the individual officer to interpret such terms as “reasonable . . . search,” “serious personal injury with likelihood of death where there is reason to suspect foul play,” and “reasonable period.” It is precisely this kind of judgmental assessment of the reasonableness and scope of a proposed search that the Fourth Amendment requires be made by a neutral and objective magistrate, not a police officer. See, e. g., United States v. United States District Court, 407 U. S. 297, 316; Coolidge v. New Hampshire, supra, at 449-453; Mancusi v. DeForte, 392 U. S. 364, 371; Wong Sun v. United States, 371 U. S. 471, 481-482. It may well be that the circumstances described by the Arizona Supreme Court would usually be constitutionally sufficient to warrant a search of substantial scope. But the Fourth Amendment requires that this judgment in each case be made in the first instance by a neutral magistrate. “The point of the Fourth Amendment, which often is not grasped by zealous officers, is not that it denies law enforcement the support of the usual inferences which reasonable men draw from evidence. Its protection consists in requiring that those inferences be drawn by a neutral and detached magistrate instead of being judged by the officer engaged in the often competitive enterprise of ferreting out crime.” Johnson v. United States, supra, at 13-14. In sum, we hold that the “murder scene exception” created by the Arizona Supreme Court is inconsistent with the Fourth and Fourteenth Amendments—that the warrantless search of Mincey’s apartment was not constitutionally permissible simply because a homicide had recently occurred there.9 9 To what extent, if any, the evidence found in Mincey’s apartment was permissibly seized under established Fourth Amendment standards will be for the Arizona courts to resolve on remand. 396 OCTOBER TERM, 1977 Opinion of the Court 437U.S. II Since there will presumably be a new trial in this case,10 it is appropriate to consider also the petitioner’s contention that statements he made from a hospital bed were involuntary, and therefore could not constitutionally be used against him at his trial. Mincey was brought to the hospital after the shooting and taken immediately to the emergency room where he was examined and treated. He had sustained a wound in his hip, resulting in damage to the sciatic nerve and partial paralysis of his right leg. Tubes were inserted into his throat to help him breathe, and through his nose into his stomach to keep him from vomiting; a catheter was inserted into his bladder. He received various drugs, and a device was attached to his arm so that he could be fed intravenously. He was then taken to the intensive care unit. At about eight o’clock that evening, Detective Hust of the Tucson Police Department came to the intensive care unit to interrogate him. Mincey was unable to talk because of the tube in his mouth, and so he responded to Detective Hust’s questions by writing answers on pieces of paper provided by the hospital.11 Hust told Mincey he was under arrest for the murder of a police officer, gave him the warnings required by Miranda v. Arizona, 384 U. S. 436, and began to ask questions about the events that had taken place in Mincey’s apartment a few hours earlier. Although Mincey asked repeatedly that the interrogation stop until he could get a lawyer, Hust continued to question him until almost midnight. 10 See also n. 2, supra. 11 Because of the way in which the interrogation was conducted, the only contemporaneous record consisted of Mincey’s written answers. Hust testified that the next day he went over this document and made a few notes to help him reconstruct the conversation. In a written report dated about a week later, Hust transcribed Mincey’s answers and added the questions he believed he had asked. It was this written report that was used to cross-examine Mincey at his subsequent trial. MINCEY v. ARIZONA 397 385 Opinion of the Court After a pretrial hearing, see Jackson v. Denno, 378 U. S. 368, the trial court found that Mincey had responded to this interrogation voluntarily.12 When Mincey took the witness stand at his trial his statements in response to Detective Hust’s questions were used in an effort to impeach his testimony in several respects.13 On appeal, the Arizona Supreme Court indicated its belief that because Detective Hust had failed to honor Mincey’s request for a lawyer, the statements would have been inadmissible as part of the prosecution’s case in chief. Miranda v. Arizona, supra. But, relying on Harris v. New York, 401 U. S. 222, and Oregon v. Hass, 420 U. S. 714, it held that since the trial court’s finding of voluntariness was not “clearfly] and manifestfly]” erroneous the statements were properly used for purposes of impeachment. 115 Ariz., at 480, 566 P. 2d, at 281. Statements made by a defendant in circumstances violating the strictures of Miranda v. Arizona, supra, are admissible for 12 The trial court made no findings of fact, nor did it make a specific finding of voluntariness, and the petitioner contends that admission of the statements therefore violated Jackson v. Denno. We agree with the Arizona Supreme Court, however, that the finding of voluntariness “appearfs] from the record with unmistakable clarity.” Sims v. Georgia, 385 U. S. 538, 544. The petitioner had originally moved to suppress his written answers to Bust’s questions on two grounds: that they had been elicited in violation of Miranda v. Arizona, 384 U. S. 436, and that they had been involuntary. During the hearing, the prosecution stipulated that the answers would be used only to impeach the petitioner if he took the witness stand. Any violation of Miranda thus became irrelevant. Oregon v. Hass, 420 U. S. 714; Harris v. New York, 401 U. S. 222. The testimony and the briefs and arguments of counsel were thereafter directed solely to whether the answers had been voluntarily given, and the court specifically ruled that they would be admissible for impeachment purposes only. The court thus necessarily held that Mincey’s responses to Bust’s interrogation were voluntary. 13 In light of our holding that Mincey’s hospital statements were not voluntarily given, it is unnecessary to reach his alternative contention that their use against him was impermissible because they were not sufficiently inconsistent with his trial testimony. 398 OCTOBER TERM, 1977 Opinion of the Court 437U.S. impeachment if their “trustworthiness ... satisfies legal standards.” Harris v. New York, supra, at 224; Oregon v. Hass, supra, at 722. But any criminal trial use against a defendant of his involuntary statement is a denial of due process of law “even though there is ample evidence aside from the confession to support the conviction.” Jackson n. Denno, supra, at 376; Haynes v. Washington, 373 U. S. 503, 518; Lynumn v. Illinois, 372 U. S. 528, 537; Stroble v. California, 343 U. S. 181, 190; see Chapman v. California, 386 U. S. 18, 23 and n. 8. If, therefore, Mincey’s statements to Detective Hust were not “ ‘the product of a rational intellect and a free will,’ ” Townsend v. Sain, 372 U. S. 293, 307, quoting Blackburn v. Alabama, 361 U. S. 199, 208, his conviction cannot stand. In making this critical determination, we are not bound by the Arizona Supreme Court’s holding that the statements were voluntary. Instead, this Court is under a duty to make an independent evaluation of the record. Davis v. North Carolina, 384 U. S. 737, 741-742; Haynes v. Washington, supra, at 515-516. It is hard to imagine a situation less conducive to the exercise of “a rational intellect and a free will” than Mincey’s. He had been seriously wounded just a few hours earlier, and had arrived at the hospital “depressed almost to the point of coma,” according to his attending physician. Although he had received some treatment, his condition at the time of Hust’s interrogation was still sufficiently serious that he was in the intensive care unit.14 He complained to Hust that the pain in his leg was “unbearable.” He was evidently confused and unable to think clearly about either the events of that afternoon or the circumstances of his interrogation, since some 14 A nurse testified at the suppression hearing that the device used to aid Mincey’s respiration was reserved for “more critical” patients. Moreover, Mincey apparently remained hospitalized for almost a month after the shooting. According to docket entries in the trial court his arraignment was postponed several times because he was still in the hospital; he was not arraigned until November 26, 1974. MINCEY v. ARIZONA 399 385 Opinion of the Court of his written answers were on their face not entirely coherent.15 Finally, while Mincey was being questioned he was lying on his back on a hospital bed, encumbered by tubes, needles, and breathing apparatus. He was, in short, “at the complete mercy” of Detective Hust, unable to escape or resist the thrust of Hust’s interrogation. Cf. Beecher v. Alabama, 389 U. S. 35, 38. In this debilitated and helpless condition, Mincey clearly expressed his wish not to be interrogated. As soon as Hust’s questions turned to the details of the afternoon’s events, Mincey wrote: “This is all I can say without a lawyer.” Hust nonetheless continued to question him, and a nurse who was present suggested it would be best if Mincey answered. Mincey gave unresponsive or uninformative answers to several more questions, and then said again that he did not want to talk without a lawyer. Hust ignored that request and another made immediately thereafter.1® Indeed, throughout the in- 15 For example, two of the answers written by Mincey were: “Do you me Did he give me some money (no)” and “Every body know Every body.” And Mincey apparently believed he was being questioned by several different policemen, not Hust alone; although it was Hust who told Mincey he had killed a policeman, later in the interrogation Mincey indicated he thought it was someone else. 16 In his reconstruction of the interrogation, see n. 11, supra, Hust stated that, after he asked Mincey some questions to try to identify one of the other victims, the following ensued: “HUST: . . . What do you remember that happened? “MINCEY: I remember somebody standing over me saying ‘move nigger, move.’ I was on the floor beside the bed. “HUST: Do you remember shooting anyone or firing a gun? “MINCEY: This is all I can say without a lawyer. “HUST: If you want a lawyer now, I cannot talk to you any longer, however, you don’t have to answer any questions if you don’t want to. Do you still want to talk to me? “MINCEY: (Shook his head in an affirmative manner.) “HUST: What else can you remember? “MINCEY: I’m going to have to put my head together. There are so 400 OCTOBER TERM, 1977 Opinion of the Court 437 U. S. terrogation Mincey vainly asked Hust to desist. Moreover, he complained several times that he was confused or unable to think clearly, or that he could answer more accurately many things that I don’t remember I. Like how did they get into the apartment? “HUST: How did who get into the apartment? “MINCEY: Police. “HUST: Did you sell some narcotics to the guy that was shot? “MINCEY: Do you mean, did he give me some money? “HUST: Yes. “MINCEY: No. “HUST: Did you give him a sample? “MINCEY: What do you call a sample? “HUST: A small amount of drug or narcotic to test? “MINCEY: I can’t say without a lawyer. “HUST: Did anyone say police or nares when they came into the apartment? “MINCEY: Let me get myself together first. You see, I’m not for sure everything happened so fast. I can’t answer at this time because I don’t think so, but I can’t say for sure. Some questions aren’t clear to me at the present time. “HUST: Did you shoot anyone? “MINCEY: I can’t say, I have to see a lawyer.” (Emphasis supplied.) While some of Mincey’s answers seem relatively responsive to the questions, it must be remembered that Hust added the questions at a later date, with the answers in front of him. See n. 11, supra. The reliability of Bust’s report is uncertain. For example, Hust claimed that imme-diately after Mincey first expressed a desire to remain silent, Hust said Mincey need not answer any questions but Mincey responded by indicating that he wanted to continue. There is no contemporaneous record supporting Bust’s statement that Mincey acted so inconsistently immAdi-ately after asserting his wish not to respond further, nor did the nurse who was present during the interrogation corroborate Bust. The Arizona Supreme Court apparently disbelieved Bust in this respect, since it stated that “after each indication from [Mincey] that he wanted to consult an attorney or that he wanted to stop answering questions, the police officer continued to question [him].” 115 Ariz., at 479, 566 P. 2d, at 280 (emphasis supplied). MINCEY v. ARIZONA 401 385 Opinion of the Court the next day.17 But despite Mincey’s entreaties to be let alone, Hust ceased the interrogation only during intervals when Mincey lost consciousness or received medical treatment, and after each such interruption returned relentlessly to his task. The statements at issue were thus the result of virtually continuous questioning of a seriously and painfully wounded man on the edge of consciousness. There were not present in this case some of the gross abuses that have led the Court in other cases to find confessions involuntary, such as beatings, see Brown v. Mississippi, 297 U. S. 278, or “truth serums,” see Townsend v. Sain, 372 IT. S. 293. But “the blood of the accused is not the only hallmark of an unconstitutional inquisition.” Blackburn v. Alabama, 361 •U. S., at 206. Determination of whether a statement is involuntary “requires more than a mere color-matching of cases.” Reck v. Pate, 367 U. S. 433, 442. It requires careful evaluation of all the circumstances of the interrogation.18 It is apparent from the record in this case that Mincey’s statements were not “the product of his free and rational choice.” Greenwald v. Wisconsin, 390 IT. S. 519, 521. To the contrary, the undisputed evidence makes clear that Mincey wanted not to answer Detective Hust. But Mincey was weakened by pain and shock, isolated from family, friends, and legal counsel, and barely conscious, and his will was simply 17 In addition to the statements quoted in n. 16, supra, Mincey wrote at various times during the interrogation: “There are a lot of things that aren’t clear,” “Thats why I have to have time to redo everything that happened in my mind,” and “I’m not sure as of now.” He also wrote: “If its possible to get a lawyer now. We can finish the talk. He could direct me in the right direction where as without a lawyer I might saw something thinking that it means something else.” And at another point he wrote: “Lets rap tomarrow, face to face. I can’t give facts. If something happins that I don’t know about.” Before the interrogation ended, Mincey made two further requests for a lawyer. 18 E. g., Boulden n. Holman, 394 U. S. 478, 480; Clewis v. Texas, 386 U. S. 707, 708; Haynes n. Washington, 373 U. S. 503, 513-514. 402 437 U.S. OCTOBER TERM, 1977 Marshall, J., concurring overbome. Due process of law requires that statements obtained as these were cannot be used in any way against a defendant at his trial. Ill For the foregoing reasons, the judgment of the Arizona Supreme Court is reversed, and the case is remanded for further proceedings not inconsistent with this opinion. It is so ordered. Mr. Justice Marshall, with whom Mr. Justice Brennan joins, concurring. I join the opinion of the Court, which holds that petitioner’s rights under the Fourth and Fourteenth Amendments have been violated. I write today to emphasize a point that is illustrated by the instant case, but that applies more generally to all cases in which we are asked to review Fourth Amendment issues arising out of state criminal convictions. It is far from clear that we would have granted certiorari solely to resolve the involuntary-statement issue in this case, for that could have been resolved on federal habeas corpus. With regard to the Fourth Amendment issue, however, we had little choice but to grant review, because our decision in Stone v. Powell, 428 U. S. 465 (1976), precludes federal habeas consideration of such issues. In Stone the Court held that, “where the State has provided an opportunity for full and fair litigation of a Fourth Amendment claim, a state prisoner may not be granted federal habeas corpus relief on the ground that evidence obtained in an unconstitutional search or seizure was introduced at his trial.” Id., at 494 (footnotes omitted). Because of this holding, petitioner would not have been able to present to a federal habeas court the Fourth Amendment claim that the Court today unanimously upholds. The additional responsibilities placed on this Court in the wake of Stone become apparent upon examination of deci MINCEY v. ARIZONA 403 385 Marshall, J., concurring sions of the Arizona Supreme Court on the Fourth Amendment issue presented here. The Arizona court created its “murder scene exception” in a 1971 case. State v. Sample, 107 Ariz. 407, 409—410,489 P. 2d 44,46—47. A year later, when the defendant in that case sought federal habeas corpus relief, the United States Court of Appeals for the Ninth Circuit ruled, as we do today, that the exception could not be upheld under the Fourth Amendment. Sample v. Eyman, 469 F. 2d 819, 821-822 (1972). When the Arizona Supreme Court next gave plenary consideration to the issue, prior to our decision in Stone, it apparently felt bound by the Ninth Circuit’s Sample decision, although it found the case before it to be distinguishable. State v. Duke, 110 Ariz. 320, 324, 518 P. 2d 570, 574 (1974)? When the Arizona Supreme Court rendered its decision in the instant case, however, it took a different approach. The decision, issued nearly a year after Stone, merely noted that the Ninth Circuit had “disagreed” with the Arizona court’s view of the validity of the murder-scene exception. 115 Ariz. 472, 482 n. 4, 566 P. 2d 273, 283 n. 4 (1977). It thus created an effective “conflict” for us to resolve. Cf. this Court’s Rule 19 (1) (b). If certiorari had not been granted, we would have left standing a decision of the State’s highest court on a question of federal constitutional law that had been resolved in a directly opposing way by the highest federal court having 1In its Mincey opinion, 115 Ariz. 472, 482, 566 P. 2d 273, 283 (1977), the Arizona Supreme Court indicated that one case other than Sample and Duke involved the murder-scene exception. State ex rel. Berger v. Superior Court, 110 Ariz. 281, 517 P. 2d 1277 (1974). The two-sentence opinion in the latter case, however, provides no explanation of the underlying facts and does not cite to either the Arizona court’s or the Ninth Circuit’s decision in Sample. There is thus no way to determine whether the situation in Berger was in any way comparable to those in Sample, Duke, and Mincey, nor any way to determine whether the Berger court simply disregarded the Ninth Circuit’s Sample decision or instead, as in Duke (decided just two weeks after Berger), viewed Sample as distinguishable. 404 OCTOBER TERM, 1977 Marshall, J., concurring 437U.S. special responsibility for the State. Regardless of which court’s view of the Constitution was the correct one, such nonuniformity on Fourth Amendment questions is obviously undesirable; it is as unfair to state prosecutors and judges— who must make difficult determinations regarding what evidence is subject to exclusion—as it is to state criminal defendants. Prior to Stone v. Powell, there would have been no need to grant certiorari in a case such as this, since the federal habeas remedy would have been available to the defendant. Indeed, prior to Stone petitioner here probably would not even have had to utilize federal habeas, since the Arizona courts were at that earlier time more inclined to follow the federal constitutional pronouncements of the Ninth Circuit, as discussed above. But Stone eliminated the habeas remedy with regard to Fourth Amendment violations, thus allowing state-court rulings to diverge from lower federal-court rulings on these issues and placing a correspondingly greater burden on this Court to ensure uniform federal law in the Fourth Amendment area. At the time of Stone my Brother Brennan wrote that “institutional constraints totally preclude any possibility that this Court can adequately oversee whether state courts have properly applied federal law.” 428 U. S., at 526 (dissenting opinion); see id., at 534. Because of these constraints, we will often be faced with a Hobson’s choice in cases of less than national significance that could formerly have been left to the lower federal courts: either to deny certiorari and thereby let stand divergent state and federal decisions with regard to Fourth Amendment rights; or to grant certiorari and thereby add to our calendar, which many believe is already overcrowded, cases that might better have been resolved elsewhere. In view of this problem and others,21 hope that the 2 The Stone holding has not eased the burden on the lower federal courts as much as the Stone majority might have hoped, since those courts have MINCEY v. NBXLONK 405 385 Opinion of Rehnquist, J. Court will at some point reconsider the wisdom of Stone n. Powell.3 Mr. Justice Rehnquist, concurring in part and dissenting in part. Petitioner was indicted for murder, assault, and three counts of narcotics offenses. He was convicted on all charges. On appeal, the Supreme Court of Arizona reversed all but the narcotics convictions. 115 Ariz. 472, 566 P. 2d 273 (1977). In his petition for certiorari, petitioner challenged the introduction of evidence material to his narcotics convictions that was seized during a lengthy warrantless search of his apartment. Petitioner also challenged on voluntariness grounds the introduction of various statements made to the police relating to the murder charge. We granted certiorari, 434 U. S. 902, and the Court today reverses the Supreme Court of Arizona on both issues. While I agree with the Court that the warrantless search was not justifiable on the grounds advanced by the Arizona Supreme Court, I dissent from the Court’s holding that Mincey’s statements were involuntary and thus inadmissible. I I join Part I of the Court’s opinion. As the Supreme Court of Arizona recognized, the four-day warrantless search of petitioner’s apartment did not, on the facts developed at trial, “fit within [any] usual ‘exigent circumstances’ exception.” 115 Ariz., at 482, 566 P. 2d, at 283. Instead, the State of had to struggle over what this Court meant by “an opportunity for full and fair litigation of a Fourth Amendment claim,” 428 U. S., at 494. See, e. g., Gates v. Henderson, 568 F. 2d 830 (CA2 1977); United States ex rel. Petillo v. New Jersey, 562 F. 2d 903 (CA3 1977); O’Berry v. Wainwright, 546 F. 2d 1204 (CA5 1977). 3 A bill currently pending in the Congress would have the effect of overruling Stone v. Powell. S. 1314, 95th Cong., 1st Sess. (1977); see 123 Cong. Rec. 11347-11353 (1977). 406 OCTOBER TERM, 1977 Opinion of Rehnquist, J. 437 U.S. Arizona asks us to adopt a separate “murder scene” exception to the warrant requirement and the Court, for the reasons stated in its opinion, correctly rejects this invitation. I write separately on this issue only to emphasize that the question of what, if any, evidence was seized under established Fourth Amendment standards is left open for the Arizona courts to resolve on remand. Ante, at 395 n. 9. Much of the evidence introduced by the State at trial was apparently removed from the apartment the same day as the shooting. App. 40. And the State’s brief suggests that some evidence— for example, blood on the floor—required immediate examination. Brief for Respondent 70-71. The question of what evidence would have been “lost, destroyed, or removed” if a warrant had been obtained, ante, at 394, otherwise required an immediate search, or was in plain view should be considered on remand by the Arizona courts. In considering whether exigencies required the search for or seizure of particular evidence, the previous events within the apartment cannot be ignored. I agree with the Court that the police’s entry to arrest Mincey, followed by the shooting and the search for victims, did not justify the later four-day search of the apartment. Ante, at 391-392. But the constitutionality of a particular search is a question of reasonableness and depends on "a balance between the public interest and the individual’s right to personal security free from arbitrary interference by law officers.” United States v. Brig-noni-Ponce, 422 U. S. 873, 878 (1975). See Terry v. Ohio, 392 U. S. 1, 19 (1968). In Pennsylvania v. Mimms, 434 U. S. 106 (1977), we held that once a motor vehicle had been lawfully detained for a traffic violation, police officers could constitutionally order the driver out of the vehicle. In so holding, we emphasized that the challenged intrusion was “occasioned not by the initial stop of the vehicle, which was admittedly justified, but by the order to get out of the car. We think this additional intrusion can only be described as de minimis’’ MINCEY v. ARIZONA 407 385 Opinion of Rehnquist, J. Id., at 111. Similarly, in the instant case, the prior intrusions occasioned by the shooting and the police’s response thereto may legitimize a search under some exigencies that in tamer circumstances might not permit a search. II The Court in Part II of its opinion advises the Arizona courts on the admissibility of certain statements made by Mincey that are relevant only to the murder charge. Because Mincey’s murder conviction was reversed by the Arizona Supreme Court, and it is not certain that there will be a retrial, I would not reach this issue. Since the Court addresses the issue, however, I must register my disagreement with its conclusion. Before trial, Mincey moved to suppress as involuntary certain statements that he had made while confined in an intensive care unit some hours after the shooting. As the Court acknowledges, the trial court found “‘with unmistakable clarity’ ” that the statements were voluntary, ante, at 397 n. 12, and the Supreme Court of Arizona unanimously affirmed. 115 Ariz., at 479^80, 566 P. 2d, at 280-281. This Court now disagrees and holds that “Mincey’s statements were not ‘the product of his free and rational choice’ ” and therefore “cannot be used in any way against [him] at his trial.” Ante, at 401,402. Because I believe that the Court both has failed to accord the state-court finding the deference that the Court has always found such findings due and also misapplied our past precedents, I dissent. As the Court notes, ante, at 398, past cases of this Court hold that a state-court finding as to voluntariness which is “not fairly supported by the record cannot be conclusive of federal rights.” Townsend v. Sain, 372 U. S. 293, 316 (1963) (emphasis added). Instead, these cases require the Court to “make an independent determination on the undisputed facts.” Stroble v. California, 343 U. S. 181, 190 (1952) (emphasis added); 408 OCTOBER TERM, 1977 Opinion of Rehnquist, J. 437U.S. Malinski v. New York, 324 U. S. 401, 404 (1945). It is well established that, “for purposes of review in this Court, the determination of the trial judge or of the jury will ordinarily be taken to resolve evidentiary conflicts and may be entitled to some weight even with respect to the ultimate conclusion on the crucial issue of voluntariness.” Haynes v. Washington, 373 U. S. 503, 515 (1963). See Lisenbav. California, 314 U. S. 219, 238 (1941); Blackburn v. Alabama, 361 U. S. 199, 205, and n. 5 (1960). Such deference, particularly on the resolution of evidentiary conflict«, “is particularly apposite because the trial judge and jury are closest to the trial scene and thus afforded the best opportunity to evaluate contradictory testimony.” Haynes, supra, at 516. The Court in this case, however, ignores entirely some evidence of voluntariness and distinguishes away yet other testimony. There can be no discounting that Mincey was seriously wounded and laden down with medical equipment. Mincey was certainly not able to move about and, because of the breathing tube in his mouth, had to answer Detective Hust’s questions on paper. But the trial court was certainly not required to find, as the Court would imply, that Mincey was “a seriously and painfully wounded man on the edge of consciousness.” Ante, at 401. Nor is it accurate to conclude that Detective Hust “ceased the interrogation only during intervals when Mincey lost consciousness or received medical treatment, and after each such interruption returned relentlessly to his task.” Ibid. As the Arizona Supreme Court observed in affirming the trial court’s finding of voluntariness, Mincey’s nurse “testified that she had not given [Mincey] any medication and that [he] was alert and able to understand the officer’s questions. ... She said that [Mincey] was in moderate pain but was very cooperative with everyone. The interrogating officer also testified that [Mincey] did not appear to be under the influence of drugs and that MINCEY v. ARIZONA 409 385 Opinion of Rehnquist, J. [his] answers were generally responsive to the questions.” 115 Ariz., at 480, 566 P. 2d, at 281. See App. 50-51 (testimony of Detective Hust), 63 and 66 (testimony of Nurse Graham).1 The uncontradicted testimony of Detective Hust also reveals a questioning that was far from “relentless.” While the interviews took place over a three-hour time span, the interviews were not “very long; probably not more than an hour total for everything.” Id., at 59. Hust would leave the room whenever Mincey received medical treatment “or if it looked like he was getting a little bit exhausted.” Ibid. According to Detective Hust, Mincey never “los[t] consciousness at any time.” Id., at 58. As the Court openly concedes, there were in this case none of the “gross abuses that have led the Court in other cases to find confessions involuntary, such as beatings ... or Truth serums.’ ” Ante, at 40b Neither is this a case, however, where the defendant’s will was “simply overborne” by “mental coercion.” Cf. Blackburn v. Alabama, supra, at 206; Davis v. North Carolina, 384 U. S. 737, 741 (1966) ; Greenwald v. Wisconsin, 390 U. S. 519, 521 (1968). As the Supreme Court of Arizona observed, it was the testimony of both Detective Hust and Nurse Graham “that neither mental or physical force nor abuse was used on [Mincey] .... Nor were any promises made.” 115 Ariz., at 480, 566 P. 2d, at 281. See App. 58-59 (testimony of Detective Hust) and 63 (testimony of Nurse Graham). According to Mincey’s own testimony, he wanted 1 The Supreme Court of Arizona also emphasized “the fact that [Mincey] was able to write his answers in a legible and fairly sensible fashion.” 115 Ariz., at 480 n. 3, 566 P. 2d, at 281 n. 3. The Court concedes that “Mincey’s answers seem relatively responsive to the questions,” ante, at 400 n. 16, but chooses to ignore this evidence on the ground that the “reliability of Hust’s report is uncertain.” Ibid. Despite the contrary impression given by the Court, ibid., the Arizona Supreme Court’s opinion casts no doubt on the testimony or report of Detective Hust. The Court is thus left solely with its own conclusion as to the reliability of various witnesses based on a re-examination of the record on appeal. 410 OCTOBER TERM, 1977 Opinion of Rehnquist, J. 437 U. S. to help Hust “the best I could” and tried to answer each question “to the best of my recollection at the time that this was going on.” Id., at 86. Mincey did not claim that he felt compelled by Detective Hust to answer the questions propounded.2 Cf. Greenwald, supra, at 521. By all of these standards enunciated in our previous cases, I think the Court today goes too far in substituting its own judgment for the judgment of a trial court and the highest court of a State, both of which decided these disputed issues differently than does this Court, and both of which were a good deal closer to the factual occurrences than is this Court. Admittedly we may not abdicate our duty to decide questions of constitutional law under the guise of wholly remitting to state courts the function of factfinding which is a necessary ingredient of the process of constitutional decision. But the authorities previously cited likewise counsel us against going to the other extreme, and attempting to extract from a cold record bits and pieces of evidence which we then treat as the “facts” of the case. I believe that the trial court was entitled to conclude that, notwithstanding Mincey’s medical condition, his statements in the intensive care unit were admissible. The fact that the same court might have been equally entitled to reach the opposite conclusion does not justify this Court’s adopting the opposite conclusion. I therefore dissent from Part II of the Court’s opinion. 2 While Mincey asked at several points to see a lawyer, he also expressed his willingness to continue talking to Detective Hust even without a lawyer. See ante, at 399-400, n. 16. As the Court notes, since Mincey’s statements were not used as part of the prosecution’s case in chief but only in impeachment, any violation of Miranda n. Arizona, 384 U. S. 436 (1966), was irrelevant.. See Harris n. New York, 401 U., S. 222 (1971); Oregon v. Hass, 420 U. S. 714 (1975). AMERICAN BROADCASTING COS. v. WRITERS GUILD 411 Syllabus AMERICAN BROADCASTING COMPANIES, INC., et al. v. WRITERS GUILD OF AMERICA, WEST, INC., ET AL. CERTIORARI TO THE UNITED STATES COURT OF APPEALS FOR THE SECOND CIRCUIT No. 76-1121. Argued December 5, 1977—Reargued March 20, 1978— Decided June 21, 1978* Respondent union, which represents persons hired to perform writing functions for motion picture and television films (hereinafter respondent), had collective-bargaining contracts with a producers association (petitioner in No. 76-1153) and three television networks (petitioners in No. 76-1121). Among respondent’s members are a large number of persons (so-called “hyphenates”) who are engaged by petitioners primarily to perform executive and supervisory functions. Though the hyphenates, who include various categories of producers, directors, and story editors, have minor writing tasks, these are not covered in the collective-bargaining contracts; only if the hyphenates are employed to perform additional writing services are the rates therefor governed by those contracts. In connection with their regular, primary duties many of the hyphenates are represented by unions other than respondent. In anticipation of an economic strike upon expiration of its contracts with petitioners, respondent distributed strike rules to its members, including the hyphenates (to whom the rules were made expressly applicable). The rules included a prohibition against crossing a picket line established by respondent at any entrance of a struck premise. After the strike began, petitioners informed the hyphenates that they were expected to continue their regular supervisory functions, though they would not be asked to perform writing duties covered by the union contract. Thereafter respondent notified a large number of the hyphenates who had returned to work that they had violated one or more of the strike rules, including in many instances the ban on crossing a picket fine. After ensuing disciplinary proceedings (at which there was no proof that hyphenates had performed any work covered by the recently expired *Together with No. 76-1153, Association of Motion Picture & Television Producers, Inc. v. Writers Guild of America, West, Inc., et al.; and No. 76-1162, National Labor Relations Board v. Writers Guild of America, West, Inc., et al., also on certiorari to the same court. 412 OCTOBER TERM, 1977 Syllabus 437 U. S. contracts) respondent imposed various penalties on the hyphenates. Meanwhile the association and network petitioners filed charges against respondent for allegedly violating §8 (b)(1)(B) of the National Labor Relations Act, which makes it an unfair labor practice for a labor organization to restrain or coerce an employer in the selection of his representatives for the purposes of collective bargaining or the adjustment of grievances. After extensive hearings, the Administrative Law Judge made findings that the hyphenates’ regular supervisory duties included the performance of grievance adjustment; that the employer insisted that hyphenates return to work, but only to perform supervisory, not rank-and-file, duties; and that the hyphenates who reported did only supervisory work and had the authority to adjust grievances, which they did when the occasion arose. He found that §8 (b)(1)(B) had been violated because, by keeping hyphenates from work, the union had deprived the employer of fully effective §8 (b)(1)(B) representatives. The National Labor Relations Board (NLRB) adopted these findings and conclusions, found that the union’s disciplinary action was an unfair labor practice under that provision, and issued a remedial order against respondent. The Court of Appeals denied enforcement. Held: Respondent’s actions against the hyphenates violated §8 (b)(1)(B). Pp. 429-438. (a) In ruling upon a § 8 (b)(1) (B) charge growing out of union discipline of a supervisory member who elects to work during a strike, the NLRB must inquire whether the sanction may adversely affect the supervisor’s performance of his collective-bargaining or grievance-adjustment tasks and thereby coerce or restrain the employer contrary to that provision. See Florida Power & Light Co. v. Electrical Workers, 417 U. S. 790. Pp. 429-431. (b) The NLRB’s findings were based on substantial evidence that the hyphenates were coerced or restrained from reporting to work; that the employer was thereby deprived of the opportunity to choose particular supervisors as his collective-bargaining or his grievance-adjustment representatives during the strike; and that as to the hyphenates who reported to work there was adequate basis for concluding that the discipline would adversely affect the performance of their grievance-adjustment duties either during or after the strike. Moreover, since as the evidence showed, the union’s policy was not to permit a member to resign during a strike and for six months thereafter, the employer could not free a supervisor from further threats of union discipline by requiring him to leave the union. Pp. 431-437. 547 F. 2d 159, reversed. AMERICAN BROADCASTING COS. v. WRITERS GUILD 413 411 Opinion of the Court White, J., delivered the opinion of the Court, in which Burger, C. J., and Blackmun, Powell, and Rehnquist, JJ., joined. Stewart, J., filed a dissenting opinion, in which Brennan, Marshall, and Stevens, JJ., joined, post, p. 438. Norton J. Come reargued the cause for petitioner in No. 76-1162. With him on the briefs were Solicitor General McCree, John S. Irving, Carl L. Taylor, and John G. Elligers. Harry J. Keaton reargued the cause and filed a brief for petitioner in No. 76-1153. Charles G. Bakcdy reargued the cause for petitioners in No. 76-1121. With him on the briefs was Gordon E. Krischer. Julius Reich reargued the cause for respondent Writers Guild of America, West, Inc., in all cases. With him on the briefs was Paul P. Selvin. Laurence Gold reargued the cause for the American Federation of Labor and Congress of Industrial Organizations as amicus curiae urging affirmance. With him on the brief was J. Albert Woll. Mr. Justice White delivered the opinion of the Court. The issue in this litigation is whether a labor union commits an unfair labor practice when it disciplines a member who is a supervisory employee for crossing the union’s picket line during a strike and performing his regular supervisory duties, which include the adjustment of grievances. I Respondent Writers Guild of America, West, Inc. (hereafter respondent), represents persons hired to perform writing functions for employers engaged in the production of motion pictures and television films, and in 1973 had contracts with certain petitioners that were about to expire. Petitioner in No. 76-1153 is the Association of Motion Picture and Television Producers, Inc., whose members are engaged in the production of motion pictures and television films. Petitioner 414 OCTOBER TERM, 1977 Opinion of the Court 437U.S. represents its members in the negotiation and administration of collective-bargaining contracts. The three television networks, NBC, CBS, and ABC, petitioners in No. 76-1121, are also engaged in the production of television films and negotiate and administer collective-bargaining contracts. In March 1973, respondent engaged in a strike against both of these groups of petitioners, picketed the various premises, and issued strike rules that it enforced against its own members. It is this action which gave rise to this case. Among respondent’s members are a substantial number of persons who were engaged by petitioners primarily to perform executive and supervisory functions including the selection and direction of writers and including certain limited writing duties. These persons are referred to as “hyphenates” and include various categories of producers, directors, and story editors.1 Although the primary function of hyphenates is not to write, they do perform minor writing tasks (referred to in the contract as “A to H” functions) that are an integral part of their primary duties and that expressly are not covered by the contracts between petitioners and respondent.2 1 Executive producers, with the help of producers and associate producers, have the primary responsibility for the production of films for motion pictures or for television. The responsibility begins with the idea or concept for the film or the series and carries through to the post-production stages after filming. Directors are in personal charge of the principal photography of the film. They are responsible for the employment of crew and actors and effectively direct such employees. Story editors, story consultants, script consultants, executive story editors, and executive story consultants principally assist the producer in the highly important function of dealing with scripts and writers. They have individual judgment, initiative, and responsibility, and their tasks are clearly supervisory. Approximately 80 hyphenate members of respondent were principally employed as producers of one kind or another, approximately 15 were directors, and another 15 were in the story editor category. 2 The finding of the Administrative Law Judge in this regard was: “The important point is that when these executives and supervisors perform those functions excluded from the Respondent’s bargaining agree- AMERICAN BROADCASTING COS. v. WRITERS GUILD 415 411 Opinion of the Court Only in the event hyphenates are assigned or employed by petitioners to perform additional writing services are the rates for such services governed by the collective-bargaining contracts with respondent. In connection with the performance of their regular, primary duties, which, with the limited exception noted, do not include writing services, many, but not all, hyphenates are represented by labor organizations other than respondent. Some of the contracts between these other organizations and petitioners contained no-strike clauses when the events involved herein occurred. Certain hyphenates were pressured by these other labor organizations to honor these no-strike pledges by reporting to work. Respondent, meanwhile, was preparing its own kinds of pressure to keep the hyphenates from working. In preparation for the strike, respondent issued and distributed to its members, including the hyphenates, some 31 strike rules. The rules, among other things, forbade any act prejudicial to the welfare of respondent such as conduct tending to defeat a strike or to weaken its effectiveness (Rule 1); prohibited all members “from crossing a picket line which is established by the Guild at any entrance” of a struck premises (Rule 12); forbade the entry of any struck premises for certain purposes and required notice to respondent when entry was made for other purposes (Rule 13);3 and obliged members to accept picket duty when assigned by respondent (Rule 28). Another ments they thereby perform functions which the parties have acknowledged do not constitute work reserved to Respondent’s non-hyphenate members under the agreements, but rather are accepted as a normal part of the duties and responsibilities of the executives and supervisors (as hereinabove discussed) employed by the employers involved.” (Footnote omitted.) App. to Pet. for Cert, in No. 76-1162, p. 35a. The contract provided that performance of any “A to H” writing “shall not constitute such person a writer hereunder.” Id., at 33a. 3 Rule 13 provided: “Members are prohibited from entering the premises of any struck producer for the purpose of discussion of the sale of material or contract of 416 OCTOBER TERM, 1977 Opinion of the Court 437U.S. rule (Rule 30), rescinded midway in the strike, provided that no member could work with any individual, including the writer-executive, who had violated union strike rules.4 The strike rules’ applicability to hyphenates was made clear in Rule 24: “All members, regardless of the capacity in which they are working, are bound by all strike rules and regulations in the same manner and to the same extent as members who confine their efforts to writing.” The rules were widely publicized, and respondent repeatedly emphasized, orally and in writing, that it would enforce the rules against hyphenates. Nor could a hyphenate escape those strictures by resigning, for it was respondent’s policy, once the strike was under way, employment, regardless of the time it is to take effect. Members are also prohibited from entering the premises of any struck producer for the purpose of viewing any film. . . . [S]hould a member find it necessary to visit the premises of a struck producer for any reason apart from the foregoing he should inform the Guild in advance of the nature of such prospective visit.” Id., at 36a-37a. 4Rule 30 provided: “No member shall work with any individual, including a writer-executive who has been suspended from Guild membership by reason of his violation of strike rules, or has been found by the Council to have violated strike rules, in the event no disciplinary action was instituted against such person.” Id., at 38a. After the issuance of the initial complaint in this case, Rule 30 was rescinded by respondent in a letter to all of its members, which stated, among other things, that “because the old rule could be misconstrued to mean that the Guild was maintaining an improper sanction, a matter of anathema to this Guild, the Board of Directors rescinded old Rule 30 . .. .” The assessment of the Administrative Law Judge was: “In particular, by threatening to blacklist in perpetuity such hyphenates who worked during the strike, the rules threatened to drive these hyphenates out of the industry. Though the mandatory effect of the rule was rescinded . . . there are other indications that Respondent’s actions encourage a voluntary blacklist. . . . [T]he fact is that Respondent did suggest it, and it is now impossible to disentangle the consequences flowing from its actions.” Id., at 69a-70a. AMERICAN BROADCASTING COS. v. WRITERS GUILD 417 411 Opinion of the Court not to permit withdrawal from the union, then or for six months following the completion of negotiations. Petitioners, however, informed the hyphenates that petitioners’ operations were continuing and that the hyphenates were expected to report for work and perform their regular supervisory functions. Petitioners were careful to assure that hyphenates would not be requested to perform writing duties covered by the union contract. Some hyphenates went to work, informing their employers, as respondent knew, that they would perform only their primary duties as producer, director, or story editor. Others refrained from reporting for work. Between April 6 and November 8, 1973, respondent notified more than 30 hyphenates who returned to work that they had been charged with violating one or more of the strike rules. Most often, the charges related to Rules 1, 12, and 13.5 Various disciplinary trials ensued. In these proceedings, the evidence was that the hyphenates who returned to duty performed only the normal functions of the supervisory positions for which they were employed. There was no proof that hyphenates performed any work covered by the recently terminated contracts between petitioners and respondent. As the Administrative Law Judge observed, respondent “for the most part professed little or no interest in what kind of work was done during the strike” 5 The Administrative Law Judge found that a typical notice of charges against a hyphenate contained the following: “Specifically,, you are charged with: (1) having crossed the Guild’s picket lines . . . during the months of March, April, May and June 1973, without having informed the Guild in advance of the nature of your business with said company and without having obtained a Guild pass to enter said premises; (2) having during the months of March, April, May and June 1973, rendered services for ... a company against whom the Guild was at such times on strike; and (3) refusing to perform picket duties during the strike after having been requested to do so by representatives of the Guild.” Id., at 45a. (Footnote omitted.) 418 OCTOBER TERM, 1977 Opinion of the Court 437U.S. by the hyphenates who chose to work.6 Between June 25 and September 28, 1973, various penalties were imposed by respondent as the result of these disciplinary proceedings. The penalties included expulsions, suspensions, and quite substantial fines.7 Meanwhile, the Association and network petitioners had filed unfair labor practice charges, and the General Counsel of the National Labor Relations Board had issued complaints against respondent charging violations of §8 (b)(1)(B) of the National Labor Relations Act, 61 Stat. 141, 29 U. S. C. § 158 (b)(1)(B), which provides that “[i]t shall be an unfair labor practice for a labor organization ... to restrain or coerce ... an employer in the selection of his representatives for the purposes of collective bargaining or the adjustment of grievances.” Extensive hearings followed, the Administra- 6 Id., at 43a-44a. Respondent, through counsel, took the position at the disciplinary hearings that the hyphenates charged were subject to discipline simply for crossing respondent’s picket line, whether or not they crossed for the purpose of performing bargaining services for a struck employer. Respondent held that charges would properly lie even against hyphenates who had given assurances not to perform any writing services for a struck employer. 7 The Administrative Law Judge noted the penalties against 10 of the hyphenates charged and tried: “Two were expelled from membership and fined $50,000 each; one was expelled from membership and fined $10,000; one was suspended from membership for 2 years and fined $10,000; one was suspended for 2 years and fined $7,500; one was suspended for 3 years and fined $5,000; one was expelled from membership and fined $2,000; one was expelled and fined $100; and one was suspended for 2 years and fined $100. These penalties received wide publicity in the local press and trade papers. The appeals of nine of these men has [sic] been voted upon by Respondent’s membership at a special meeting and the penalties were drastically reduced. Apparently all remaining actions with respect to discipline of hyphenate-members for working during the strike are now being held in abeyance pending resolution of these cases.” Id., at 46a. AMERICAN BROADCASTING COS. v. WRITERS GUILD 419 411 Opinion of the Court tive Law Judge ultimately recommending that the charges be sustained and making findings and conclusions that were adopted by the National Labor Relations Board. These findings included an analysis of the primary functions for which the hyphenates were employed. It was concluded that all of the producers, directors, and story editors involved were employed to perform supervisory functions and were supervisors within the meaning of § 2 (11) of the Act, 29 U. S. C. § 152 (11). It was also found that the hyphenates in each of these categories regularly had the authority and the task of adjusting grievances.8 “It is clear, as has 8 The Administrative Law Judge found: “The producer has substantial responsibility and authority in adjusting grievances between directors and craft employees, directors and actors and actresses, between two or more actors or actresses, and in other similar situations. Producers also have responsibility and authority to adjust grievances involving writers, as in the case of disputes between writers and story editors.” Id., at 26a. Executive producers supervise one or more producers, and associate producers assist the producer. “Without distinguishing among them in detail, it is clear on this record that persons occupying these positions in the motion picture or television industries have the authority to hire, terminate, and responsibly direct other employees, and to adjust employee grievances, or to effectively recommend such action, and are thus supervisors within the meaning of Section 2 (11) of the Act.” Id., at 27a. With respect to directors, the Administrative Law Judge determined that they “hire or effectively recommend the employment of crew and actors, effectively direct such employees, and may discharge or effectively recommend the discharge of employees. They have authority to and do adjust grievances of such employees. It is found that persons performing the functions of director in the television and motion picture industries are supervisors and adjust grievances of employees within the meaning of the Act.” Id., at 28a. Story editors supervise writers in the development of ideas and the preparation of scripts. They interview and recommend the hiring of new 420 OCTOBER TERM, 1977 Opinion of the Court 437U.S. been found, that the normal performance of the hyphenates’ primary functions involves the adjustment of employee grievances, and, in the case of producers on distant location, to engage in collective bargaining with labor organizations.” 9 Furthermore, the record indicated that “during the strike, where the situation arose, the hyphenates dealt with grievances of employees who worked during the strike, or, in any event, were available to deal with such matters in their normal capacities when and if such grievances arose.”10 It was also found that the hyphenates who reported for duty during the strike performed only their primary functions and did not engage in writing or do any work that had been covered by respondent’s collective-bargaining contract. Significantly, none of the hyphenates was charged with violating the strike rule forbidding the performance of writing functions for a struck employer. During the disciplinary hearings, respondent was “not concerned with what work the hyphenates did when working during the strike,”11 although it would have been quite easy to determine these facts from testimony of union writers about what work was found completed upon their return. writers, and advise the producer concerning writers who should not be retained. “On a television series, the story editor may participate with the producer in the initial determination of any dispute over screen credits. He also may serve as a buffer between management and the writer, as in ameliorating a writer’s distress over material that has been rewritten. . . . “On the basis of the entire record, it is found that those persons in the television and motion picture industries performing the functions of story editor, story consultant, script consultant, executive story editors, and executive story consultants are supervisors and adjust grievances of employees within the meaning of the Act.” Id., at 29a-30a. 9 Id., at 57a. 10 Id., at 60a. 11 Id., at 59a. AMERICAN BROADCASTING COS. v. WRITERS GUILD 421 411 Opinion of the Court The ultimate factual conclusions of the Administrative Law Judge were that the hyphenates were supervisors “selected by their employers to adjust grievances”;12 that in issuing strike rules and engaging in other conduct designed to compel the hyphenates to refrain from working respondent had “restrained and coerced the hyphenates from performing managerial and supervisory services for their employers during the strike, including the adjustment of employee grievances and participation in collective bargaining,” and had thus “coerced and restrained those employers in the selection of representatives for collective bargaining and the adjustment of grievances within the meaning of Section 8 (b)(1)(B)”;13 and that by charging, trying, and disciplining the hyphenates who chose to work and who, the Administrative Law Judge found, “performed managerial and supervisory functions including the adjustment of grievances on collective bargaining as required, and did not perform rank and file work,” respondent “further coerced and restrained the employers” within the meaning of § 8 (b)(1)(B).14 In arriving at these conclusions, the Administrative Law Judge rejected the claim that Florida Power & Light Co. v. Electrical Workers, 417 U. S. 790 (1974) (FP&L), required a contrary result, saying that respondent’s conduct “violated the plain meaning of the statute without the necessity of resort to statutory exegesis.”15 On exceptions and supporting briefs, a majority of a three-member panel of the Board, except in one respect,16 adopted as its own the rulings, findings, and conclusions of the Admin 12 Id., at 62a. 13 Ibid. 14 Id., at 63a. 15 Ibid. 16 The Board held that there had been a violation with respect to certain hyphenates in addition to those in the categories of producer, director, and story editor. 422 OCTOBER TERM, 1977 Opinion of the Court 437U.S. istrative Law Judge. The Board also reasoned that FP&L, which involved supervisors who performed bargaining-unit work, did not extend to cases where union discipline was imposed upon supervisors who performed only their ordinary supervisorial functions (including the adjustment of grievances). The Board relied upon two of its cases decided subsequent to FP&L: Chicago Typographical Union No. 16 (Hammond Publishers, Inc.), 216 N. L. R. B. 903 (1975); New York Typographical Union No. 6, International Typographical Union, AFL-CIO (Daily Racing Form, a subsidiary of Triangle Publishers, Inc.), 216 N. L. R. B. 896 (1975). On application to review by the networks and the Board’s application to enforce, a divided panel of the Court of Appeals for the Second Circuit denied enforcement in a brief per curiam opinion indicating that, like the dissenting member of the Board, it considered FP&L, supra, to bar the results reached by the Board in this case. 547 F. 2d 159 (1976). We granted the petitions for certiorari of the Board as well as of the Association and the networks because of an apparent conflict between the decision below and decisions in other Courts of Appeals and because of the recurring nature of the issue.17 430 U. S. 982 (1977). II As the Court has set out in greater detail in its comprehensive review of § 8 (b)(1) (B) in FP&L, the prohibition against restraining or coercing an employer in the selection of his bargaining representative was, until 1968, applied primarily to pressures exerted by the union directly upon the employer 17 In Chicago Typographical Union No. 16 n. NLRB, 176 U. 8. App. D. C. 240, 539 F. 2d 242 (1976), the Court of Appeals for the District of Columbia Circuit enforced the Board’s order in Hammond Publishers, relied on by the Board in this case. In Wisconsin River Valley Dist. Council v. NLRB, 532 F. 2d 47 (1976), the Court of Appeals for the Seventh Circuit also took a position seemingly at odds with the judgment under review here. The issue is also a recurring one before the Board. AMERICAN BROADCASTING COS. v. WRITERS GUILD 423 411 Opinion of the Court to force him into a multiemployer bargaining unit or otherwise to dictate or control the choice of his representative for the purpose of collective bargaining or adjusting grievances in the course of administering an existing contract. In San Francisco-Oakland Mailers’ Union No. 18, International Typographical Union {Northwest Publications, Inc.), 172 N. L. R. B. 2173 (1968), however, the Board applied the section to prohibit union discipline of one of its member-supervisors for the manner in which he had performed his supervisory task of grievance adjustment. Although the union “sought the substitution of attitudes rather than persons, and may have exerted its pressures upon the [employer] by indirect rather than direct means,” the ultimate fact was that the pressure interfered with the employer’s control over his representative. “Realistically, the Employer would have to replace its foremen or face de facto nonrepresentation by them.” Oakland Mailers, supra, at 2173. The application of the section to indirect coercion of employers through pressure applied to supervisory personnel continued to evolve until the FP&L and Illinois Bell18 cases reached the Court of Appeals for the District of Columbia Circuit and then this Court. In each of those cases, the union disciplined supervisor-members who had performed rank-and-file work behind a union picket line during a strike. In a companion case to Illinois Bell,19 upon which Illinois Bell explicitly relied,20 the Board found an infraction of § 8 (b) 18 IBEW, Local 134 v. NLRB, 159 U. S. App. D. C. 242, 487 F. 2d 1113, rev’d on rehearing en banc, 159 U. S. App. D. C. 272, 487 F. 2d 1143 (1973), refusing to enforce IBEW, Local 134, 192 N. L. R. B. 85 (1971) (Illinois Bell), and IBEW Systems Council U~4, 193 N. L. R. B. 30 (1971) (FP&L). 19 Local Union No. 2150, IBEW, and Wisconsin Electric Power Co., 192 N. L. R. B. 77 (1971). 20 “We find no discernible difference between the two cases, and for the reasons set forth in that case, we find that, in the instant case, the Union violated Section 8 (b) (1) (B) . . . .” Illinois Bell, 192 N. L. R. B., at 86. 424 OCTOBER TERM, 1977 Opinion of the Court 437 U. S. (1)(B), broadly construing its purpose “to assure to the employer that its selected collective-bargaining representatives will be completely faithful to its desires” and holding that this could not be achieved “if the union has an effective method, union disciplinary action, by which it can pressure such representatives to deviate from the interests of the employer.” 21 In like fashion, in FP&L, the Board held that fining supervisors for doing rank-and-file work during a work stoppage “struck at the loyalty an employer should be able to expect from its representatives for the adjustment of grievances and therefore restrained and coerced employers in their selection of such representatives.”22 The Court of Appeals overturned both decisions of the Board, holding that although the section could be properly applied to union efforts to discipline supervisors for their performance as collective-bargaining or grievance-adjustment representatives, it could not reasonably be applied to prohibit union discipline of supervisors crossing picket lines to perform bargaining-unit work: “When a supervisor forsakes his supervisory role to do rank-and-file work ordinarily the domain of nonsupervisory employees, he is no longer acting as a management representative and no longer merits any immunity from discipline.” 159 U. S. App. D. C., at 286, 487 F. 2d, at 1157. This Court affirmed the judgment of the Court of Appeals: “The conclusion is thus inescapable that a union’s discipline of one of its members who is a supervisory employee can constitute a violation of §8 (b)(1)(B) only when that discipline may adversely affect the supervisor’s conduct in performing the duties of, and acting in his capacity as, grievance adjuster or collective bargainer on behalf of the employer.” 417 U. S., at 804-805. 21 Id., at 78. 22 193 N. L. R. B., at 31. AMERICAN BROADCASTING COS. v. WRITERS GUILD 425 411 Opinion of the Court The Court thus rejected the claim that “even if the effect of [union] discipline did not carry over to the performance of the supervisor’s grievance adjustment or collective bargaining functions,” it was enough to show that the result would be “to deprive the employer of the full allegiance of, and control over, a representative he has selected for grievance adjustment or collective bargaining purposes.” Id., at 807. Assuming without deciding that the Board’s decision in Oakland Mailers fell within the outer reaches of § 8 (b)(1)(B), the Court concluded that the Illinois Bell and FP&L decisions did not, because it was “certain that these supervisors were not engaged in collective bargaining or grievance adjustment, or in any activities related thereto, when they crossed union picket lines during an economic strike to engage in rank-and-file struck work.” 417 U. S., at 805. Subsequent to FP&L, in applying §8 (b)(1)(B) to cases involving union discipline of supervisor-members, the Board directed its attention, as it understood FP&L to require, to the question whether the discipline may adversely affect the supervisor’s conduct in performing his grievance-adjustment or collective-bargaining duties on behalf of the employer. In Hammond Publishers, supra, and Triangle Publishers, supra, the Board held that it was an unfair practice under § 8 (b)(1)(B) for a union to discipline a supervisor-member whose regular duties included the adjustment of grievances for crossing a picket line to perform his regular functions during a strike. See also Wisconsin River Valley Dist. Council (Skippy Enterprises, Inc.), 218 N. L. R. B. 1063 (1975). These cases rested on the Board’s conclusion that such discipline imposed on the supervisor would have a “carryover” effect and would influence the supervisor in the performance of his adjustment functions after the strike and hence interfere with and coerce the employer in the choice of his grievance representative. See Triangle, 216 N. L. R. B., at 897; Hammond, 216 N. L. R. B., at 904. The Triangle decision 426 437 U. S. OCTOBER TERM, 1977 Opinion of the Court was not challenged in the courts, but Hammond was enforced, 176 U. S. App. D. C. 240, 539 F. 2d 242 (1976), as was Skippy Enterprises, 532 F. 2d 47 (CA7 1976).23 III This case was tried to the Administrative Law Judge prior to the issuance of this Court’s decision in FP&L, but hearings continued and the record was not closed until after the Court of Appeals’ final decision in that case; and the FP&L opinion here was handed down on June 24, 1974, some three months before the Administrative Law Judge issued his recommended decision. As we have already indicated, the findings of the Administrative Law Judge, accepted by the Board, were that the hyphenates’ regular supervisory duties included the performance of grievance adjustment; that the employer insisted that hyphenates return to work but only to perform supervisory, not rank-and-file, duties;24 and that the hypen-ates who reported did only supervisory work and had the 23 In Hammond and Skippy, the supervisor also performed some rank-and-file work during the strike. The Board in Hammond characterized the amount of rank-and-file work as minimal, and only incidental to the supervisory functions, but in Skippy, the supervisor performed ra.nk-a.nd-file work for about 30% of his time. In light of the finding that the supervisors performed no rank-and-file writing in this case, we are not presented with that element of the Board’s reasoning in Hammond and Skippy. 24 We note also respondent’s argument that the limited writing duties— the A-to-H functions—normally performed by the hyphenates should be considered rank-and-file work within the meaning of FP&L. The Admin-istrative Law Judge gave careful attention to the issue and concluded to the contrary, App. to Pet. for Cert, in No. 76-1162, p. 59a, and the Board accepted his findings and conclusions in this respect. We also find them unexceptionable. The dissenting Board member did not premise his opinion on the A-to-H issue. We thus do not have here the situation where the disciplined supervisor has performed not only supervisory duties, including grievance adjustment, but also has done some rank-and-file tasks. See Hammond and Triangle, and also Wisconsin River Valley. AMERICAN BROADCASTING COS. v. WRITERS GUILD 427 411 Opinion of the Court authority to adjust grievances which they did when the occasion arose.25 After analyzing this Court’s pronouncements in FP&L, the Administrative Law Judge rejected the claim that union discipline of a supervisor-member for working during a strike can never bea§8(b)(l)(B) violation and went on to hold that under the test prescribed by FP&L, there was a violation here. His conclusions were that through its strike rules and other pressures “designed to compel such hyphenates from going to work during the strike,” regardless of the tasks that they might perform, the union had “restrained and coerced the hyphenates from performing managerial and supervisory services for their employers during the strike, including the adjustment of employee grievances and participation in collective bargaining . . . 26 By “coercing or restraining” hyphenates from going in to do their normal work, which included grievance adjustment, or in the case of producers, on distant location, the task of collective bargaining, the union had “actually coerced and restrained their employers from selecting those persons as the employers’ representatives for the adjustment of grievances and for collective bargaining during the strike.” 27 He also concluded that by charging, trying, and disciplining those hyphenates who did report for work and by “threatening to blacklist in perpetuity . . . [and] to drive [them] out of the industry,” 28 the union had coerced and restrained these hyphenates from performing their regular 25 It is suggested that there was insufficient proof that the hyphenates who worked actually engaged in grievance adjustment of any kind during the strike. But the findings were to the contrary; and, in any event, there is no question that they were authorized to do so and were available for that purpose when and if the occasion arose. Section 8(b)(1)(B) obviously can be violated by attempting coercively to control the choice of the employer’s representative, before, as well as after, the representative has actually dealt with the grievance. 26 App. to Pet. for Cert, in No. 76-1162, p. 62a. 27 Id., at 64a. 28 Id., at 69a. 428 OCTOBER TERM, 1977 Opinion of the Court 437U.S. duties in the normal manner, including the adjustment of grievances and collective bargaining. The employer, in turn, had been further coerced and restrained in the free selection of those hyphenates as his collective-bargaining and grievanceadjustment representatives. The Administrative Law Judge thus found the section violated according to the test as elaborated in FP&L because, by keeping hyphenates from work, the union had deprived the employer of any opportunity to select those particular supervisors as his grievance-adjusting or collective-bargaining representatives29 and because disciplining and threatening those supervisors who had reported for duty deprived the employer of fully effective §8 (b)(1)(B) representatives. Although 29 The Administrative Law Judge reasoned, as follows, in support of his conclusion. “To illustrate: A person performing the function of a director acts in a managerial or supervisory capacity, which normally includes the adjustment of grievances of actors, actresses, craft employees and others. One occupying the position of a producer normally has a similar capacity and similar duties with respect to employee grievances. In addition, if the film is being shot on distant location the producer has authority to negotiate on the spot agreements with local unions. Thus when Respondent prevented or sought to prevent, such hyphenate members from going to work in their managerial and supervisory capacities as producers and directors during the strike, Respondent obviously coerced and restrained their employers in the selection of those specific producers and directors for the purpose of collective bargaining and the adjustment of grievances of employees working during the strike within the plain meaning of the statute. Similarly, those persons employed as story editors or in like classifications perform executive functions normally, and appear to have done so during the strike, in which the record indicates they were engaged as supervisors and actual or potential representatives of their employers for the adjustment of grievances. Respondent, by coercing or restraining persons in these classifications from going in to do their normal work thereby actually coerced and restrained their employers from selecting those persons as the employers’ representatives for the adjustment of grievances and for collective bargaining during the strike.” Id., at 63a-64a. (Footnote omitted.) AMERICAN BROADCASTING COS. v. WRITERS GUILD 429 411 Opinion of the Court the Board embraced these findings and conclusions of the Administrative Law Judge,30 it also found that the disciplinary action taken by the union against those hyphenates who crossed the picket line was an unfair practice under § 8 (b) (1)(B) as that section had been construed in Hammond and Triangle and that threats of such illegal discipline against others also violated the section. IV We cannot agree with what appears to be the fundamental position of the Court of Appeals and the union that under §8 (b)(1)(B), as the section was construed in FP&L, it is never an unfair practice for a union to discipline a supervisormember for working during a strike, regardless of the work that he may perform behind the picket line. The opinion in FP&L expressly refrained from questioning Oakland Mailers or the proposition that an employer could be coerced or restrained within the meaning of §8 (b)(1)(B) not only by picketing or other direct actions aimed at him but also by debilitating discipline imposed on his collective-bargaining or grievance-adjustment representative. Indeed, after focusing on the purposes of the section, the Court in FP&L delineated the boundaries of when that “carryover” effect would violate §8(b)(l)(B): whenever such discipline may adversely affect the supervisor’s conduct in his capacity as a grievance adjustor or collective bargainer. In these situations—that is, when such impact might be felt—the employer would be deprived of the full services of his representatives and hence would be restrained and coerced in his selection of those representatives. Furthermore, because this was the test prescribed and employed by the Court to adjudicate the very situation where 30 It is suggested by respondent that the Board did not fully adopt the approach of the Administrative Law Judge, but it is plain that, with the single exception noted above, the Board adopted all of the findings and conclusions of the Administrative Law Judge. 430 437 U. S. OCTOBER TERM, 1977 Opinion of the Court union discipline was imposed for crossing a picket line, it is unlikely that the Court anticipated that the test could never be satisfied in such disciplinary cases, that it could never be true that the sanction could or would affect the supervisor’s collective-bargaining or grievance-adjustment functions, or that the employer in such circumstances could never be restrained or coerced in the selection of his representatives. This is not to say that every effort by a union to discipline a supervisor for crossing a picket line to do supervisory rather than rank-and-file work would satisfy the standards specified by FP&L, or that on facts present here there is necessarily a violation of § 8 (b)(1)(B). But we are of the view that the Board correctly understood FP&L to mean that in ruling upon a § 8 (b)(1)(B) charge growing out of union discipline of a supervisory member who elects to work during a strike, it may indeed, it must—inquire whether the sanction may adversely affect the supervisor’s performance of his collectivebargaining or grievance-adjustment tasks and thereby coerce or restrain the employer contrary to §8 (b)(1)(B). The Board addressed those issues here, and if its ultimate factual conclusions in this regard are capable of withstanding judicial review, it seems to us that its construction of the section fairly recognizes and respects the outer boundaries established by FP&L, and represents an “acceptable reading of the statutory language and a reasonable implementation of the purposes of the relevant statutory sections.” NLRB v. Iron Workers, 434 U. S. 335, 341 (1978). Respondent objects that this construction of the Act impermissibly intrudes on the union’s right to resort to economic sanctions during a strike. However, an employer also has economic rights during a strike, and the statute declares that, in the unrestrained freedom to select a grievance-adjustment and collective-bargaining representative, the employer’s rights dominate. Ample leeway is already accorded to a union in permitting it to discipline any member, even a supervisor, for AMERICAN BROADCASTING COS. v. WRITERS GUILD 431 411 Opinion of the Court performing struck work—to carry that power over to the case of purely supervisory work is an inappropriate extension and interference with the employer’s prerogative. The Board has so ruled, and as the Court has often observed, “ ‘[t]he function of striking [the] balance to effectuate national labor policy is often a difficult and delicate responsibility, which the Congress committed primarily to the National Labor Relations Board, subject to limited judicial review.’ ” NLRB n. Iron Workers, supra, at 350, quoting NLRB v. Truck Drivers, 353 U. S. 87, 96 (1957) ; NLRB v. Insurance Agents, 361 U. S. 477, 499 (1960). Here, in adjudicating as it did the intertwining interests of union, employer, and supervisor-member during an economic strike, we cannot say that the Board has moved into a new area of regulation not committed to it by Congress, ibid., or conclude that the role assumed by the Board is “fundamentally inconsistent with the structure of the Act and the function of the sections relied upon.” American Ship Building Co. v. NLRB, 380 U. S. 300, 318 (1965) ; NLRB n. Iron Workers, supra.31 V We are also unpersuaded that the Board’s findings and conclusions are infirm on any of the grounds submitted. First, it is urged that there was an insufficient showing and insufficient findings that any hyphenates were coerced or restrained from reporting for work. But the Administrative Law Judge carefully detailed the strike rules that he expressly found were designed and enforced with the intent of restraining hyphenates from going to work and from performing the normal duties of their positions, which included the adjustment of 31 The Board’s decision holding the union responsible under § 8 (b) (1) (B) for the foreseeable course and consequences of its actions is not inconsistent with Teamsters v. NLRB, 365 U. 8. 667 (1961), and NLRB v. News Syndicate Co., 365 U. S. 695 (1961). The holding does not rest on any assumption that the union will act illegally in the future. 432 OCTOBER TERM, 1977 Opinion of the Court 437U.S. grievances.32 It was also found that the hyphenates were especially vulnerable to pressure from the union and that many of them were actually restrained and prevented from performing their normal duties, including the adjustment of grievances. These are sufficiently clear findings that union pressures kept many hyphenates from the job, and, on the record before us, it approaches the frivolous to argue that there is insufficient evidence to support them. It also follows, as the Administrative Law Judge and the Board concluded, that as to those hyphenates whom the union kept from work, •the employer was restrained and coerced within the meaning of §8 (b)(1)(B) by being totally deprived of the opportunity to choose these particular supervisors as his collectivebargaining or grievance-adjustment representatives during the strike. Second, as to those hyphenates who reported for work, it is strenuously urged that there is no basis for concluding that the discipline imposed upon them would adversely affect the performance of their grievance-adjustment duties either during or after the strike. Again, however, we are unwilling to differ with the Board in these respects. The inquiry whether union conduct would or might adversely affect the performance of the hyphenates’ grievance-adjustment duties is, as petitioners assert, necessarily a matter of probabilities, and its resolution depends much on what experience would suggest are the justifiable inferences from the known facts. This seems to us to be peculiarly the kind of determination that Congress has assigned to the Board: An administrative agency with power after hearings to 32 The findings were also that: The record is convincing that Respondent, well aware of the primary supervisory, management, and executive functions of its hyphenate-members, drafted its strike rules and enforced them with the intent of compelling those hyphenate-members from going to work during the strike, without regard to the capacity in which they performed or the work done.” App. to Pet. for Cert, in No. 76-1162, p. 69a. AMERICAN BROADCASTING COS. v. WRITERS GUILD 433 411 Opinion of the Court determine on the evidence in adversary proceedings whether violations of statutory commands have occurred may infer within the limits of the inquiry from the proven facts such conclusions as reasonably may be based upon the facts proven. One of the purposes which lead to the creation of such boards is to have decisions based upon evidential facts under the particular statute made by experienced officials with an adequate appreciation of the complexities of the subject which is entrusted to their administration.” Republic Aviation Corp. n. NLRB, 324 U. S. 793, 800 (1945); Radio Officers v. NLRB, 347 U. S. 17, 48-49 (1954). See also NLRB v. Erie Resistor Corp., 373 U. S. 221, 227 (1963); Teamsters v. NLRB, 365 U. S. 667, 675 (1961). The Board’s findings are “entitled to the greatest deference in recognition of its special competence in dealing with labor problems.” American Ship Building Co. v. NLRB, supra, at 316. Furthermore, it does not strike us as groundless or lacking substantial evidence for the Board to conclude on this record that the discipline imposed would have the necessary adverse effect. Strike rules were distributed in February; the strikes against the Association began on March 4 and terminated June 24; the strikes against the networks began on March 29 and ended on July 12. Between April 6 and November 8— both during and after the strikes—some 31 hyphenates who had worked during the strikes were charged with violating union rules,33 15 hearings had been held prior to the closing of evidence in November 1973, and from June 25 to September 28, very substantial penalties were imposed in 10 cases although 9 have already been reduced on appeal. These penalties were widely publicized at the time of their imposi 33 Violations of Rules 1, 12, 13, and 28 were alleged. See, supra, at 415, 416, 417, and n. 3. 434 OCTOBER TERM, 1977 Opinion of the Court 437 U. S. tion. Other charges were pending and remained to be tried when the record was closed in this case. These penalties were meted out at least in part because the accused hyphenates had complied with the orders of their employers by reporting for work and performing only their normal supervisory functions, including the adjustment of grievances, during the strike. Hyphenates who worked were thus faced not only with threats but also with the actuality of charges, trial, and severe discipline simply because they were working at their normal jobs. And if this were not enough, they were threatened with a union blacklist that might drive them from the industry. How long such hyphenates would remain on the job under such pressure was a matter no one, particularly the employer, could predict. Moreover, after the strike, with the writers back at work, the hyphenates who had worked during the strike still faced charges and trials or were appealing large fines and long suspensions. At the same time, they were expected to perform their regular supervisory duties and to adjust grievances whenever the occasion demanded, functions requiring them to deal with the same union which was considering the appeal of their personal sanctions. As to these supervisors, who had felt the union’s wrath, not for doing rank-and-file work contrary to union rules, but for performing only their primary supervisory duties during the strike and who were in a continuing controversy with the union, it was not untenable for the Board to conclude that these disciplined hyphenates had a dimin-ished capacity to carry out their grievance-adjustment duties effectively and that the employer was deprived of the full range of services from his supervisors.34 Such a hyphenate 34 In determining that the Board had exceeded the limitations of the statute in the FP&L and Illinois Bell cases, the Court of Appeals for the District of Columbia Circuit recognized that when a supervisor acts as a grievance adjustor, “he is a representative of management, and as such he should be immune from union discipline. The unions participating in the present cases conceded as much at oral argument when they agreed that AMERICAN BROADCASTING COS. v. WRITERS GUILD 435 4U Opinion of the Court might be tempted to give the union side of a grievance a more favorable slant while the threat of discipline remained, or while his own appeal of a union sanction was pending. At the very least, the employer could not be certain that a fined hyphenate would willingly answer the employer’s call to duty during a subsequent work stoppage, particularly if it occurred in the near future.35 For an employer in these circumstances to insure having satisfactory collective-bargaining and grievance-adjustment services would require a change in his representative. As the Board has construed the Act from Oakland Mailers to Triangle, Hammond, and the cases now before us, such a likely impact on the employer constitutes sufficient restraint and coercion in connection with the selection of collectivebargaining and grievance-adjustment representatives to vio-when a supervisor crosses a picket line to perform supervisory work he remains immune from discipline. . . . The dividing line between supervisory and nonsupervisory work in the present context is sharply defined and easily understood.” 159 U. S. App. D. C., at 286, 487 F. 2d, at 1157. As the Court of Appeals for the Seventh Circuit said: “[W]here supervisors cross picket lines to perform rank-and-file struck work, union discipline does not violate Section 8 (b) (1) (B) since it merely deprives the employer of services normally rendered by strikebreaking replacement employees.” Skippy Enterprises, 532 F. 2d 47, 53 (1976). On the other hand, Where supervisors cross picket lines to perform regular supervisory duties, union discipline violates Section 8(b)(1)(B) since it tends to deprive the employer of its supervisors’ services—including their §8 (b)(1)(B) services and because the supervisors would reasonably anticipate that union discipline would also be imposed if future performance of their § 8 (b) (1) (B) functions did not meet with union approval.” Ibid. Union discipline might even result in depriving the employer of the supervisors’ services forever, if the blacklist involved in this case had been successful. The employer would have had no choice but to let the hyphenate go, since the positions of director, producer, and script editor unavoidably require working with rank-and-file writers. 436 OCTOBER TERM, 1977 Opinion of the Court 437 U. S. late § 8 (b) (1)(B). In FP&L the Court declined the invitation to overrule Oakland Mailers, and we do so again. Union pressure on supervisors can affect either their willingness to serve as grievance adjustors or collective bargainers, or the manner in which they fulfill these functions; and either effect impermissibly coerces the employer in his choice of representative.36 Third, it is further urged that union discipline could not adversely affect a supervisor’s later performance of his §8(b)(l)(B) duties because the employer could require him to leave the union and thus free himself from further threats of union discipline. This submission has little force in this case, since, as the Administrative Law Judge found, the union’s known policy was not to permit a member to resign during a strike and for a period of six months thereafter. For the entire period to which the Board’s findings were addressed, hyphenates could not terminate their membership, and the . 36 In the FP&L and Illinois Bell cases, the Court of Appeals for the District of Columbia Circuit noted that its consistent view has been that the “basic rationale [of Oakland Mailers'] is consistent with the purposes of Section 8 (b)(1)(B) . . . [for] management’s right to a free selection would be hollow indeed if the union could dictate the manner in which the selected representative performed his collective bargaining and grievance adjustment duties.” 159 U. S. App. D. C., at 282, 283, 487 F. 2d, at 1153, 1154. The court also noted its agreement with New Mexico District Council of Carpenters and Joiners of America ( A. S. Horner, Inc.), 177 N. L. R. B. 500 (1969), enf’d, 454 F. 2d 1116 (CAIO 1972), where a union member worked as a supervisor for a company which had no contract with the union. 159 U. S. App. D. C., at 284 n. 19, 487 F. 2d, at 1155 n. 19. A fine imposed in these circumstances violated the section because compliance by the supervisor with the union’s demands would have required his leaving his job and thus have “the effect of depriving the Company of the services of its selected representative for the purposes of collective bargaining or the adjustment of grievances.” 177 N. L. R. B., at 502. The Court of Appeals said that A. S. Horner “thus falls close to the original rationale of § 8 (b)(1)(B) which was to permit the employer to keep the bargaining representative of his own choosing.” 159 U. S. App. D. C., at 284 n. 19, 487 F. 2d., at 1155 n. 19. AMERICAN BROADCASTING COS. v. WRITERS GUILD 437 411 Opinion of the Court employer’s only recourse would have been to replace them as his grievance representatives. Carried to its logical end, this submission is simply another argument that union sanctions applied to supervisor-members who work during a strike can never violate §8 (b)(1)(B), because the employer could always insist that his supervisors either terminate union affiliation or face discharge. Yet, as we have noted, the test posited by this Court in FP&L plainly recognizes the possibility of a § 8 (b)(1)(B) violation arising from union fines imposed during a strike. Moreover, if the argument were to be accepted, indirect pressures on the employer by sanctioning supervisor-members for the manner in which they perform their grievance-adjusting function (as in Oakland Mailers) would never be a violation because the supervisor could, at the employer’s request, escape from union threats and sanctions. The Board’s construction of the Act is to the contrary, however, and, as we have said, we are not prepared at this juncture to override it.37 37 It is also argued that at the very least the Board erred with respect to director-hyphenates because there is no evidence and no finding that directors ever dealt with writers or adjusted their grievances even if producers and story editors did. Hence, it is alleged that union discipline of directors could not possibly affect their adjustment of writers’ grievances during or after the strike for the simple reason that they had none to adjust. But during the strike, no supervisor, writer, director, producer, or story editor had writer grievances to adjust—at least no new grievances—because there were no writers on the job and only the possibility that there might be replacements or a few strikebreakers. Nevertheless, directors, as well as others, had adjustment duties with respect to other employees. The Administrative Law Judge found that directors “hire or effectively recommend the employment of crew and actors, effectively direct such employees, and . . . have authority to and do adjust grievances of such employees.” App. to Pet. for Cert, in No. 76-1162, p. 28a. Directors’ willingness to work and to perform these duties subjected them to sanctions and financial loss, making them less than completely reliable and effective employer representatives for the duration of the strike, and 438 OCTOBER TERM, 1977 Stewart, J., dissenting 437 U. S. Because we have concluded that the Board’s construction of § 8 (b)(1) (B) is not an unreasonable reading of its language or inconsistent with its purposes, and because we cannot say that the Board’s findings lacked substantial evidence, we must reverse the judgment of the Court of Appeals. So ordered. Mr. Justice Stewart, with whom Mr. Justice Brennan, Mr. Justice Marshall, and Mr. Justice Stevens join, dissenting. The Court holds today that a labor union locked in a direct economic confrontation with an employer is powerless to impose sanctions on its own members who choose to pledge their loyalty to the adversary. Nothing in §8 (b)(1)(B) or any other provision of the National Labor Relations Act permits such a radical alteration of the natural balance of less likely to perform any supervisory task during future strikes. A union may no more interfere with the employer’s choice of a grievance representative with respect to employees represented by other unions than with respect to those employees whom it itself represents. International Organization of Masters, Mates and Pilots, International Marine Division, 197 N. L. R. B. 400 (1972), enf’d, 159 U. S. App. D. C. 11, 14, 486 F. 2d 1271, 1274 (1973), cert, denied, 416 U. S. 956 (1974), and International Organization of Masters, Mates and Pilots n. NLRB, 539 F. 2d 554, 559-560 (CA5 1976). We note also that all hyphenates, including directors, were threatened with a permanent blacklist—a refusal by other Guild members, including producers, other directors, and story editors, as well as writers, to work with the offending director—and that revocation of the formal rule on April 30 did not completely remove the threat. Because of his central role, refusal to work with a director means refusal to participate at all in a particular film. The union thus threatened a strike by all of its members against the employer who permitted director-hyphenates to work, plainly an independent violation of §8 (b)(1)(B). The Administrative Law Judge found that of the 15 union members employed as directors by petitioners, 3 were charged with strike rule violations, and 1 was brought before a trial panel and disciplined. App. to Pet. for Cert, in No. 76-1162, p. 29a. AMERICAN BROADCASTING COS. v. WRITERS GUILD 439 411 Stewart, J., dissenting power between labor and management. I therefore respectfully dissent. A union’s ability to maintain a unified front in its confrontations with management and to impose disciplinary sanctions on those who “adherfe] to the enemy in time of struggle are essential to its survival as an effective organization. See Summers, Legal Limitations on Union Discipline, 64 Harv. L. Rev. 1049, 1066 (1951). An employer also has an interest in securing the loyalty of those who represent him in dealings with the union, and that interest is protected by specific provisions of the Act.1 Thus, as the Court observed in Florida Power & Light Co. v. Electrical Workers, 417 U. S. 790 (FP&L), very real concerns are raised on both sides when supervisory employees with collective-bargaining and grievance-adjustment responsibilities are also union members. But § 8 (b)(1) (B) is not “any part of the solution to the generalized problem of supervisor-member conflict of loyalties ” 417 U. 8., at 813. That statutory provision was enacted for the primary purpose of prohibiting a union from exerting direct pressure on an employer to force him into a multiemployer bargaining unit or to dictate his choice of representatives for the settlement of employee grievances. S. Rep. No. 105, 80th Cong., 1st Sess., pt. 1, p. 21 (1947). The Court in FP&L reserved decision on whether union pressure expressly aimed at affecting the manner in which supervisor-members performed their collective-bargaining or grievance-adjustment functions might 1 This interest is protected by § 2 (3) of the National Labor Relations Act, which excludes “supervisors” as defined in §2 (11) from the definition of “employees,” thereby excluding them from the coverage of the Act. Thus an employer may discharge or otherwise penalize a supervisory employee for engaging in what would otherwise be protected concerted activity under the Act. In addition, § 14 (a) of the Act provides that “no employer . . . shall be compelled to deem . . . supervisors as employees for the purpose of any law . . . relating to collective bargaining.” See Florida Power & Light Co. v. Electrical Workers, 417 U. S. 790, 808-811. 440 OCTOBER TERM, 1977 Stewart, J., dissenting 437U.S. fall within the “outer limits” of the proscription of § 8 (b) (1)(B). 417 U. S., at 805. See San Francisco-Oakland Mailers’ Union No. 18 {Northwest Publications, Inc.), 172 N. L. R. B. 2173. But it flatly rejected the argument that union discipline aimed at enforcing uniform rules violated §8 (b)(1)(B) simply because it might have the ancillary effect of “depriv[ing] the employer of the full allegiance of, and control over, a representative he has selected for grievance adjustment or collective bargaining purposes.” 417 U. S., at 807. In the present cases it is entirely clear that the union had no interest in restraining or coercing the employers in the selection of their bargaining or grievance-adjustment representatives, or in affecting the manner in which supervisory employees performed those functions. As the Court notes, ante, at 417-418, and n. 6, the union expressed no interest at the disciplinary trials in the kind of work that was done behind its picket lines. Its sole purpose was to enforce the traditional kinds of rules that every union relies on to maintain its organization and solidarity in the face of the potential hardship of a strike. Cf. NLRB v. Allis-Chalmers Mjg. Co., 388 U. S. 175, 181-184. In reversing the judgment of the Court of Appeals, this Court today forbids a union from disciplining a supervisormember who crosses its picket line—who clearly gives “aid and comfort to the enemy” during a strike, see Summers, supra, at 1066—solely because that action may have the incidental effect of depriving the employer of the hypothetical grievance-adjustment services of that particular supervisor for the duration of the strike. This ruling quite simply gives the employer the superior right to call on the loyalty of any supervisor with grievance-adjustment responsibilities,2 when- 2 Since the power to adjust employee grievances is one of the statutory indicia of supervisory status under §2 (11) of the Act, many if not most AMERICAN BROADCASTING COS. v. WRITERS GUILD 441 411 Stewart, J., dissenting ever the union to which the supervisor belongs calls him out on strike. In short, the Court’s decision prevents a union with supervisory members from effectively calling and enforcing a strike.3 Nothing in § 8 (b)(1)(B) permits such a sweeping limitation on the choice of economic weapons by unions that include supervisory employees among their members. On the contrary, as the Court clearly held in FP&L, supra, an employer’s remedy if he does not want to share the loyalty of his supervisors with a union is to insist that his supervisory personnel not belong to a union; or if he does not welcome the consequences of his supervisors’ union membership he may legally penalize them for engaging in union activities, see n. 1, supra, or “resolvfe] such conflicts as arise through the traditional procedures of collective bargaining.” FP&L, supra, at 813.4 The sole function of §8 (b)(1)(B) is to protect an employer from any union coercion of the free choice of his bargaining or grievance-adjustment representative. In prohibiting union interference in his choice of representatives for dealings with thef union, this statutory provision does not in any supervisory employees will fall within the Court’s ruling when they are restrained] . . . from going to work and from performing the normal duties of their positions, which includfe] the adjustment of grievances.” Ante, at 431-432. 3 Under this rule, it would appear that a separate union consisting entirely of supervisory employees would commit an unfair labor practice if it ordered its members not to cross the picket lines of another union, or indeed, if it called an economic strike entirely on its own, since the employer would thereby be deprived of the services of his chosen grievanceadjustment representatives. 4 Alternatively, the employer may ease the dilemma of his supervisory employees by offering to provide their defense or to indemnify them against any fines that might be imposed by the union for a breach of strike discipline. Several of the employers in this case did in fact extend such offers to the hyphenates. See decision of the Administrative Law Judge, App. to Pet. for Cert, in No. 76-1162, p. 42a. 442 OCTOBER TERM, 1977 Stewart, J., dissenting 437U.S. way grant him a right to interfere in the union’s relationship with its supervisor-members.5 The statute leaves the balance of power in equipoise. The Court’s decision, by contrast, tips it measurably in favor of the employer at the most delicate point of direct confrontation, by completely preventing the union from enlisting the aid of its supervisor-members in a strike effort. It seems to me that the Court’s reading of § 8 (b)(1)(B) is “fundamentally inconsistent with the structure of the Act and the function of the sections relied upon.” American Ship Building Co. n. NLRB, 380 U. S. 300, 318. Accordingly, I would affirm the judgment of the Court of Appeals. 5 In San Francisco-Oakland Mailers Union No. 18 {Northwest Publications, Inc.), 172 N. L. R. B. 2173, the Board found a violation of § 8 (b) (1)(B) when a union expelled member-foremen for allegedly assigning bargaining-unit work in violation of the collective-bargaining agreement. It reasoned that the employer’s statutory right to choose his bargaining representative would be rendered illusory if the union could effectively control the actions of any individual who happened to occupy the position. I adhere to the view expressed by the Court in FP&L, 417 U. S., at 805, that this ruling is at best within the “outer limits” of §8 (b)(1)(B). ZENITH RADIO CORP. v. UNITED STATES 443 Syllabus ZENITH RADIO CORP. v. UNITED STATES CERTIORARI TO THE UNITED STATES COURT OF CUSTOMS AND PATENT APPEALS No. 77-539. Argued April 25, 1978—Decided June 21, 1978 Petitioner, an American manufacturer of consumer electronic products, filed a petition with the Commissioner of Customs, requesting assessment under § 303 of the Tariff Act of 1930 of countervailing duties on various consumer electronic products exported from Japan to this country. Petitioner contended that the products benefited from bounties or grants paid or conferred by Japan because Japan imposes a commodity tax (an “indirect” tax) on those products when they are sold in that country but “remits” the tax when the products are exported, any tax paid on the shipment of a product being refunded upon the subsequent exportation. Section 303 provides that whenever a foreign country pays a “bounty or grant” upon the exportation of a product from that country, the Secretary of the Treasury (Secretary) must levy a countervailing duty “equal to the net amount of such bounty or grant” upon the importation of the product into the United States. After rejection of its request petitioner filed suit in the Customs Court, claiming that the Treasury Department had erred in concluding that remission of the Japanese tax was not a bounty or grant within the purview of § 303. The Secretary contended that since the remission of the tax was “nonexcessive” (i. e., not above the amount of the tax paid or otherwise due), § 303 did not require assessment of a countervailing duty. Relying on Downs v. United States, 187 U. S. 496, the Customs Court ruled in petitioner’s favor. The Court of Customs and Patent Appeals reversed. Held: Japan does not confer a “bounty or grant” within the meaning of § 303 on the consumer electronic products by failing to impose a commodity tax on those products when they are exported to this country, while imposing the tax on the products when they are sold in Japan. Downs v. United States, supra, distinguished Pp. 450-462. (a) The Secretary’s statutory interpretation that was followed in this case has been consistently maintained since the basic countervailing-duty statute was enacted in 1897, and that administrative interpretation is entitled to great weight. See Udall v. Tailman, 380 U. S. 1, 16. Pp. 450-451. (b) The legislative history of the statute suggests that the term 444 OCTOBER TERM, 1977 Syllabus 437 U. S. “bounty” was not intended to encompass the nonexcessive remission of an indirect tax. Pp. 451-455. (c) The Secretary’s interpretation was reasonable in light of the statutory purpose of the countervailing duty, viz., offsetting the unfair competitive advantage that foreign products would otherwise enjoy from export subsidies paid by their governments. In deciding in 1898 that a nonexcessive remission of indirect taxes did not give the exporter an unfair competitive advantage, the Secretary permissibly viewed the remission as a reasonable measure for avoiding double taxation of exports—once by the foreign country and once upon sale in this country. Pp. 455-457. (d) The Secretary’s interpretation is as permissible today as it was in 1898. The statute has been re-enacted five times with no modification of the relevant language, and the Secretary’s position has been incorporated into an international agreement followed by every major trading nation in the world. It is not for the judiciary to substitute its views as to the fairness and economic effect of remitting indirect taxes. Pp. 457-459. (e) Downs n. United States, supra, did not involve the issue of whether a nonexcessive remission of taxes, standing alone, would have constituted a bounty on exportation, and is not dispositive of this case. Pp. 459-462. 64 C. C. P. A. 130, 562 F. 2d 1209, affirmed Marshall, J., delivered the opinion for a unanimous Court. Frederick L. Ikenson argued the cause for petitioner. With him on the briefs were Eugene L. Stewart and Philip J. Curtis. Solicitor General McCree argued the cause for the United States. With him on the brief were Assistant Attorney General Babcock, Deputy Solicitor General Easterbrook, Richard A. Allen, Leonard Schaitman, David M. Cohen, and Robert H. Mundheim* *Briefs of amici curiae urging affirmance were filed by Saul L. Sherman for the American Importers Assn., Inc.; and by N. David Palmeter and David P. Houlihan for the Union des Industries de la Communaute Europeenne. Briefs of amici curiae were filed by Marjorie M. Shostak, S. Richard Shostak, Theodore B. Olson, and James F. O’Hara for Craig Corp, et al.; and by Robert E. Herzstein for Ford Motor Co. ZENITH RADIO CORP. v. UNITED STATES 445 443 Opinion of the Court Mr. Justice Marshall delivered the opinion of the Court. Under § 303 (a) of the Tariff Act of 1930, 46 Stat. 687, as amended, 19 U. S. C. § 1303 (a) (1976 ed.), whenever a foreign country pays a “bounty or grant” upon the exportation of a product from that country, the Secretary of the Treasury is required to levy a countervailing duty, “equal to the net amount of such bounty or grant,” upon importation of the product into the United States.1 The issue in this case is whether Japan confers a “bounty” or “grant” on certain consumer electronic products by failing to impose a commodity tax on those products when they are exported, while imposing the tax on the products when they are sold in Japan. 1 Section 303 (a) provides in relevant part: “(1) Whenever any country, dependency, colony, province, or other political subdivision of government, person, partnership, association, cartel, or corporation, shall pay or bestow, directly or indirectly, any bounty or grant upon the manufacture or production or export of any article or merchandise manufactured or produced in such country, dependency, colony, province, or other political subdivision of government, then upon the importation of such article or merchandise into the United States, whether the same shall be imported directly from the country of production or otherwise, and whether such article or merchandise is imported in the same condition as when exported from the country of production or has been changed in condition by remanufacture or otherwise, there shall be levied and paid, in all such cases, in addition to any duties otherwise imposed, a duty equal to the net amount of such bounty or grant, however the same be paid or bestowed. “(5) The Secretary shall from time to time ascertain and determine, or estimate, the net amount of each such bounty or grant, and shall declare the net amount so determined or estimated. “(6) The Secretary shall make all regulations he deems necessary for the identification of articles and merchandise subject to duties under this section and for the assessment and collection of such duties. All determinations by the Secretary under this section, and all determinations by the Commission under subsection (b)(1) of this section (whether affirmative or negative) shall be published in the Federal Register.” 19 U. S. C. §1303 (a) (1976 ed.). 446 OCTOBER TERM, 1977 Opinion of the Court 437U.S. I Under the Commodity Tax Law of Japan, Law No. 48 of 1962, see App. 44-48, a variety of consumer goods, including the electronic products at issue here, are subject to an “indirect” tax—a tax levied on the goods themselves, and computed as a percentage of the manufacturer’s sales price rather than the income or wealth of the purchaser or seller. The Japanese tax applies both to products manufactured in Japan and to those imported into Japan.2 On goods manufactured in Japan, the tax is levied upon shipment from the factory; imported products are taxed when they are withdrawn from the customs warehouse. Only goods destined for consumption in Japan are subject to the tax, however. Products shipped for export are exempt, and any tax paid upon the shipment of a product is refunded if the product is subsequently exported. Thus the tax is “remitted” on exports.3 In April 1970 petitioner, an American manufacturer of consumer electronic products, filed a petition with the Commissioner of Customs,4 requesting assessment of countervailing duties on a number of consumer electronic products exported from Japan to this country.5 Petitioner alleged that Japan 2 See App. 12-13, 30-31; An Outline of Japanese Taxes 128-129 (Tax Bureau, Japanese Ministry of Finance, 1976). For the products at issue here, the rate of taxation apparently ranges from 15% to 20%. See App. 13-14; An Outline of Japanese Taxes, supra, at 131. 3 For purposes of this opinion, we adopt the convention followed by the parties and use the term “remission” to encompass both the exemption of exports from initial taxation and the refund to the exporter of any taxes already paid. 4 The Secretary of the Treasury has delegated the authority to make countervailing-duty determinations to the Commissioner of Customs, subject to the Secretary’s approval. See 19 CFR § 159.47 (1977). 5 The products included television receivers, radio receivers, radio-phonograph combinations, radio-television-phonograph combinations, radiotape-recorder combinations, record players and phonographs complete with amplifiers and speakers, tape recorders, tape players, and color television ZENITH RADIO CORP. v. UNITED STATES 447 443 Opinion of the Court had bestowed a “bounty or grant” upon exportation of these products by, inter alia, remitting the Japanese Commodity Tax that would have been imposed had the products been sold within Japan. In January 1976, after soliciting the views of interested parties and conducting an investigation pursuant to Treasury Department regulations, see 19 CFR § 159.47 (c) (1977), the Acting Commissioner of Customs published a notice of final determination, rejecting petitioner’s request. 41 Fed. Reg. 1298 (1976).6 Petitioner then filed suit in the Customs Court, claiming that the Treasury Department had erred in concluding that remission of the Japanese Commodity Tax was not a bounty or grant within the purview of the countervailing-duty statute.7 The Department defended on the ground that, since the remission of indirect taxes was “nonexcessive,” the statute did not require assessment of a countervailing duty. In the Department’s terminology, a remission of taxes is “nonexcessive” if it does not exceed the amount of tax paid or otherwise due; thus, for example, if a tax of $5 is levied on goods at the factory, the return of the $5 upon exportation would be “nonexcessive,” whereas a payment of $8 from the government to the manufacturer upon exportation would be “excessive” by $3. The Department pointed out that the current picture tubes. See 37 Fed. Reg. 10087, App. A-(1972), as amended, 37 Fed. Reg. 11487 (1972). 6 The notice stated in relevant part that “on the basis of the . . . facts gathered and the investigation conducted pursuant to . . . Customs Regulations ... a final determination is hereby made . . . that ... no bounty or grant is being paid or bestowed, directly or indirectly, within the meaning of section 303 . . . upon the . . . exportation of certain consumer electronic products from Japan.” 41 Fed. Reg. 1298 (1976). 7 Suit was filed pursuant to a provision, enacted in 1975, authorizing American manufacturers, producers, and wholesalers to seek review in the Customs Court of administrative decisions not to impose countervailing duties under § 303. Tariff Act of 1930, as amended, § 516 (d), 19 U. S. C. §1516 (d) (1976 ed.). 448 437 U. S. OCTOBER TERM, 1977 Opinion of the Court version of § 303 is in all relevant respects unchanged from the countervailing-duty statute enacted by Congress in 1897/ and that the Secretary—in decisions dating back to 1898—has always taken the position that the nonexcessive remission of an indirect tax is not a bounty or grant within the meaning of the statute.9 On cross-motions for summary judgment, the Customs Court ruled in favor of petitioner and ordered the Secretary to assess countervailing duties on all Japanese consumer elec- Section 5 of the Tariff Act of July 24, 1897, 30 Stat. 205, provided in full: “That whenever any country, dependency, or colony shall pay dr bestow, directly or indirectly, any bounty or grant upon the exportation of any article or merchandise from such country, dependency, or colony, and such article or merchandise is dutiable under the provisions of this Act, then upon the importation of any such article or merchandise into the United States, whether the same shall be imported directly from the country of production or otherwise, and whether such article or merchandise is imported in the same condition as when exported from the country of production or has been changed in condition by remanufacture or otherwise, there shall be levied and paid, in all such cases, in addition to the duties otherwise imposed by this Act, an additional duty equal to the net amount of such bounty or grant, however the same be paid or bestowed. The net amount of all such bounties or grants shall be from time to time ascertained, determined, and declared by the Secretary of the Treasury, who shall make all needful regulations for the identification of such articles and merchandise and for the assessment and collection of such additional duties.” The current version of §303 represents the fifth re-enactment of the 1897 provision without any changes relevant here. Tariff Act of 1909 § 6, 36 Stat. 85; Tariff Act of 1913, § IV (E), 38 Stat. 193; Tariff Act of 1922, § 303, 42 Stat. 935; Tariff Act of 1930, § 303, 46 Stat. 687; Trade Act of 1974, § 331 (a), 88 Stat. 2049. $ There is no dispute here regarding either the nonexcessive nature of the remission or the indirect nature of the tax. Moreover, although the Department did not so state in the notice of final determination, see n. 6, supra, petitioner does not dispute that the Department’s decision in this case was based on its longstanding position that the nonexcessive remission of an indirect tax is not a bounty or grant. ZENITH RADIO CORP. v. UNITED STATES 449 443 Opinion of the Court tronic products specified in petitioner’s complaint. 430 F. Supp. 242 (1977). The court acknowledged the Secretary’s longstanding interpretation of the statute. It concluded, however, that this administrative practice could not be sustained in light of this Court’s decision in Downs v. United States, 187 U. S. 496 (1903), which held that an export bounty had been conferred by a complicated Russian scheme for the regulation of sugar production and sale, involving, among other elements, remission of excise taxes in the event of exportation. On appeal by the Government, the Court of Customs and Patent Appeals, dividing 3-2, reversed the judgment of the Customs Court and remanded for entry of summary judgment in favor of the United States. 64 C. C. P. A. 130, 562 F. 2d 1209 (1977). The majority opinion distinguished Downs on the ground that it did not decide the question of whether non-excessive remission of an indirect tax, standing alone, constitutes a bounty or grant upon exportation. The court then examined the language of § 303 and the legislative history of the 1897 provision and concluded that, “in determining whether a bounty or grant has been conferred, it is the economic result of the foreign government’s action which controls.” 64 C. C. P. A., at 138-139, 562 F. 2d, at 1216. Relying primarily on the “long-continued” and “uniform” administrative practice, id., at 142-143, 146-147, 562 F. 2d, at 1218-1219, 1222-1223, and secondarily on congressional “acquiescence” in this practice through repeated re-enactment of the controlling statutory language, id., at 143-144, 562 F. 2d, at 1220, the court held that interpretation of “bounty or grant” so as not to include a nonexcessive remission of an indirect tax is “a lawfully permissible interpretation of § 303.” Id., at 147, 562 F. 2d, at 1223. We granted certiorari, 434 U. S. 1060 (1978), and we now affirm. 450 OCTOBER TERM, 1977 Opinion of the Court 437U.S. II It is undisputed that the Treasury Department adopted the statutory interpretation at issue here less than a year after passage of the basic countervailing-duty statute in 1897, see T. D. 19321, 1 Synopsis of [Treasury] Decisions 696 (1898), and that the Department has uniformly maintained this position for over 80 years.10 This longstanding and consistent administrative interpretation is entitled to considerable weight. “When faced with a problem of statutory construction, this Court shows great deference to the interpretation given the statute by the officers or agency charged with its administration. ‘To sustain [an agency’s] application of [a] statutory term, we need not find that its construction is the only reasonable one, or even that it is the result we would have reached had the question arisen in the first instance in judicial proceedings.’ ” Udall v. Tallman, 380 U. S. 1, 16 (1965), quoting Unemployment Compensation Comm’n v. Aragon, 329 U. S. 143, 153 (1946). Moreover, an administrative “practice has peculiar weight when it involves a contemporaneous construction of a statute by the [persons] charged with the responsibility of setting its machinery in motion, of making the parts work efficiently and smoothly while they are yet untried and new.” Norwegian Nitrogen Products Co. v. United States, 288 U. S. 294, 315 (1933); see, e. g., Power Reactor Co. v. Electricians, 367 U. S. 396, 408 (1961). The question is thus whether, in light of the normal aids to statutory construction, the Department’s interpretation is sufficiently reasonable” to be accepted by a reviewing court. Train v. Natural Resources Defense Council, 421 U. S. 60, 10 See, e. g., T. D. 19729, 2 Synopsis of Decisions 157 (1898); T. D. 20039, 2 Synopsis of Decisions 534 (1898); T. D. 43634, 56 Treas. Dec. 342 (1929); T. D. 49355, 73 Treas. Dec. 107 (1938). ZENITH RADIO CORP. v. UNITED STATES 451 443 Opinion of the Court 75 (1975). Our examination of the language, the legislative history, and the overall purpose of the 1897 provision persuades us that the Department’s initial construction of the statute was far from unreasonable; and we are unable to find anything in the events subsequent to that time that convinces us that the Department was required to abandon this interpretation. A The language of the 1897 statute evolved out of two earlier countervailing-duty provisions that had been applicable only to sugar imports. The first provision was enacted in 1890, apparently for the purpose of protecting domestic sugar refin-ers from unfair foreign competition; it provided for a fixed countervailing duty on refined sugar imported from countries that “pay, directly or indirectly, a [greater] bounty on the exportation of” refined sugar than on raw sugar. Tariff Act of 1890, H 237, 26 Stat. 584. Although the congressional debates did not focus sharply on the meaning of the word “bounty,” what evidence there is suggests that the term was not intended to encompass the nonexcessive remission of an indirect tax. Thus, one strong supporter of increased protection for American sugar producers heavily criticized the export “bounties” conferred by several European governments, and attached a concise description of “The Bounty Systems in Europe”; both the remarks and the description indicated that the “bounties” consisted of the amounts by which government payments exceeded the excise taxes that had been paid upon the beets from which the sugar was produced. See 21 Cong. Rec. 9529, 9532 (1890) (remarks of Sen. Gibson); id., at 9537 (description). According to the description, for example, French sugar manufacturers paid an “excise tax [of] $97.06 per gross ton[,] [b]ut upon the export of a ton of sugar . . . received back as a drawback $117.60, making a clear bounty of $20.54 per gross ton of sugar exported.” Ibid. 452 437 U. S. OCTOBER TERM, 1977 Opinion of the Court This concept of a “net” bounty—that is, a remission in excess of taxes paid or otherwise due—as the trigger for a countervailing-duty requirement emerged more clearly in the second sugar provision, enacted in 1894. Tariff Act of 1894, fl 182^, 28 Stat. 521. The 1894 statute extended the countervailing-duty requirement to all imported sugar, raw as well as refined, and provided for payment of a fixed duty on all sugar coming from a country which “pays, directly or indirectly, a bounty on the export thereof.” A proviso to the statute made clear, however, that no duties were to be assessed in the event that the “bounty” did not exceed the amount of taxes already paid.11 The author of the 1894 provision, Senator Jones, expressly characterized this difference between the amounts received upon exportation and the amounts already paid in taxes as the “net bounty” on exportation. 26 Cong. Rec. 5705 (1894) (discussing German export bounty system). The 1897 statute greatly expanded upon the coverage of the 1894 provision by making the countervailing-duty requirement applicable to all imported products. Tariff Act of 1897, § 5, 30 Stat. 205, quoted in n. 8, supra. There are strong indications, however, that Congress intended to retain the “net bounty” concept of the 1894 provision as the criterion for determining when a countervailing duty was to be imposed. Although the proviso in the 1894 law was deleted, the 1897 statute did provide for levying of duties equal to the “net amount” of any export bounty or grant. And the legislative 11 The proviso specified that “the importer of sugar produced in a foreign country, the Government of which grants such direct or indirect bounties, may be relieved from this additional duty under such regulations as the Secretary of the Treasury may prescribe, in case said importer produces a certificate of said Government that no indirect bounty has been received upon said sugar in excess of the tax collected upon the beet or cane from which it was produced, and that no direct bounty has been or shall be paid.” 28 Stat. 521 (emphasis added). ZENITH RADIO CORP. v. UNITED STATES 453 443 Opinion of the Court history suggests that this language, in addition to establishing a responsive mechanism for determining the appropriate amount of countervailing duty, was intended to incorporate the prior rule that nonexcessive remission of indirect taxes would not trigger the countervailing-duty requirement at all. There is no question that the prior rule was carried forward in the version of the 1897 statute that originally passed the House. This version did not extend the countervailing-duty requirement to all imports. Instead, it merely modified the 1894 sugar provision so that the amount of the countervailing duty, rather than being fixed, would be “equal to [the export] bounty, or so much thereof as may be in excess of any tax collected by [the foreign] country upon [the] exported [sugar], or upon the beet or cane from which it was produced . . . .” See 30 Cong. Rec. 1634 (1897). The House Report unequivocally stated that the countervailing duty was intended to be “equivalent to the net export bounty paid by any country.” H. R. Rep. No. 1, 55th Cong., 1st Sess., 4-5 (1897) (emphasis supplied). The Senate deleted the House provision from the bill and replaced it with the more general provision that was eventually enacted into law. See 30 Cong. Rec. 1733 (1897) (striking House provision); id., at 2226 (adopting general provision); id., at 2705, 2750 (House agreement to Senate amendment). The debates in the Senate indicate, however, that—aside from extending the coverage of the House provision—the Senate did not intend to change its substance. Senator Allison, the sponsor of the Senate amendment, explained that the House provision was being “stricken from the bill,” because “the same paragraph in substance [is] being inserted [in] section [5], making this countervailing duty apply to all articles instead of to [sugar] alone.” Id., at 1635. See also id., at 1732 (remarks of Sen. White). Senator Allison twice remarked that the countervailing duty that he was proposing was an “imitation” of the one provided in the 1894 statute, 454 OCTOBER TERM, 1977 Opinion of the Court 437U.S. id., at 1719; see id., at 1674, and later in the debates he stated in response to a question as to whether the countervailing duty would be equal to “the whole amount of the export bounty”—that “[the bounty contemplated] is the net bounty, less the taxes and reductions . . . id., at 1721 (answering question from Sen. Vest). An additional indication of the Senate’s intent can be found in the extended discussion of the effect that the statute would have with respect to German sugar exports. Time after time the amount of the German “bounty”—and, correspondingly, the amount of the countervailing duty that would be imposed under the statute—was stated to be 380 per 100 pounds of refined sugar, and 270 per 100 pounds of raw sugar. See, e. g., id., at 1650 (remarks of Sens, Allison, Vest, and Caffery),’ 1658 (Sens. Allison and Jones), 1680 (Sen. Jones), 1719 (Sens. Allison and Lindsay), 1729 (Sen. Caffery), 2823-2824 (Sens. Aldrich and Jones). These figures were supplied by the Treasury Department itself, see id., at 1719 (remarks of Sen. Allison), 1722 (letter from Treasury Department to Sen. Caf-fery), and were utilized by both proponents and opponents of the measure. And yet it was frequently acknowledged during the debates that Germany exempted sugar exports from its domestic consumption tax of $2.16 per 100 pounds, an amount far in excess of the 380 and 270 figures. See, e. g., id., at 1646 (remarks of Sen. Vest), 1651 (Sen. Caffery), 1697 (same), 2205 (same). Had the Senators considered the mere remission of an indirect tax to be a “bounty,” it seems unlikely that they would have stated that the German “bounties” were only 380 and 270 per 100 pounds.12 Especially in The figures of 380 and 270 per 100 pounds apparently represented the amount of direct bounty paid upon exportation. See, e. g., 30 Cong. Rec. 1722 (1897) (letter from Treasury Department). Petitioner argues that the Senate musk have intended the term “bounty” to include nonexcessive remissions of indirect taxes, since Germany collected a tax on the output of sugar factories that was not remitted upon exporta- ZENITH RADIO CORP. v. UNITED STATES 455 443 Opinion of the Court light of the strong opposition to countervailing duties even of the magnitude of 380 and 270, see, e. g., id., at 1719 (remarks of Sen. Lindsay), 2203-2205 (remarks of Sen. Gray), it seems reasonable to infer that Congress did not intend to impose countervailing duties of many times this magnitude. B Regardless of whether this legislative history absolutely compelled the Secretary to interpret “bounty or grant” so as not to encompass any nonexcessive remission of an indirect tax, there can be no doubt that such a construction was reasonable in light of the statutory purpose. Cf. Mourning v. Family Publications Service, Inc., 411 U. S. 356, 374 (1973). This purpose is relatively clear from the face of the statute and is confirmed by the congressional debates: The counter-tion and yet was not subtracted from the figures of 380 and 270 cited as the “bounties” paid by Germany. The sole evidence cited by petitioner to show that Germany in fact collected such a tax is an exhibit to the testimony of a single witness during hearings conducted by the House in 1896. See Tariff Hearings before the House Committee on Ways and Means, 54th Cong., 2d Sess., 617-618 (1896-1897). We have been unable to find any references to this tax anywhere in the Senate debates; moreover, to the extent that anyone contemplated the existence of German taxes that were not remitted upon exportation, the assumption appears to have been that they would be deducted from the 380 and 270 figures in determining the net amount of the bounty to be countervailed. The following exchange between Senators Allison and Vest is illustrative: “Mr. VEST. What ... is the amount of export bounty, taking out taxes, etc., granted by Germany? “Mr. ALLISON. ... Of course it can not exceed three-eighths of a cent a pound—thirty-eight one-hundredths on refined sugar—nor can it exceed twenty-seven one-hundredths upon raw sugar. But it may be very much less.” 30 Cong. Rec. 1721 (1897). We note in any event that the amount of the tax cited by petitioner was less than 20 per 100 pounds, see Tariff Hearings, supra, at 617, whereas the consumption tax—which concededly was remitted upon exportation and yet not added to the figures of 380 and 270—was in the vicinity of $2.16 per 100 pounds. 456 OCTOBER TERM, 1977 Opinion of the Court 437U.S. vailing duty was intended to offset the unfair competitive advantage that foreign producers would otherwise enjoy from export subsidies paid by their governments. See, e. g., 30 Cong. Rec. 1674 (remarks of Sen. Allison), 2205 (Sen. Caffery), 2225 (Sen. Lindsay) (1897). The Treasury Department was well positioned to establish rules of decision that would accurately carry out this purpose, particularly since it had contributed the very figures relied upon by Congress in enacting the statute. See Zuber v. Allen, 396 U S 168 192 (1969). In deciding in 1898 that a nonexcessive remission of indirect taxes did not result in the type of competitive advantage that Congress intended to counteract, the Department was clearly acting in accordance with the shared assumptions of the day as to the fairness and economic effect of that practice. The theory underlying the Department’s position was that a foreign country’s remission of indirect taxes did not constitute subsidization of that country’s exports. Rather, such remis-sion was viewed as a reasonable measure for avoiding double taxation of exports—once by the foreign country and once upon sale in this country. As explained in a recent study prepared by the Department for the Senate Committee on Finance: “ [The Department’s construction was] based on the principle that, since exports are not consumed in the country of production, they should not be subject to consumption taxes in that country. The theory has been that the application of countervailing duties to the rebate of consumption [and other indirect] taxes would have the effect of double taxation of the product, since the United States would not only impose its own indirect taxes, such as Federal and state excise taxes and state and local sales taxes, but would also collect, through the use of the countervailing duty, the indirect tax imposed by the ZENITH RADIO CORP. v. UNITED STATES 457 443 Opinion of the Court exporting country on domestically consumed goods.” Senate Committee on Finance, Executive Branch GATT Studies, 93d Cong., 2d Sess., 17-18 (1974). This intuitively appealing principle regarding double taxation had been widely accepted both in this country and abroad for many years prior to enactment of the 1897 statute. See, e. g., Act of July 4, 1789, § 3, 1 Stat. 26 (remission of import duties upon exportation of products); 4 Works and Correspondence of D. Ricardo 216-217 (pamphlets and papers first published in 1822); A. Smith, An Inquiry Into the Nature and Causes of the Wealth of Nations, Book Four, ch. IV (1776). C The Secretary’s interpretation of the countervailing-duty statute is as permissible today as it was in 1898. The statute has been re-enacted five times by Congress without any modification of the relevant language, see n. 8, supra, and, whether or not Congress can be said to have “acquiesced” in the administrative practice, it certainly has not acted to change it. At the same time, the Secretary’s position has been incorporated into the General Agreement on Tariffs and Trade (GATT),13 which is followed by every major trading nation in the world; foreign tax systems as well as private expectations thus have been built on the assumption that countervailing duties would not be imposed on nonexcessive remissions of indirect taxes. In light of these substantial reliance interests, the longstanding administrative construction of the statute should “not be 13 Article VI (3) of the GATT, adopted in 1947, 61 Stat. A24, provides that “[n]o product . . . imported into the territory of any other contracting party shall be subject to . . . countervailing duty by reason of the exemption of such product from . . . taxes borne by the like product when destined for consumption in the country of origin or exportation, or by reason of the refund of such . . . taxes.” The Government does not contend that the GATT provision would supersede § 303 in the event of conflict between the two. Brief for United States 19 n. 11. 458 OCTOBER TERM, 1977 Opinion of the Court 437 U. g. disturbed except for cogent reasons.” McLaren v. Fleischer, 256 U. 8. 477, 481 (1921); see Udall v. Tallman, 380 U S' at 18. ’ Aside from the contention, discussed in Part III, infra, that the Department’s construction is inconsistent with this Court’s decisions, petitioner’s sole argument is that the Department’s position is premised on false economic assumptions that should be rejected by the courts. In particular, petitioner points to ‘modern” economic theory suggesting that remission of indirect taxes may create an incentive to export in some circumstances, and to recent criticism of the GATT rules as favoring producers in countries that rely more heavily on indirect than on direct taxes.14 But, even assuming that these arguments are at all relevant in view of the legislative history of the 1897 provision and the longstanding administrative construction of the statute, they do not demonstrate the unreasonableness of the Secretary’s current position. Even “modem” economists do not agree on the ultimate economic effect of remitting indirect taxes, and—given the present state of economic knowledge—it may be difficult, if not impossible, to measure the precise effect in any particular case. See, e. g., Executive Branch GATT Studies, supra, at 13-14, 17 ‘ Marks & Malmgren, Negotiating Nontariff Distortions to Trade, 7 L. & Policy in Int’l Bus. 351 (1975). More fundamentally, as the Senate Committee with responsibility in this 14 See e g., Marks & Malmgren, Negotiating Nontariff Distortions to Trade, 7 L. & Policy in Int’l Bus. 327, 351-355 (1975); The United States Submission on Border Tax Adjustments to Working Party No. 4 of the Council on Border Tax Adjustments, Organisation for Economic Cooperation and Development (1966), reprinted in App. 93-116; Paper Submitted by John R. Petty, Assn’t Sec’y of the Treasury, Twenty-First Annual Conference of the Canadian Tax Foundation (1968), reprinted in App. 117-138. Both the Secretary and GATT apparently consider remis-slons of direct taxes (e. g., income taxes) to be countervailable export subsidies. See Brief for United States 18 n. 10, 37-38; GATT Basic Instruments and Selected Documents 186-187 (Supp. 1961). ZENITH RADIO CORP. v. UNITED STATES 459 443 Opinion of the Court area recently stated, “the issues involved in applying the countervailing duty law are complex, and . . . internationally, there is [a] lack of any satisfactory agreement on what constitutes a fair, as opposed to an ‘unfair,’ subsidy.” S. Rep. No. 93-1298, p. 183 (1974). In this situation, it is not the task of the judiciary to substitute its views as to fairness and economic effect for those of the Secretary. Ill Notwithstanding all of the foregoing considerations, this would be a very different case if, as petitioner contends, the Secretary’s practice were contrary to this Court’s decision in Downs v. United States, 187 U. S. 496 (1903).15 Upon close examination of the admittedly opaque opinion in that case, however, we do not believe that Downs is controlling on the question presented here. The Russian sugar laws at issue in Downs were, as the Court noted, “very complicated.” Id., at 502. Much of the Court’s opinion was devoted to an exposition of these provisions, see id., at 502-512, but for present purposes only two features are relevant: (1) excise taxes imposed on sugar sales within Russia were remitted on exports; and (2) the exporter received, in addition, a certificate entitling its bearer to sell an amount of sugar in Russia, equal to the quantity exported, without paying the full excise tax otherwise due. This certificate was transferable and had a substantial market value related to the amount of tax forgiveness that it carried with it. 15 Petitioner also relies on language in G. S. Nicholas & Co. v. United States, 249 U. S. 34 (1919), suggesting that the countervailing-duty statute was intended to be read broadly. See id., at 39-41. As petitioner concedes, however, the only question before the Court in that case was whether a direct bounty on exportation of liquor from Great Britain was a “bounty or grant” within the meaning of the statute, see Brief for Petitioner 16-17, and the Court did not address the question of whether nonexcessive remission of an indirect tax fell within the statute. 460 OCTOBER TERM, 1977 Opinion of the Court 437U.S. The Secretary, following the same interpretation of the statute that he followed here, imposed a countervailing duty based on the value of the certificates alone, and not on the excise taxes remitted on the exports themselves.16 Downs, the importer, sought review, claiming that the Russian system did not confer any countervailable bounty or grant within the meaning of the 1897 statute. He did not otherwise challenge the amount of the duty assessed by the Secretary.17 The issue as it came before this Court, therefore, was whether a nonexcessive remission of an indirect tax, together with the granting of an additional benefit represented by the value of the certificate, constituted a “bounty or grant.” Since the amount of the bounty was not in question, neither the parties nor this Court focused carefully on the distinction between remission of the excise tax and conferral of the certificate. Petitioner argues, however, that certain broad language in the Court’s opinion suggests that mere remission of a tax, even if nonexcessive, must be considered a bounty or grant within the meaning of the statute. Petitioner relies in particular on the following language: “The details of this elaborate procedure for the production, sale, taxation and exportation of Russian sugar are of much less importance than the two facts which appear clearly through this maze of regulations, viz.: that no sugar is permitted to be sold in Russia that does not pay an excise tax of R. 1.75 per pood, and that sugar exported pays no tax at all. . . . When a tax is imposed 16 See Memorandum from the Secretary of the Treasury (1901), reprinted in App. 49-51; T. D. 20407, 2 Synopsis of Decisions 996, 997-998 (1898); T. D. 22814, 4 Treas. Dec. 184 (1901); Downs v. United States, 113 F 144, 145 (CA4 1902). 17 In rejecting Downs’ claim, both the United States Board of General Appraisers and the Fourth Circuit Court of Appeals identified the “bounty” as residing in the value of the certificates granted upon exportation. See T« D. 22984, 4 Treas. Dec. 405, 410—411, 413 (1901); Downs v. United States, supra, at 145. ZENITH RADIO CORP. v. UNITED STATES 461 443 Opinion of the Court upon all sugar produced, but is remitted upon all sugar exported, then, by whatever process, or in whatever manner, or under whatever name it is disguised, it is a bounty upon exportation.” Id., at 515. This passage is inconsistent with both preceding and subsequent language which suggests that the Court understood the “bounty” to reside in the value of the certificates. At one point the Court stated that “[t]he amount [the exporter] receives for his export certificate [on the market], say, R. 1.25, is the exact amount of the bounty he receives upon exportation . . . .” Ibid™ And the Court in conclusion specifically endorsed the Fourth Circuit’s holding to the same effect, see n. 17, supra: “[T]he Circuit Court of Appeals found: ‘That the Russian exporter of sugar obtained from his government a certificate, solely because of such exportation, which is worth in the open market of that country from R. 1.25 to R. 1.64 per pood, or from 1.8 to 2.35 cents per pound. Therefore we hold that the government of Russia does secure to the exporter of that country, as the inevitable result of its action, a money reward or gratuity whenever he exports sugar from Russia.’ We all concur in this expression of opinion.” 187 U. S., at 516. Given this other language, we cannot read for its broadest implications the passage on which petitioner relies. In our view the passage does no more than establish the proposition 18 The Court also noted that “[i]t is practically admitted in this case that a bounty equal to the value of [the] certificates is paid by the Russian government, and the main argument of the petitioner is addressed to the proposition that this bounty is paid, not upon exportation, but upon production.” 187 U. S., at 512. This latter argument was based on the fact that the 1897 statute covered only bounties on exportation and not those on production. In 1922, Congress amended the statute to cover bounties on production and manufacture as well as exportation. Tariff Act of 1922, supra, n. 8. 462 437 U. S. OCTOBER TERM, 1977 Opinion of the Court that an excessive remission of taxes—there, the combination of the exemption with the certificates—is an export bounty within the meaning of the statute. As the court below noted, “ ‘ [i]t is a maxim, not to be disregarded, that general expressions, in every opinion, are to be taken in connection with the case in which those expressions are used.’ ” 64 C. C. P. A., at 134, 562 F. 2d, at 1213, quoting Cohens v. Virginia, 6 Wheat. 264, 399 (1821). No one argued in Downs that a nonexcessive remission of taxes, standing alone, would have constituted a bounty on exportation, and indeed that issue was not presented on the facts of the case. It must also be remembered, of course, that the Court did affirm the Secretary’s decision, and that decision rested on the conclusion that a bounty had been paid only to the extent that the remission exceeded the taxes otherwise due. In light of all these circumstances, the isolated statement in Downs relied upon by petitioner cannot be dispositive here. The judgment of the Court of Customs and Patent Appeals is, accordingly, Affirmed. COOPERS & LYBRAND v. LIVESAY 463 Syllabus COOPERS & LYBRAND v. LIVESAY et al. CERTIORARI TO THE UNITED STATES COURT OF APPEALS FOR THE EIGHTH CIRCUIT No. 76-1836. Argued March 22, 1978—Decided June 21, 1978 Respondents, who had purchased securities in reliance on a prospectus, brought this action on behalf of themselves and a class of similarly situated purchasers, alleging that petitioner accounting firm had violated the federal securities laws. The District Court first certified the action as a class action under Fed. Rule Civ. Proc. 23, and then, after further proceedings, decertified the class. Respondents then filed a notice of appeal pursuant to 28 U. S. C. § 1291, under which courts of appeals have jurisdiction of appeals from all “final decisions” of the district courts except where a direct review may be had in the Supreme Court. After examining the amount of respondents’ claims in relation to their financial resources and the probable cost of the litigation, the Court of Appeals concluded that they would not pursue their claims individually. On the basis of the “death knell” doctrine (which assumes that without the incentive of a possible group recovery the individual plaintiff may find it economically imprudent to pursue his lawsuit to a final judgment and then seek appellate review of an adverse class determination), the Court of Appeals held that it had jurisdiction to hear the appeal, and reversed the District Court’s order decertifying the class. Respondents contend in this Court that an order denying class certification is appealable under both the “death knell” doctrine and the “collateral order” exception articulated in Cohen v. Beneficial Industrial Loan Corp., 337 U. S. 541. Held: 1. The collateral order” exception does not apply to a prejudgment order denying class certification because such an order is subject to revision in the District Court, Fed. Rule Civ. Proc. 23 (c)(1); involves considerations that are “enmeshed in the factual and legal issues comprising the plaintiff’s cause of action,” Mercantile Nat. Bank v. Langdeau, 371 U. S. 555, 558; and is subject to effective review after final judgment at the behest of the named plaintiff or intervening class members. United Airlines, Inc. v. McDonald, 432 U. S. 385 Pp 468-469. 2. Nor does the “death knell” doctrine support appellate jurisdiction of a prejudgment order denying class certification. Pp. 469-476. (a) The formulation of an appealability rule that turns on the 464 437 U. S. OCTOBER TERM, 1977 Opinion of the Court amount of the plaintiff’s claim is plainly a legislative, not a judicial, function. Pp. 472-473. (b) The alternative approach to the “death knell” rule that is based on a thorough study of the possible impact of the class order on the fate of the litigation would have a seriously debilitating effect on the administration of justice. The district court would have to take evidence, entertain argument, and make findings, which the court of appeals would have to review simply to determine whether a discretionary class determination is subject to appellate review, with the possibility of remand for further factual development. Further appeals from adverse rulings on other grounds could likewise be anticipated Pp. 473-474. (c) Perhaps the principal vice of the doctrine is that it authorizes indiscriminate interlocutory review of the trial judge’s decisions, circumventing restrictions imposed by the Interlocutory Appeals Act of 1958. Pp. 474-475. (d) The doctrine favors only plaintiffs even though the class issue will often be critically important to defendants as well. P. 476. (e) Allowing appeals as a matter of right from nonfinal orders that turn on the facts of a particular case thrusts appellate courts indis-criimnately into the trial process, thus defeating a vital purpose of the final-judgment rule of maintaining the appropriate relationship between the respective courts. P. 476. 550 F. 2d 1106, reversed. Stevens, J., delivered the opinion for a unanimous Court. Thomas C. Walsh argued the cause for petitioner. With him on the briefs were Veryl L. Riddle, John J. Hennelly, Jr., and Harris J. Amhowitz. Melvyn I. Weiss argued the cause for respondents. With him on the brief were Lawrence Milbery, Jared Specthrie, and Richard L. Ross. Mr. Justice Stevens delivered the opinion of the Court. The question in this case is whether a district court’s determination that an action may not be maintained as a class action pursuant to Fed. Rule Civ. Proc. 23 is a “final decision” COOPERS & LYBRAND v. LIVESAY 465 463 Opinion of the Court within the meaning of 28 U. S. C. § 12911 and therefore appealable as a matter of right. Because there is a conflict in the Circuits over this issue,2 we granted certiorari and now hold that such an order is not appealable under § 1291. Petitioner, Coopers & Lybrand, is an accounting firm that certified the financial statements in a prospectus issued in connection with a 1972 public offering of securities in Punta Gorda Isles for an aggregate price of over $18 million. Respondents purchased securities in reliance on that prospectus. In its next annual report to shareholders, Punta Gorda restated the earnings that had been reported in the prospectus for 1970 and 1971 by writing down its net income for each year by over $1 million. Thereafter, respondents sold their Punta Gorda securities and sustained a loss of $2,650 on their investment. Respondents filed this action on behalf of themselves and a class of similarly situated purchasers. They alleged that petitioner and other defendants3 had violated various sections of 1 “The courts of appeals shall have jurisdiction of appeals from all final decisions of the district courts of the United States . . . except where a direct review may be had in the Supreme Court.” 2 Compare Hackett v. General Host Corp., 455 F. 2d 618 (CA3 1972), cert, denied, 407 U. S. 925; King n. Kansas City Southern Industries, Inc., 479 F. 2d 1259 (CA7 1973) (holding that such an order is not immediately appealable under § 1291), with Hartmann v. Scott, 488 F. 2d 1215 (CA8 1973); Ott n. Speedwriting Pub. Co., 518 F. 2d 1143 (CA6 1975) ; Eisen n. Carlisle & Jacquelin, 370 F. 2d 119 (CA2 1966), cert, denied, 386 U. S. 1035 (holding that such an order is immediately appealable under §1291). 3 The other defendants, Punta Gorda and several of its officers and directors, also filed a petition for writ of certiorari in this Court. Punta Gorda Isles, Inc. v. Livesay, No. 76-1837. After we granted certiorari in this case and No. 76-1837, 434 U. S. 954, the parties entered into a tentative settlement agreement. Respondents and petitioners in No. 76-1837 agreed to dismiss that petition; petitioner in this case, however, did not stipulate to dismissal of its petition. In view of the tentative nature of the settlement, this case is not moot. 466 OCTOBER TERM, 1977 Opinion of the Court 437U.S. the Securities Act of 1933 and the Securities Exchange Act of 1934.4 The District Court first certified, and then, after further proceedings, decertified the class. Respondents did not request the District Court to certify its order for interlocutory review under 28 U. S. C. § 1292 (b).5 Rather, they filed a notice of appeal pursuant to § 1291.6 The Court of Appeals regarded its appellate jurisdiction as depending on whether the decertification order had sounded the ‘death knell” of the action. After examining the amount of respondents claims in relation to their financial resources and the probable cost of the litigation, the court concluded that they would not pursue their claims individually.7 The Court 4 §§ 11, 12 (2) and 17 (b) of the Securities Act of 1933, 15 U. S. C. §§ 77k, 77Z (2), and 77q (b) (1976 ed.), and § 10 (b) of the Securities Exchange Act of 1934, 15 U. S. C. § 78j (b) (1976 ed.). 5 Section 1292 (b) provides: When a district judge, in making in a civil action an order not otherwise appealable under this section, shall be of the opinion that such order mvolves a controlling question of law as to which there is substantial ground for difference of opinion and that an immediate appeal from the order may materially advance the ultimate termination of the litigation, he shall so state in writing in such order. The Court of Appeals may thereupon, in its discretion, permit an appeal to be taken from such order, if application is made to it within ten days after the entry of the order: Provided, however, That application for an appeal hereunder shall not stay proceedings in the district court unless the district judge or the Court of Appeals or a judge thereof shall so order.” 6 Respondents also petitioned for a writ of mandamus directing the District Court to recertify the class. Since the Court of Appeals accepted appellate jurisdiction, it dismissed the petition for a writ of mandamus. Plaintiffs, both of whom are employed, have an aggregate yearly gross income of $26,000. Their total net worth is approximately $75,000, but only $4,000 of this sum is in cash. The remainder consists of equity in their home and investments. f Dumber 1974 plaintiffs had already incurred expenses in excess of $1,200 in connection with this lawsuit. Plaintiffs’ new counsel has estimated expenses of this lawsuit to be $15,000. The nature of this case will require extensive discovery, much of which must take place in Florida, COOPERS & LYBRAND v. LI VESA Y Opinion of the Court 467 463 of Appeals therefore held that it had jurisdiction to hear the appeal and, on the merits, reversed the order decertifying the class. Livesay v. Punta Gorda Isles, Inc., 550 F. 2d 1106. Federal appellate jurisdiction generally depends on the existence of a decision by the District Court that “ends the litigation on the merits and leaves nothing for the court to do but execute the judgment.” Catlin v. United States, 324 U. S. 229, 233.8 An order refusing to certify, or decertifying, a class does not of its own force terminate the entire litigation because the plaintiff is free to1 proceed on his individual claim. Such an order is appealable, therefore, only if it comes within an appropriate exception to the final-judgment rule. In this where most defendants reside. Moreover, the allegations regarding the prospectus and financial statements will likely require expert testimony at trial. “After considering all the relevant information in the record, we are convinced that plaintiffs have sustained their burden of showing that they will not pursue their individual claim if the decertification order stands. Although plaintiffs’ total net worth could absorb the cost of this litigation, fit [takes] no great understanding of the mysteries of high finance to m’ake obvious the futility of spending a thousand dollars to get a thousand dollars—or even less.’ Douglas, Protective Committees in Railroad Reorganizations, 47 Harv. L. Rev. 565, 567 (1934). We conclude we have jurisdiction to hear the appeal.” Livesay v. Punta Gorda Isles, Inc 550 F. 2d 1106, 1109-1110. 8 For a unanimous Court in Cobbledick v. United States, 309 U. S. 323, 325, Mr. Justice Frankfurter wrote: Since the right to a j’udgment from more than one court is a matter of grace and not a necessary ingredient of justice, Congress from the very beginning has, by forbidding piecemeal disposition on appeal of what for practical purposes is a single controversy, set itself against enfeebling judicial administration. Thereby is avoided the obstruction to just claims that would come from permitting the harassment and cost of a succession of separate appeals from the various rulings to which a litigation may give rise, from its initiation to entry of judgment. To be effective, judicial administration must not be leaden-footed. Its momentum would be arrested by permitting separate reviews of the component elements in a unified cause.” 468 OCTOBER TERM, 1977 Opinion of the Court 437U.S. case respondents rely on the “collateral order” exception articulated by this Court in Cohen v. Beneficial Industrial Loan Corp., 337 U.K 541, and on the “death knell” doctrine adopted by several Circuits to determine the appealability of orders denying class certification. I In Cohen, the District Court refused to order the plaintiff in a stockholder’s derivative action to post the security for costs required by a New Jersey statute. The defendant sought immediate review of the question whether the state statute applied to derivative suits in federal court. This Court noted that the purpose of the finality requirement “is to combine in one review all stages of the proceeding that effectively may be reviewed and corrected if and when final judgment results. Id., at 546. Because immediate review of the District Court’s order was consistent with this purpose, the Court held it appealable as a “final decision” under § 1291. The ruling had settled conclusively the corporation’s claim that it was entitled by state law to require the shareholder to post security for costs . . . [and] concerned a collateral matter that could not be reviewed effectively on appeal from the final judgment.” 9 To come within the “small class” of decisions excepted from the final-judgment rule by Cohen, the order must conclusively determine the disputed question, resolve an important issue completely separate from the merits of the action, and be effectively unreviewable on appeal from a final judgment.10 Abney v. United States, 431 U. S. 651, 658; United States v. 9 Eisen v. Carlisle & Jacquelin, 417 U. S. 156,171. 10 As the Court summarized the rule in Cohen: “This decision appears to fall in that small class which finally determine claims of right separable from, and collateral to, rights asserted in the action, too important to be denied review and too independent of the cause itself to require that appellate consideration be deferred until the whole case is adjudicated.” 337 U. S., at 546. COOPERS & LYBRAND v. LIVESAY 469 463 Opinion of the Court MacDonald, 435 U. S. 850, 855. An order passing on a request for class certification does not fall in that category. First, such an order is subject to revision in the District Court. Fed. Rule Civ. Proc. 23 (c)(1).11 Second, the class determination generally involves considerations that are “enmeshed in the factual and legal issues comprising the plaintiff’s cause of action.” Mercantile Nat. Bank v. Langdeau, 371 U. S. 555, 558.12 Finally, an order denying class certification is subject to effective review after final judgment at the behest of the named plaintiff or intervening class members. United Airlines, Inc. v. McDonald, 432 U. S. 385. For these reasons, as the Courts of Appeals have consistently recognized,13 the collateral-order doctrine is not applicable to the kind of order involved in this case. II Several Circuits, including the Court of Appeals in this case, have held that an order denying class certification is appealable if it is likely to sound the “death knell” of the litigation.14 The “death knell” doctrine assumes that without the incentive of a possible group recovery the individual plaintiff may find it economically imprudent to pursue his lawsuit to a final 11 The Rule provides that an order involving class status may be “altered or amended before the decision on the merits.” Thus, a district court’s order denying or granting class status is inherently tentative. 12 Evaluation of many of the questions entering into determination of class action questions is intimately involved with the merits of the claims The typicality of the representative’s claims or defenses, the adequacy of the representative, and the presence of common questions of law or fact are obvious examples. The more complex determinations required in Rule 23 (b) (3) class actions entail even greater entanglement with the merits ... .” 15 C. Wright, A. Miller, & E. Cooper, Federal Practice and Procedure §3911, p. 485 n. 45 (1976). 13 See, e. g., King v. Kansas City Southern Industries, Inc., 479 F. 2d 1259 (CA7 1973); Williams v. Mumford, 167 U. S. App. D. C. 125, 511 F. 2d 363 (1975), cert, denied, 423 U. S. 828. 14 See n. 2, supra. 470 OCTOBER TERM, 1977 Opinion of the Court. 437U.S. judgment and then seek appellate review of an adverse class determination. Without questioning this assumption, we hold that orders relating to class certification are not independently appealable under § 1291 prior to judgment. In addressing the question whether the “death knell” doctrine supports mandatory appellate jurisdiction of orders refusing to certify class actions, the parties have devoted a portion of their argument to the desirability of the small-claim class action. Petitioner’s opposition to the doctrine is based in part on criticism of the class action as a vexatious kind of litigation. Respondents, on the other hand, argue that the class action serves a vital public interest and, therefore, special rules of appellate review are necessary to ensure that district judges are subject to adequate supervision and control. Such policy arguments, though proper for legislative consideration, are irrelevant to the issue we must decide. There are special rules relating to class actions and, to that extent, they are a special kind of litigation. Those rules do not, however, contain any unique provisions governing appeals. The appealability of any order entered in a class action is determined by the same standards that govern appealability in other types of litigation. Thus, if the “death knell” doctrine has merit, it would apply equally to the many interlocutory orders in ordinary litigation—rulings on discovery, on venue, on summary judgment^-that may have such tactical economic significance that a defeat is tantamount to a “death knell” for the entire case. Though a refusal to certify a class is inherently interlocutory, it may induce a plaintiff to abandon his individual claim, On the other hand, the litigation will often survive an adverse class determination. What effect the economic disincentives created by an interlocutory order may have on the fate of any litigation will depend on a variety of factors.15 Under the 15 E- Q-> the plaintiff’s resources; the size of his claim and his subjective willingness to finance prosecution of the claim; the probable cost of the COOPERS & LYBRAND v. LIVESAY 471 463 Opinion of the Court “death knell” doctrine, appealability turns on the court’s perception of that impact in the individual case. Thus, if the court believes that the plaintiff has adequate incentive to continue, the order is considered interlocutory; but if the court concludes that the ruling, as a practical matter, makes further litigation improbable, it is considered an appealable final decision. The finality requirement in § 1291 evinces a legislative judgment that “ [restricting appellate review to ‘final decisions’ prevents the debilitating effect on judicial administration caused by piecemeal appeal disposition of what is, in practical consequence, but a single controversy.” Eisen v. Carlisle & Jacquelin, 417 U. S. 156, 170. Although a rigid insistence on technical finality would sometimes conflict with the purposes of the statute, Cohen v. Beneficial Industrial Loan Corp., 337 U. S. 541, even adherents of the “death knell” doctrine acknowledge that a refusal to certify a class does not fall in that limited category of orders which, though nonfinal, may be appealed without undermining the policies served by the general rule. It is undisputed that allowing an appeal from such an order in the ordinary case would run “directly contrary to the policy of the final judgment rule embodied in 28 U. S. C. § 1291 and the sound reasons for it. . . 16 Yet several Courts of Appeals have sought to identify on a case-by-case basis those few interlocutory orders which, when viewed from the standpoint of economic prudence, may induce a plaintiff to abandon the litigation. These orders, then, become appealable as a matter of right. In administering the “death knell” rule, the courts have used two quite different methods of identifying an appealable class ruling. Some courts have determined their jurisdiction litigation and the possibility of joining others who will share that cost; and the prospect of prevailing on the merits and reversing an order denying class certification. 16 Korn v. Franchard Corp., 443 F. 2d 1301, 1305 (CA2 1971). 472 437 U. S. OCTOBER TERM, 1977 Opinion of the Court by simply comparing the claims of the named plaintiffs with an arbitrarily selected jurisdictional amount;17 others have undertaken a thorough study of the possible impact of the class order on the fate of the litigation before determining their jurisdiction. Especially when consideration is given to the consequences of applying these tests to pretrial orders entered in non-class-action litigation, it becomes apparent that neither provides an acceptable basis for the exercise of appellate jurisdiction. The formulation of an appealability rule that turns on the amount of the plaintiff’s claim is plainly a legislative, not a judicial, function. While Congress could grant an appeal of right to those whose claims fall below a specific amount in controversy, it has not done so. Rather, it has made “finality” the test of appealability. Without a legislative prescription, an amount-in-controversy rule is necessarily an arbitrary measure of finality because it ignores the variables that inform a litigant’s decision to proceed, or not to proceed, in the face of an adverse class ruling.18 Moreover, if the jurisdictional 17 Thus, orders denying class certification have been held nonappealable because the plaintiffs alleged damages in the $3,OO(M8,OOO range. Shayne v. Madison Square Garden, 491 F. 2d 397 (CA2 1974); Korn v. Franchard Corp., supra; Gosa v. Securities Inv. Co., 449 F. 2d 1330 (CA5 1971); Domaco Venture Capital Fund v. Teltronics Services, Inc., 551 F. 2d 508 (CA2 1977). Smaller claims, however, have been held sufficient to support appellate jurisdiction in other cases. See, e. g., Green v. Wolf Corp., 406 F. 2d 291 (CA2 1968), cert, denied, 395 U. S. 977. 18 See n. 15, supra. Thus, it is not at all clear that the prospect of recovering $3,000 would provide more incentive to sustain complex litigation against corporate defendants than the prospect of recovering $1,000. Yet the amount-in-controversy test allows an appeal in the latter case but not in the former. Compare Green v. Wolf Corp., supra, at 295 n. 6, with Gosa v. Securities Inv. Co., supra. The arbitrariness of this approach is exacerbated by the fact that the Courts of Appeals have not settled on a specific jurisdictional amount; rather, they have simply determined on an ad hoc basis whether the plaintiff’s claim is too small to warrant individual prosecution. COOPERS & LYBRAND v. LIVESAY 473 463 Opinion of the Court amount is to be measured by the aggregated claims of the named plaintiffs, appellate jurisdiction may turn on the joinder decisions of counsel rather than the finality of the order.19 While slightly less arbitrary, the alternative approach to the “death knell” rule would have a serious debilitating effect on the administration of justice. It requires class-action plaintiffs to build a record in the trial court that contains evidence of those factors deemed relevant to the “death knell” issue and district judges to make appropriate findings.20 And one Court of Appeals has even required that the factual inquiry be extended to all members of the class because the policy against interlocutory appeals can be easily circumvented by joining “only those whose individual claims would not warrant the cost of separate litigation”;21 to avoid this possibility, the named plaintiff is required to prove that no member of the purported class has a claim that warrants individual litigation. A threshold inquiry of this kind may, it is true, identify some orders that would truly end the litigation prior to final judgment; allowing an immediate appeal from those orders may enhance the quality of justice afforded a few litigants. But this incremental benefit is outweighed by the impact of such an individualized jurisdictional inquiry on the judicial system’s overall capacity to administer justice. The potential waste of judicial resources is plain. The district court must take evidence, entertain argument, and make findings; and the court of appeals must review that record and those findings simply to determine whether a discretionary class determination is subject to appellate review. And if the record provides an inadequate basis for this determination, a 19 Cf. Milberg v. Western Pacific R. Co., 443 F. 2d 1301 (CA2 1971). 20 See, e. g., Hooley n. Red Carpet Corp., 549 F. 2d 643 (CA9 1977) ; Ott v. Speedwriting Pub. Co., 518 F. 2d 1143 (CA6 1975). 21 Hooley v. Red Carpet Corp., supra, at 645. 474 437 U. S. OCTOBER TERM, 1977 Opinion of the Court remand for further factual development may be required.22 Moreover, even if the court makes a “death knell” finding and reviews the class-designation order on the merits, there is no assurance that the trial process will not again be disrupted by interlocutory review. For even if a ruling that the plaintiff does not adequately represent the class is reversed on appeal, the district court may still refuse to certify the class on the ground that, for example, common questions of law or fact do not predominate. Under the “death knell” theory, plaintiff would again be entitled to an appeal as a matter of right pursuant to § 1291. And since other kinds of interlocutory orders may also create the risk of a premature demise, the potential for multiple appeals in every complex case is apparent and serious. Perhaps the principal vice of the “death knell” doctrine is that it authorizes indiscriminate interlocutory review of decisions made by the trial judge. The Interlocutory Appeals Act of 1958, 28 U. S. C. § 1292 (b),23 was enacted to meet the recognized need for prompt review of certain nonfinal orders. However, Congress carefully confined the availability of such review. Nonfinal orders could never be appealed as a matter of right. Moreover, the discretionary power to permit an interlocutory appeal is not, in the first instance, vested in the courts of appeals.24 A party seeking review of a nonfinal order must first obtain the consent of the trial judge. This screening procedure serves the dual purpose of ensuring that such review will be confined to appropriate cases and avoiding time-consuming jurisdictional determinations in the court 22 See, e. g., Jelfo v. Hickok Mfg. Co., 531 F. 2d 680, 681 (CA2 1976). 23 See n. 5, supra. 24 Thus, Congress rejected the notion that the courts of appeals should be free to entertain interlocutory appeals whenever, in their discretion, it appeared necessary to avoid unfairness in the particular case. H. R. Rep, No. 1667, 85th Cong, 2d Sess, 4-6 (1958); Note, Interlocutory Appeal in the Federal Courts under 28 U. S. C. § 1292 (b), 88 Harv. L. Rev 607 610 (1975). COOPERS & LYBRAND v. LI VESA Y 475 463 Opinion of the Court of appeals.25 Finally, even if the district judge certifies the order under § 1292 (b), the appellant still “has the burden of persuading the court of appeals that exceptional circumstances justify a departure from the basic policy of postponing appellate review until after the entry of a final judgment.” Fisons, Ltd. v. United States, 458 F. 2d 1241,1248 (CA7 1972). The appellate court may deny the appeal for any reason, including docket congestion.26 By permitting appeals of right from class-designation orders after jurisdictional determinations that turn on questions of fact, the “death knell” doctrine circumvents these restrictions.27 25 H. R. Rep. No. 1667, supra, at 5-6: nWe also recognize that such savings may be nullified in practice by indulgent extension of the amendment to inappropriate cases or by enforced consideration in Courts of Appeals of many ill-founded applications for review. The problem, therefore, is to provide a procedural screen through which only the desired cases may pass, and to avoid the wastage of a multitude of fruitless applications to invoke the amendment contrary to its purpose. . . . . . . Requirement that the Trial Court certify the case as appropriate for appeal serves the double purpose of providing the Appellate Court with the best informed opinion that immediate review is of value, and at once protects appellate dockets against a flood of petitions in inappropriate cases. ... [AJvoidance of ill-founded applications in the Courts of Appeals for piecemeal review is of particular concern. If the consequence of change is to be crowded appellate dockets as well as any substantial number of unjustified delays in the Trial Court, the benefits to be expected from the amendment may well be outweighed by the lost motion of preparation, consideration, and rejection of unwarranted applications for its benefits.” 26 Hearings on H. R. 6238 and H. R. 7260 before Subcommittee No. 3 of the House Committee on the Judiciary, 85th Cong., 2d Sess., 21 (1958). 27 Several Courts of Appeals have heard appeals from discretionary class determinations pursuant to § 1292 (b). See, e. g., Lukenas v. Bryce’s Mountain Resort, Inc., 538 F. 2d 594 (CA4 1976); Susman v. Lincoln American Corp., 561 F. 2d 86 (CA7 1977). See also Samuel v. University of Pittsburgh, 506 F. 2d 355 (CA3 1974). As Judge Friendly has noted: [T]he best solution is to hold that appeals from the grant or denial of class action designation can be taken only under the procedure for inter 476 437 U. S. OCTOBER TERM, 1977 Opinion of the Court Additional considerations reinforce our conclusion that the “death knell” doctrine does not support appellate jurisdiction of prejudgment orders denying class certification. First, the doctrine operates only in favor of plaintiffs even though the class issue—whether to certify, and if so, how large the class should be—will often be of critical importance to defendants as well. Certification of a large class may so increase the defendant’s potential damages liability and litigation costs that he may find it economically prudent to settle and to abandon a meritorious defense. Yet the Courts of Appeals have correctly concluded that orders granting class certification are interlocutory. Whatever similarities or differences there are between plaintiffs and defendants in this context involve questions of policy for Congress.28 Moreover, allowing appeals of right from nonfinal orders that turn on the facts of a particular case thrusts appellate courts indiscriminately into the trial process and thus defeats one vital purpose of the final-judgment rule—“that of maintaining the appropriate relationship between the respective courts. . . . This goal, in the absence of most compelling reasons to the contrary, is very much worth preserving.” 29 locutory appeals provided by 28 U. S. C. § 1292 (b). . . . Since the need for review of class action orders turns on the facts of the particular case, this procedure is preferable to attempts to formulate standards which are necessarily so vague as to give rise to undesirable jurisdictional litigation with concomitant expense and delay.” Parkinson v. April Industries, Inc., 520 F. 2d 650, 660 (CA2 1975) (concurring opinion). 28 The Congress is in a position to weigh the competing interests of the dockets of the trial and appellate courts, to consider the practicability of savings in time and expense, and to give proper weight to the effect on litigants. . . . This Court ... is not authorized to approve or declare judicial modification. It is the responsibility of all courts to see that no unauthorized extension or reduction of jurisdiction, direct or indirect, occurs in the federal system. . . . Any such ad hoc decisions disorganize practice by encouraging attempts to secure or oppose appeals with a consequent waste of tune and money. The choices fall in the legislative domain.” Baltimore Contractors v. Bodinger, 348 U. S. 176, 181-182. 29 Parkinson v. April Industries, Inc., supra, at 654. COOPERS & LYBRAND v. LIVESAY 477 463 Opinion of the Court Accordingly, we hold that the fact that an interlocutory order may induce a party to abandon his claim before final judgment is not a sufficient reason for considering it a “final decision” within the meaning of § 1291.30 The judgment of the Court of Appeals is reversed with directions to dismiss the appeal. It is so ordered. 30 Respondents also suggest that the Court’s decision in Gillespie v. United States Steel Corp., 379 U. S. 148, supports appealability of a classdesignation order as a matter of right. We disagree. In Gillespie, the Court upheld an exercise of appellate jurisdiction of what it considered a marginally final order that disposed of an unsettled issue of national significance because review of that issue unquestionably “implemented the same policy Congress sought to promote in § 1292 (b),” id., at 154, and the arguable finality issue had not been presented to this Court until argument on the merits, thereby ensuring that none of the policies of judicial economy served by the finality requirement would be achieved were the case sent back with the important issue undecided. In this case, in contrast, respondents sought review of an inherently nonfinal order that tentatively resolved a question that turns on the facts of the individual case; and, as noted above, the indiscriminate allowance of appeals from such discretionary orders is plainly inconsistent with the policies promoted by § 1292 (b). If Gillespie were extended beyond the unique facts of that case, § 1291 would be stripped of all significance. 478 437 U. S. OCTOBER TERM, 1977 Opinion of the Court GARDNER v. WESTINGHOUSE BROADCASTING CO. CERTIORARI TO THE UNITED STATES COURT OF APPEALS FOR THE THIRD CIRCUIT No. 77-560. Argued March 22, 1978—Decided June 21, 1978 Petitioner, who had been denied employment by respondent’s radio station, brought an action seeking injunctive relief against respondent on behalf of herself and other females adversely affected by respondent’s alleged practice of discriminating against women. The District Court denied petitioner’s motion for class certification under Fed. Rule Civ. Proc. 23 (b). Claiming that since the relief that could be granted in favor of the class would be broader than the relief she might obtain as an individual, the denial of class certification in effect refused a substantial portion of the injunctive relief sought, petitioner immediately appealed under 28 U. S. C. § 1292 (a)(1), which gives courts of appeals jurisdiction of appeals from interlocutory orders refusing injunctions, but the Court Appeals held that it had no jurisdiction. Held: The order denying class certification was not appealable under § 1292 (a) (1). Pp. 480-482. 559 F. 2d 209, affirmed. Stevens, J., delivered the opinion for a unanimous Court. Robert N. Hackett argued the cause and filed a brief for petitioner. Leonard L. Scheinholtz argued the cause for respondent. With him on the brief were Peter D. Post, Wendell G. Freeland, Richard F. Kronz, and Stuart I. Saltman. Mr. Justice Stevens delivered the opinion of the Court. The United States Court of Appeals for the Third Circuit held that the denial of a class certification could not be appealed immediately under 28 U. S. C. § 1292 (a)(1) 1 as an 1 “§ 1292. Interlocutory decisions. (a) The courts of appeals shall have jurisdiction of appeals from: *‘(1) Interlocutory orders of the district courts of the United States granting, continuing, modifying, refusing or dissolving injunctions, or GARDNER v. WESTINGHOUSE BROADCASTING CO. 479 478 Opinion of the Court order refusing an injunction. 559 F. 2d 209. Because there is a conflict among the Circuits on the question whether § 1292 (a)(1) authorizes such an appeal,2 we granted certiorari. 434 U. S. 984. We affirm. Petitioner unsuccessfully applied for employment as a radio talk-show host at a station owned by respondent. She then brought this civil rights action on behalf of herself and other females adversely affected by respondent’s alleged practice of discriminating against women. The class she sought to represent included respondent’s past, present, and future female employees; unsuccessful female applicants; females deterred by respondent’s reputation from applying for employment; and females who will not in the future be considered for employment by respondent on account of their sex. Her complaint prayed for equitable relief for the entire class.3 Petitioner moved for a class certification pursuant to Fed. Rule Civ. Proc. 23 (b) .4 The District Court denied the motion refusing to dissolve or modify injunctions, except where a direct review may be had in the Supreme Court....” 2 Compare Williams n. Wallace Silversmiths, Inc., 566 F. 2d 364 (CA2 1977); Williams v. Mumford, 167 U. S. App. D. C. 125, 511 F. 2d 363 (1975), cert, denied, 423 U. S. 828 (holding that such orders are not immediately appealable under § 1292 (a)(1)), with Smith n. Merchants & Farmers Bank, 574 F. 2d 982 (CA8 1978); Jones v. Diamond, 519 F. 2d 1090 (CA5 1975); Price v. Lucky Stores, Inc., 501 F. 2d 1177 (CAO 1974); Yaffe v. Powers, 454 F. 2d 1362 (CAI 1972); Brunson v. Board of Trustees of School District 1, 311 F. 2d 107 (CA4 1962), cert, denied, 373 U. S. 933 (holding that such orders are appealable). 3 Petitioner did not file a motion for a preliminary injunction; for that reason, the issue decided in Jenkins n. Blue Cross Mutual Hospital Insurance, Inc., 538 F. 2d 164 (CA7 1976), cert, denied, 429 U. S. 986 (plaintiff’s appeal from denial of class certification and denial of preliminary injunction held within appellate jurisdiction), is not before us. 4 On the same day that she filed her motion for class-action certification, petitioner also filed a motion to compel respondent to answer interrogatories concerning its employee rosters at other radio stations, owned and operated by respondent and located in other cities. The District 480 OCTOBER TERM, 1977 Opinion of the Court 437U.S. on the grounds that petitioner’s claim was not typical and that the case did not present questions of law or fact common to the class. She immediately appealed, invoking the jurisdiction of the Court of Appeals under § 1292 (a)(1).5 Petitioner argues that the relief that could be granted in favor of the class if she prevails would be broader than the relief that she may obtain as an individual. The practical effect of the denial of class certification is, therefore, to refuse a substantial portion of the injunctive relief requested in the complaint. Relying on our decision in General Electric Co. v. Marvel Rare Metals Co., 287 U. S. 430, petitioner then argues that this sort of effect on a request for injunctive relief establishes appealability under § 1292 (a)(1). We cannot agree; indeed the argument misconceives both the scope of § 1292 (a)(1) and the import of decisions such as General Electric. The history of § 1292 (a)(1), which we reviewed in Baltimore Contractors V. Bodinger, 348 U. S. 176, 178-181, need not be repeated. It is sufficient to note that the statute creates an exception from the long-established policy against piecemeal appeals, which this Court is not authorized to enlarge or extend. The exception is a narrow one and is keyed to the “need to permit litigants to effectually challenge interlocutory orders of serious, perhaps irreparable, consequence.” Id., at 181. The order denying class certification in this case did not have any such “irreparable” effect. It could be reviewed both prior to and after final judgment;6 it did not affect the merits Court did not pass on this second motion because it denied class-action certification. 5 Petitioner did not seek certification of her appeal pursuant to § 1292 (b). 6 As the Court of Appeals noted, a decision on class-action status “may be conditional, subject to alteration or amendment prior to final judgment, F. R. Civ. P. 23 (c) (1) . . . . If, after judgment on the merits, the relief granted is deemed unsatisfactory, the question of class status is fully GARDNER v. WESTINGHOUSE BROADCASTING CO. 481 478 Opinion of the Court of petitioner’s own claim; and it did not pass on the legal sufficiency of any claims for injunctive relief.7 This stands in sharp contrast to the order in General Electric.3 In that case the Court held that an order dismissing a counterclaim for an injunction was appealable. The order, therefore, entirely disposed of the defendant’s prayer for injunctive relief; here, the order merely limits the scope of the relief that may ultimately be granted. While it may have a significant effect on the litigation, “[m]any interlocutory orders are equally important, . . . but they are not for that reason converted into injunctions.” Morgantown n. Royal Insurance Co., 337 U. S. 254, 258. As we stated in Switzerland Cheese Assn., Inc. v. E. Horne’s Market, Inc., 385 U. S. 23, 24, “we approach this reviewable.” 559 F. 2d 209, 212; see also United Airlines, Inc. v. McDonald, 432 U. S. 385, 393. 7 There is an important distinction between an order denying an injunction on the merits and “one based on alleged abuse of a discretionary power over the scope of the action.” St ewart- Warner Corp. v. Westinghouse Electric Corp., 325 F. 2d 822, 829 (CA2 1963) (Friendly, J., dissenting). “Where the order is of the former type, the danger of serious harm from the court’s erroneous belief in the existence of a legal barrier to its entertaining a claim for an injunction has been thought to outweigh the general undesirability of interlocutory appeals. The very fact that the second type of order hinges on the trial court’s discretion is itself an indication that such orders, relating primarily to convenience in litigation, carry a lesser threat of harm.” Ibid. 8 In addition to General Electric, petitioner relies on Endow v. New York Life Insurance Co., 293 U. S. 379, and Ettelson v. Metropolitan Life Insurance Co., 317 U. S. 188. Both of those cases, however, rest on the distinction between “legal” and “equitable” claims and supply no precedential weight for petitioner’s argument. Our characterization of those cases in Morgantown v. Royal Insurance Co., 337 U. S. 254, 258, is equally applicable here: “[Distinctions from common-law practice which supported our conclusions in the Endow and Ettelson cases supply no analogy competent to make an injunction of what in any ordinary understanding of the word is not one.” 482 437 U. S. OCTOBER TERM, 1977 Opinion of the Court statute [§ 1292 (a)(1)] somewhat gingerly lest a floodgate be opened that brings into the exception many pretrial orders.” The exception does not embrace orders that have no direct or irreparable impact on the merits of the controversy. The order in this case, like the order in Switzerland Cheese, had no such impact; it “in no way touch[ed] on the merits of the claim but only relatefd] to pretrial procedures . . . ” Id., at 25.9 A holding that such an order falls within § 1292 (a)(1) would compromise “the integrity of the congressional policy against piecemeal appeals.” 385 U. S., at 25. The judgment is affirmed. It is so ordered. 9 In Switzerland Cheese we held that an order denying a motion for summary judgment was not within § 1292 (a). Inasmuch as the requested summary judgment would have included an injunction against trademark infringement, that order was, if anything, a more direct refusal of an injunction than the order denying class certification in this case. Of course, in one sense, the denial of class certification, like the denial of a summary judgment, does “touch on the merits,” since a court must consider whether the complaint reveals common questions of law and fact, or whether there is a material issue of disputed fact. But this determina-tion does not otherwise reflect on the legal sufficiency of the claim for injunctive relief. BETH ISRAEL HOSPITAL v. NLRB 483 Syllabus BETH ISRAEL HOSPITAL v. NATIONAL LABOR RELATIONS BOARD CERTIORARI TO THE UNITED STATES COURT OF APPEALS FOR THE FIRST CIRCUIT No. 77-152. Argued April 24, 1978—Decided June 22, 1978 Petitioner nonprofit hospital had a written rule that prohibited employees from soliciting and distributing literature except in certain employee locker rooms and certain adjacent restrooms. The cafeteria was the common gathering place of employees and had been used by petitioner or with its approval for solicitation and distribution of literature to employees for various nonunion purposes. After an employee had made general distribution in the cafeteria to other employees of a union newsletter and had been warned that she had violated the hospital’s rule and would be dismissed if she did so again, the National Labor Relations Board (NLRB), following a charge by the union, issued an unfair labor practice complaint against petitioner. The NLRB applied to petitioner the rule that it had adopted in St. John’s Hospital & School of Nursing, Inc., 222 N. L. R. B. 1150, that since “the primary function of a hospital is patient care,” and “a tranquil atmosphere is essential to the carrying out of that function,” a hospital may be warranted in imposing more stringent restrictions on employee solicitation and distribution in immediate patient-care areas than are generally permitted other employers, but the balance should be struck against such restrictions in other areas such as lounges and cafeterias, absent a showing of disruption to patients. The NLRB held that petitioner’s ban violated § 8 (a) (1) of the National Labor Relations Act (Act), which by amendments to the Act in 1974 was made applicable to employees of nonprofit healthcare institutions, and that the disciplining of employees for not observing the prohibition violated § 8 (a) (3). The NLRB ordered petitioner to cease and desist from interfering with “concerted union activities” and employees’ § 7 rights, and to rescind its written rule. The Court of Appeals accepted as settled law that restrictions on employee solicitation and distribution during nonworking hours are presumptively invalid absent special circumstances and that here petitioner had not satisfied its burden of justifying the ban on protected activities in the eating areas. While narrowing the scope of the remedies ordered by the NLRB, the court upheld the NLRB’s action rescinding that part of petitioner’s rule applicable to those areas. Held: The Court of Appeals 484 OCTOBER TERM, 1977 Syllabus 437 U.S. did not err in enforcing the NLRB’s order to petitioner to rescind its rule as applied to the hospital’s eating facilities. Pp. 491-508. (a) Freedom of employees effectively to communicate with one another regarding self-organization on the jobsite is essential to their right to self-organize and to bargain collectively established by § 7 of the Act, Republic Aviation Corp. n. NLRB, 324 U. S. 793, and in the light of its experience the NLRB is free to adopt a rule that, absent special circumstances, an employer’s restriction on employee solicitation during nonworking time and distribution during such time in nonworking areas is presumptively an unreasonable interference with § 7 rights constituting an unfair labor practice under §8 (a)(1), without the necessity of proving the underlying generic facts that persuaded it to reach that conclusion. Pp. 491-493. (b) Nothing in the legislative history of the 1974 amendments shows a congressional policy inconsistent with the NLRB’s approach to enforcement of § 7 organizational rights in the hospital context. Pp. 496-500. (c) The NLRB by those amendments is responsible for administering the federal national labor relations policy in the health-care industry. Though the NLRB is no more an expert in that industry than it is in other enterprises within its jurisdiction, it is the NLRB’s function to strike the balance in all areas within its jurisdiction between conflicting legitimate interests in order to effectuate the national labor policy. Hence petitioner’s argument that the NLRB lacks expertise to make judgments involving hospitals and that the principle of limited judicial review should not apply in that area, is without merit. Pp. 500-501. (d) The NLRB’s conclusion that “the possibility of any disruption in patient care resulting from solicitation or distribution of literature is remote” as applied to petitioner’s cafeteria is rational and fully supported by the record, as indicated by much cogent evidence, including the facts that only 1.56% of the cafeteria’s patrons are patients and that petitioner itself permitted nonunion solicitation and distribution in the cafeteria. Moreover, petitioner introduced no evidence of untoward effects on patients during the period when the rules permitted limited union solicitation in the cafeteria. Pp. 501-505. (e) Contrary to petitioner’s argument, it is not irrational for the NLRB to uphold, as it has, a ban against solicitation in the dining area of a public restaurant, where such solicitation tends to upset patrons, while prohibiting a ban on such activity in a hospital cafeteria like petitioner’s, 77% of whose patrons are employees, absent evidence that nonemployee patrons would be upset. That argument fails to consider that the NLRB’s position struck the appropriate balance between orga- BETH ISRAEL HOSPITAL v. NLRB 485 483 Opinion of the Court nizational and employer rights in the particular industry to which each solicitation rule applied. Pp. 505-507. 554 F. 2d 477, affirmed. Brennan, J., delivered the opinion of the Court, in which Stewart, White, Marshall, and Stevens, JJ., joined. Blackmun, J., post, p. 508, and Powell, J., post, p. 509, filed opinions concurring in the judgment, in which Burger, C. J. and Rehnquist, J., joined. Louis Chandler argued the cause for petitioner. With him on the brief was Robert Chandler. Norton J. Come argued the cause for respondent. With him on the brief were Solicitor General McCree, John S. Irving, and Carl L. Taylor. Laurence Gold argued the cause for intervenor Massachusetts Hospital Workers’ Union Local 880, Service Employees’ International Union. With him on the brief were Lester Asher, J. Albert Woll, and George Kaufmann* Mr. Justice Brennan delivered the opinion of the Court. The National Labor Relations Act, 49 Stat. 449, as amended, 61 Stat. 136, 29 U. S. C. §§ 151 to 168, was further amended in 1974 to extend its coverage and protection to employees of nonprofit health-care institutions.1 Act of July 26, 1974, Pub. L. No. 93—360, 88 Stat. 395. Petitioner is a Boston nonprofit hospital whose employees are covered by the amended Act. This case presents the question whether the Court of Appeals for the First Circuit erred in ordering *Richard Dorn filed a brief for the National Union of Hospital and Health Care Employees, RWDSU, AFL-CIO, as amicus curiae urging affirmance. 1 Coverage was achieved by deleting from the definition of “employer” in § 2 (2) of the Act, 29 U. S. C. § 152 (2), the provision that an employer shall not include “any corporation or association operating a hospital, if no part of the net earnings inures to the benefit of any private shareholder or individual . . . .” Act of June 23, 1947, ch. 120, 61 Stat. 136. 486 OCTOBER TERM, 1977 Opinion of the Court 437U.S. enforcement of that part of an order of the National Labor Relations Board based on the Board’s finding that petitioner, in violation of §§ 8 (a)(1) and (3), 29 U. S. C. §§ 158 (a)(1) and (3), interfered with its employees’ rights guaranteed by § 7 of the Act, 29 U. S. C. § 157, by issuing and enforcing a rule that prohibits employees from soliciting union support and distributing union literature during nonworking time in the hospital cafeteria and coffeeshop used primarily by employees but also used by patients and visitors. In 1970, prior to the advent of any union organizational activity at the hospital, petitioner announced a rule barring solicitation and distribution of literature in any area to which patients or visitors have access. Petitioner permitted these activities only in certain employee locker rooms and certain adjacent restrooms. App. 59. In July 1974, however, as a result of a proceeding instituted against it before the Massachusetts Labor Relations Commission, petitioner announced a rule permitting solicitation in the cafeteria on a one-to-one basis while maintaining the total ban on distribution. Id., at 67. On March 6, 1975, shortly after the NLRB acquired jurisdiction, petitioner reinstated its previous rule limiting employee solicitation and distribution to certain employee locker rooms and restrooms. Id., at 70.2 That rule provides: “There is to be no soliciting of the general public (patients, visitors) on Hospital property. Soliciting and the distribution of literature to B. I. employees may be done by other B. I. employees, when neither individual is on his or her working time, in employee-only areas— employee locker rooms and certain adjacent rest rooms. Elsewhere within the Hospital, including patient-care and 2 The July 1974 rule was in effect at the time the complaint was filed. Prior to the hearing before the Administrative Law Judge, however, the Board amended its complaint to encompass the March 6, 1975, policy which prohibited all solicitation and distribution in the cafeteria. BETH ISRAEL HOSPITAL v. NLRB 487 483 Opinion of the Court all other work areas, and areas open to the public such as lobbies, cafeteria and coffee shop, corridors, elevators, gift shop, etc., there is to be no solicitation nor distribution of literature. “Solicitation or distribution of literature on Hospital property by non-employees is expressly prohibited at all times. “Consistent with our long-standing practices, the annual appeal campaigns of the United Fund and of the Combined Jewish Philanthropies for voluntary charitable gifts will continue to be carried out by the Hospital.” Id., at 70-71. Upon a charge filed by the union,3 the Board issued a complaint and the matter was tried before an Administrative Law Judge. The Board affirmed the rulings, findings, and conclusions of the Administrative Law Judge that petitioner’s issuance and maintenance of the rules violated § 8 (a)(1) and the disciplining of an employee for an infraction of them violated §8 (a)(3). 223 N. L. R. B. 1193 (1976). The Administrative Law Judge found that there were few places in which employees’ § 7 rights effectively could be exercised, that petitioner had not offered any convincing evidence that the rule was necessary to prevent disruptions in patient care, and that, on balance, the rule was an unjustified infringement of § 7 rights. See 223 N. L. R. B., at 1198. The Board issued an order, paragraph 1 of which broadly required petitioner to cease and desist from interfering with “concerted union activities” and “exercise of [employees’] rights guaranteed in Section 7 of the Act,” and paragraph 2 (b) of which required petitioner to “[r]escind its written rule prohibiting distribution of union literature and union solicitation in its cafeteria 3 The charges leading to the complaint were filed by Massachusetts Hospital Workers’ Union, Local 880, Service Employees International Union, AFL-CIO. 488 OCTOBER TERM, 1977 Opinion of the Court 437U.S. and coffeeshop.” 223 N. L. R. B., at 1199, as modified, id., at 1193. The Court of Appeals accepted as settled law that rules restricting employee solicitation during nonworking time, and distribution during nonworking time in nonworking areas are presumptively invalid in the absence of special circumstances to justify them, 554 F. 2d 477, 480 (1977), and held that, since “[i]n this case, the application of the employer’s nosolicitation, no-distribution rules to the cafeteria and coffee shop banned concerted activities in non-working areas during non-working time . . . [t] he burden, therefore, was on the hospital to show that special circumstances justified its curtailment of protected activities in these two places.” Ibid. After review of the record, the court held that “the Board did not err in finding that the hospital had not justified its nosolicitation, no-distribution rule as it related to the cafeteria and coffee shop.” Id., at 481. The court refused to enforce paragraph 1 of the Board’s order, however, on the ground that no proclivity to violate the Act had been shown to support that broad cease-and-desist order. It also enforced paragraph 2 (b) only after adding to the order the clarifying words “that part of” so that petitioner was required to “[r]escind that part of its written rule prohibiting distribution [of union literature and union solicitation in its cafeteria and coffeeshop],” id., at 482 (emphasis in original), to make clear that the validity of the rules as applied to areas outside the cafeteria and coffeeshop remained open. The Board has not sought review of the Court of Appeals’ rulings in these respects.4 The narrow question for decision, therefore, is whether the Court of Appeals erred in enforcing the Board’s order requiring petitioner to rescind the rules as applied to the hospital’s eating 4 Petitioner’s application of the rules to other areas not devoted to immediate patient care has since been litigated before the Board in another case. Beth Israel Hospital, 228 N. L. R. B. 1495, 95 LRRM 1087 (1977). BETH ISRAEL HOSPITAL v. NLRB 489 483 Opinion of the Court facilities. Because of a suggested conflict among Courts of Appeals as to the validity of restrictions upon solicitation and distribution in patient-access areas of the hospital, such as petitioner’s cafeteria and coffeeshop, we granted certiorari.5 434 U. S. 1033 (1978). We affirm. I Although petitioner employs approximately 2,200 regular employees,6 only a fraction of them have access to many of the areas in which solicitation is permitted. Solicitation and distribution are not permitted in all locker areas. Rather, of the total number of looker areas only six separate and scattered locker areas containing 613 lockers are accessible to all employees for these purposes.7 Moreover, most of these rooms are divided and restricted on the basis of sex, and in any event 5 The Court of Appeals in this case, and the Court of Appeals for the Seventh Circuit, Lutheran Hosp. v. NLRB, 564 F. 2d 208 (1977), cert, pending, No. 77-1289, have enforced Board orders protecting solicitation and distribution in cafeterias and coffeeshops. In Lutheran Hospital, the order enforced extended beyond cafeterias to all areas other than “immediate patient care areas.” The Court of Appeals for the Tenth Circuit, St. John’s Hospital & School of Nursing, Inc. v. NLRB, 557 F. 2d 1368 (1977), together with the Courts of Appeals for the District of Columbia and Sixth Circuits, have denied enforcement to similar Board orders applicable to cafeterias as well as to other patient-access areas. Baylor Univ. Medical Center v. NLRB, 188 U. S. App. D. C. 109, 578 F. 2d 351 (1978); NLRB v. Baptist Hospital, Inc., 576 F. 2d 107 (CA6 1978). 6 This number is exclusive of house staff, attending physicians, students, and employees of Harvard University who work at the hospital. App. 28. 7 There are four categories of locker rooms. The first, in which there are a total of 613 lockers, are areas in which any employee may engage in solicitation and distribution. The second, in which there are a total of 470 lockers, are areas in which, for security reasons, only the employees to whom the lockers have been assigned have access. The other two categories which comprise the remainder of the hospital’s lockers are off limits to solicitation and distribution because they are located in working areas or in areas in which patients or the general public have access. 223 N. L. R. B. 1193, 1197 (1976); App. 127-134. 490 OCTOBER TERM, 1977 Opinion of the Court 437U.S. are not generally used even by petitioner to communicate messages to employees. The cafeteria,8 on the other hand, is a common gathering room for employees. A 3-day survey conducted by petitioner revealed that 77% of the cafeteria’s patrons were employees while only 9% were visitors and 1.56% patients. The cafeteria is also equipped with vending machines used by employees for snacks during coffeebreaks and other nonworking time. Petitioner itself has recognized that the cafeteria is a natural gathering place for employees on nonworking time, for it has used and permitted use of the cafeteria for solicitation and distribution to employees for purposes other than union activity. For example, petitioner maintains an official bulletin board in the cafeteria for communicating certain messages to employees. On occasion it has set up special tables in or near the cafeteria entrance to aid solicitation of contributions for the United Way or United Fund charities, the Jewish Philanthropies Organization Drive, the Israel Emergency Fund, and to recruit members for the credit union. When petitioner embarked upon an intensive cost-reduction program, styled “Save a Buck a Day” or “BAD,” it used the cafeteria to post banners and distribute informational literature touting the program to employees, and, significantly, generally did not use the locker rooms and restrooms for this purpose. In addition to these official uses, petitioner maintains an unofficial bulletin board in the cafeteria for the employees’ use, a rack and small table which display commercial literature, such as travel brochures, and information of interest only to employees, such as carpool openings. “[T]here are relatively few places where employees can congregate or meet on hospital grounds or in the nearby vicinity for the purpose of discussing nonwork related matters other than in the cafeteria; secondly, the area in the neighbor- 8 During the pendency of this litigation, the coffeeshop was dismantled, and the space added to the cafeteria. BETH ISRAEL HOSPITAL v. NLRB 491 483 Opinion of the Court hood of the hospital is congested and provides no ready access to employees”; 223 N. L. R. B., at 1198 (opinion of Administrative Law Judge). Petitioner, moreover, has adopted the policy of refusing to make available to unions the names and addresses of employees unless ordered to do so by the Board. App. 33. Petitioner has also made antiunion statements in a newsletter distributed to employees with their paychecks at their work stations. On October 25, 1974, Ann Schunior, a medical technician in the Department of Medicine, was distributing the union newsletter As We See It by circulating from table to table. She approached only persons she thought were employees, and if not sure of their employee status, inquired whether they were, explaining that she was distributing literature for employees. Petitioner’s general director witnessed this activity, advised Schunior that she was violating the hospital’s nodistribution rule, and demanded that she cease the distribution. A written warning notice was issued to Schunior the same day advising that she had been in flagrant violation of the hospital’s rules and that further violations would result in dismissal. 223 N. L. R. B., at 1195-1196. The publication As We See It was objectionable to petitioner because certain issues were said to contain remarks which disparaged the hospital’s ability to provide adequate patient care, primarily because of under-staffing. Id., at 1196. II A We have long accepted the Board’s view that the right of employees to self-organize and bargain collectively established by § 7 of the NLRA, 29 U. S. C. § 157, necessarily encompasses the right effectively to communicate with one another regarding self-organization at the jobsite.9 Republic Aviation 9 We recently reiterated this principle in Central Hardware Co. v. NLRB, 407 U.S. 539 (1972): “[Section 7] organization rights are not viable in a vacuum; their 492 437 U.S. OCTOBER TERM, 1977 Opinion of the Court Corp. v. NLRB, 324 U. S. 793 (1945), articulated the broad legal principle which must govern the Board’s enforcement of this right in the myriad factual situations in which it is sought to be exercised: “[The Board must adjust] the undisputed right of self-organization assured to employees under the Wagner Act and the equally undisputed right of employers to maintain discipline in their establishments. Like so many others, these rights are not unlimited in the sense that they can be exercised without regard to any duty which the existence of rights in others may place upon employer or employee.” Id., at 797-798. That principle was further developed in NLRB v. Babcock & Wilcox Co., 351 U. S. 105 (1956), where the Court stated: “Accommodation between [employee-organization rights and employer-property rights] must be obtained with as little destruction of one as is consistent with the main-tenance of the other.” Id., at 112. Based on its experience in enforcing the Act, the Board developed legal rules applying the principle of accommodation. The effect of these rules is to make particular restrictions on employee solicitation and distribution presumptively lawful or unlawful under §8 (a)(1) subject to the introduction of evidence sufficient to overcome the presumption. Thus, the Board has held that restrictions on employee solicitation during nonworking time, and on distribution during nonworking time in non working areas, are violative of §8 (a)(1) unless the employer justifies them by a showing of special circum- effectiveness depends in some measure on the ability of employees to learn the advantages and disadvantages of organization from others. Early in the history of the administration of the Act the Board recognized the importance of freedom of communication to the free exercise of organization rights.” Id., at 542-543 (citation omitted). BETH ISRAEL HOSPITAL v. NLRB 493 483 Opinion of the Court stances which make the rule necessary to maintain production or discipline.10 In the case of retail marketing establishments, including public restaurants, however, the Board has held that solicitation and distribution may be prohibited on the selling floor at all times.11 Republic Aviation Corp., supra, sustained the Board’s general approach to adjudication of §8 (a)(1) charges. There we held that the Board is free to adopt, in light of its experience, a rule that, absent special circumstances, a particular employer restriction is presumptively an unreasonable interference with § 7 rights constituting an unfair labor practice under § 8 (a)(1), without the necessity of proving the underlying generic facts which persuaded it to reach that conclusion. The validity of such a rule “[l]ike a statutory presumption or one established by regulation, . . . perhaps in varying degree, depends upon the rationality between what is proved and what is inferred.” Republic Aviation, supra, at 804-805 (footnote omitted). The Board here relied on, and petitioner challenges, the fashioning of a similar presumption applicable to hospitals. 10 The Board’s solicitation rule was first announced in Peyton Packing Co., 49 N. L. R. B. 828, 843 (1943). The Board’s decision in LeTourneau Co. of Ga., 54 N. L. R. B. 1253 (1944), which applied the presumption to a no-distribution rule enforced against employee organizers distributing literature in the employer’s parking lot, was affirmed with Republic Aviation Corp. v. NLRB, 324 U. S. 793 (1945), without separate discussion. In Stoddard Quirk Mjg. Co., 138 N. L. R. B. 615 (1962), however, the Board established the distinction between distribution and solicitation, limiting the presumption as applicable to distribution only in non working areas. For purposes of that rule, the Board considers the distribution of signature cards to be solicitation and not distribution. See id., at 620 n. 6. 11 See Marriott Corp. (Children’s Inn), 223 N. L. R. B. 978 (1976); Bankers Club, Inc., 218 N. L. R. B. 22 (1975); McDonald’s Corp., 205 N. L. R. B. 404 (1973); Marshall Field & Co., 98 N. L. R. B. 88 (1952), enf’d, 200 F. 2d 375 (CA7 1953); Goldblatt Bros., Inc., 77 N. L. R. B. 1262 (1948); May Dept. Stores Co., 59 N. L. R. B. 976 (1944), enf’d as modified, 154 F. 2d 533 (CA8 1946). 494 OCTOBER TERM, 1977 Opinion of the Court 437U.S. B Although, prior to the 1974 amendments, the Board had considered the validity of no-solicitation and no-distribution rules in the context of proprietary hospitals, no clear rule emerged from its decisions. In Summit Nursing & Convalescent Home, Inc., 196 N. L. R. B. 769 (1972), enf. denied, 472 F. 2d 1380 (CA6 1973), a divided panel, reversing the Administrative Law Judge, held unlawful a rule prohibiting solicitation or distribution “at any time in the patient or public area within the [nursing] home, or in the nurses’ stations.” Another divided panel, in Guyan Valley Hospital, Inc., 198 N. L. R. B. 107 (1972), affirming the Trial Examiner, held lawful a rule prohibiting “soliciting in working areas during working hours.” In Guyan Valley the Trial Examiner noted that the employer’s rule did not interfere with “solicitation ... in the waiting room, the employees’ dining room, and the parking lot.” Id., at 111. The Board apparently relied upon this fact to distinguish it from Summit Nursing, supra. See 198 N. L. R. B., at 107 n. 2. Finally, in Bellaire General Hospital, 203 N. L. R. B. 1105 (1973), the panel which had split in Summit Nursing, unanimously held unlawful a rule prohibiting solicitation and distribution “by employees while off duty or during working hours.” 203 N. L. R. B., at 1108. This series of somewhat inconclusive decisions was the background against which, after the 1974 amendments, the full Board considered development of a rule establishing the permissible reach of employer rules prohibiting solicitation and distribution in all health-care institutions. In a unanimous opinion, in St. John’s Hospital & School of Nursing, Inc., 222 N. L. R. B. 1150 (1976), the Board concluded that the special characteristics of hospitals justify a rule different from that which the Board generally applies to other employers. On the basis of evidence and aided by the briefs amici curiae filed by the American Hospital Association and District 1199 BETH ISRAEL HOSPITAL v. NLRB 495 483 Opinion of the Court of the National Union of Hospital and Health Care Employees, the Board found: “that the primary function of a hospital is patient care and that a tranquil atmosphere is essential to the carrying out of that function. In order to provide this atmosphere, hospitals may be justified in imposing somewhat more stringent prohibitions on solicitation than are generally permitted. For example, a hospital may be warranted in prohibiting solicitation even on nonworking time in strictly patient care areas, such as the patients’ rooms, operating rooms, and places where patients receive treatment, such as x-ray and therapy areas. Solicitation at any time in those areas might be unsettling to the patients—particularly those who are seriously ill and thus need quiet and peace of mind.” Ibid, (emphasis added). The Board concluded that prohibiting solicitation in such situations was justified and required striking the balance against employees’ interests in organizational activity. The Board determined, however, that the balance should be struck against the prohibition in areas other than immediate patientcare areas such as lounges and cafeterias absent a showing that disruption to patient care would necessarily result if solicitation and distribution were permitted in those areas. The Board concluded, on a record devoid of evidence which contradicted that assessment, that the possibility of disruption to patient care in those areas must be deemed remote. Ill Petitioner challenges the qualified extension of the rule affirmed in Republic Aviation to hospitals on several grounds: First, it argues that the Board’s decision conflicts with the congressional policy evinced in the 1974 hospital amendments that the “self-organizational activities of health care employees not be allowed to ‘disrupt the continuity of patient care.’ ” Brief for Petitioner 10. Second, it argues that the basis for 496 OCTOBER TERM, 1977 Opinion of the Court 437U.S. that rule, the principle of limited judicial review of agency action, is inapposite here because the Board is acting outside of its area of expertise. Third, it argues that the Board’s decision is unsupported by evidence and is irrational. Finally, it argues that it is irrational to distinguish between the nonemployee-access cafeteria involved here and the public-access restaurants in which the Board has upheld solicitation bans. A Contrary to petitioner’s assertion, nothing in the legislative history of the 1974 amendments indicates a congressional policy inconsistent with the Board’s general approach to enforcement of § 7 self-organizational rights in the hospital context. First, there is no reason to believe, as petitioner asserts, that Congress intended either to prohibit solicitation entirely in the health-care industry or to limit it to the extent the Board had required at the time the 1974 amendments were enacted. In extending coverage of the Act to nonprofit hospitals, Congress enacted special provisions for strike notice and mediation, applicable solely to the health-care industry, intended to avoid disruptions of patient care caused by strikes.12 12 Section 1 (b) of the 1974 Act, 88 Stat. 395, amended § 2 of the NLRA by adding a definition of “health care institution” to which the special provisions would be applicable. Section 1 (d), 88 Stat. 396, amended the notice provisions of § 8 (d) of the NLRA by requiring, with respect to health-care institutions, 90-day notice of termination or expiration of a contract, 60-day notice to the Federal Mediation and Conciliation Service (FMCS) of contract termination or expiration, and 30-day notice to FMCS with respect to initial contract negotiation disputes arising after recognition, and by requiring that the health-care institution and the labor organization participate in mediation at the direction of the FMCS. Section 1 (e), 88 Stat. 396, added a new § 8 (g) to the NLRA, requiring labor organizations to give a 10-day written notice to the health-care institution and to FMCS before engaging in picketing, strikes, or other concerted refusals to work. Section 2 of the 1974 Act added a new § 213 to the Labor Management Relations Act, 1947, 29 U. S. C. § 183 (1970 ed., Supp. V), which authorizes upon certain conditions the constitution of a BETH ISRAEL HOSPITAL v. NLRB 497 483 Opinion of the Court It is significant that, although, as indicated, supra, at 494, at the time the 1974 amendments were enacted, the Board had spoken with neither clarity nor one voice on the issue, Congress did not enact any special provision regarding solicitation and distribution in particular or disruption of patient care in general other than through strikes. We can only infer, therefore, that Congress was satisfied to rely on the Board to continue to exercise the responsibility to strike the appropriate balance between the interests of hospital employees, patients, and employers. Second, nothing in the legislative history supports petitioner’s argument that the particular approach to enforcement of § 7 rights in the hospital context adopted by the Board is inconsistent with congressional policy. The elimination of the nonprofit-hospital exemption reflected Congress’ judgment that hospital care would be improved by extending the protection of the Act to nonprofit health-care employees.13 Congress found that wages were low and working conditions poor in the health-care industry, and that as a result, employee morale was low and employment turnover high.14 Congress deter- Special Board of Inquiry to investigate and report concerning the labor dispute. For a more detailed explanation of these provisions, see Vernon, Labor Relations in the Health Care Field under the 1974 Amendments to the National Labor Relations Act, 70 Nw. U. L. Rev. 202 (1975). 13 See Id., at 203-204. 14 See, e. g., the remarks of Senator Cranston, the floor manager of the bill: “During the last 2% years, hospital wage increases have lagged far behind those received by workers in other industries. . . . “Today, hospital workers are still notoriously underpaid. . . . “The long hours worked and the small monetary reward received by hospital workers result in a constant turnover with a consequent threat to the maintenance of an adequate standard of medical care. This was emphasized over and over again by many of the witnesses. Turnover rates for employees in several hospitals that were studied were reported by witnesses to be as high as 1,200 to 1,500 [percent] a year. “Mr. President, both management and union witnesses reported lower 498 437 U.S. OCTOBER TERM, 1977 Opinion of the Court mined that the extension of organizational and collectivebargaining rights would ameliorate these conditions and elevate the standard of patient care.15 Congress also found that “the exemption . . . had resulted in numerous instances of recognition strikes and picketing. Coverage under the Act should completely eliminate the need for such activity, since the procedures of the Act will be available to resolve organizational and recognition disputes.” S. Rep. No. 93-766, p. 3 (1974). It is true, as petitioner argues, that Congress felt that “the needs of patients in health care institutions required special consideration in the Act . . . ibid., and that among the witnesses before the Committee on Labor and Public Welfare, “[t]here was a recognized concern for the need to avoid disruption of patient care wherever possible.” Id., at 6. But these statements do not support petitioner’s further contention that congressional policy establishes that the very fact that hospitals are involved justifies, without more, a restrictive no-solicitation rule the validity of which must be sustained unless the Board proves that patient care will not be disrupted. To begin with, the congressional statements quoted, when placed in context, offer no support for such an argument.16 turnover after unionization than before. . . . [T]he turnover rates at the two hospitals which had been 1,200 to 1,500 percent a year before unionization dropped to 24 to 30 percent a year after unionization. Indeed it has been convincingly argued that when hospital employees are unionized . . . the result is better job stability and security than is possible without such collective bargaining arrangements. This will also mean a better job done in terms of the quality of patient care provided. “Mr. President, I urge all those who want improved health care and increased stability for labor-management relations in health care institutions to support this bill.” 120 Cong. Rec. 12936-12938 (1974). 15See ibid.; id., at 16899-16900 (remarks of Rep. Thompson). 16 The statements in full are as follows: In the Committee’s deliberations on this measure, it was recognized that the needs of patients in health care institutions required special consideration in the Act including a provision requiring hospitals to have BETH ISRAEL HOSPITAL v. NLRB 499 483 Opinion of the Court Moreover, Congress addressed its concern for the unique problems presented by labor disputes in the health-care industry by adding specific strike-notice and mediation provisions designed to avert interruption in the delivery of critical health-care services; none expresses a policy in favor of curtailing self-organizational rights.17 Indeed, although Congress recognized that strikes could cause complete disruption of patient care and enacted provisions designed to forestall them, it apparently felt that extension of the right to strike was sufficiently important to fulfillment of its goals to permit strikes despite that result. If Congress was willing to countenance the total, albeit temporary, disruption of patient care caused by strikes in order to achieve harmonious employer-employee relations and long-term improved health care, we cannot say it necessarily regarded appropriately regulated solicitation and distribution in areas such as the cafeteria as undesirable without evidence of a substantial threat of harm to patients. In light of Congress’ express finding that improvements in health care would result from the right to organize, and that unionism is necessary to overcome the poor working conditions sufficient notice of any strike or picketing to allow for appropriate arrangements to be made for the continuance of patient care in the event of a work stoppage.” S. Rep. No. 93-766, p. 3 (1974). “PRIORITY CASE HANDLING “Many of the witnesses before the Committee, including both employee and employer witnesses, stressed the uniqueness of health care institutions. There was a recognized concern for the need to avoid disruption of patient care wherever possible. “It was this sensitivity to the need for continuity of patient care that led the Committee to adopt amendments with regard to notice requirements and other procedures related to potential strikes and picketing. “Because of the need for continuity of patient care, the Committee expects the NLRB to give special attention and priority to all charges of employer, employee and labor organization unfair practices involving health care institutions consistent with [existing priorities].” Id., at 6-7. 17 See n. 12, supra. 500 OCTOBER TERM, 1977 Opinion of the Court 437U.S. retarding the delivery of quality health care, we therefore cannot say that the Board’s policy—which requires that absent such a showing solicitation and distribution be permitted in the hospital except in areas where patient care is likely to be disrupted—is an impermissible construction of the Act’s policies as applied to the health-care industry by the 1974 amendments. Even if the legislative history arguably pointed toward a contrary view, the Board’s construction of the statute’s policies would be entitled to considerable deference. NLRB v. Iron Workers, 434 U. S. 335, 350 (1978); NLRB v. Weingarten, Inc., 420 U. S. 251, 266-267 (1975). B Petitioner disputes the applicability of the principle of limited judicial review of Board action generally and of the principle announced in Republic Aviation, regarding the Board’s authority to fashion generalized rules in light of its experience, in particular, to the Board’s decision involving hospitals. Arguing that the Board’s conclusion regarding the likelihood of disruption to patient care which solicitation in a patient-access cafeteria would produce is essentially a medical judgment outside of the Board’s area of expertise, it contends that the Board’s decision is not entitled to deference. Rather, since it, not the Board, is responsible for establishing hospital policies to ensure the well-being of its patients, the Board may not set aside such a policy without specifically disproving the hospital’s judgment that solicitation and distribution in the cafeteria would disrupt patient care. Brief for Petitioner 18. We think that this argument fundamentally misconceives the institutional role of the Board. It is the Board on which Congress conferred the authority to develop and apply fundamental national labor policy. Because it is to the Board that Congress entrusted the task of “applying the Act’s general prohibitory language in the light of the infinite combinations of events which might be charged BETH ISRAEL HOSPITAL v. NLRB 501 483 Opinion of the Court as violative of its terms,” Republic Aviation, 324 U. S., at 798, that body, if it is to accomplish the task which Congress set for it, necessarily must have authority to formulate rules to fill the interstices of the broad statutory provisions. It is true that the Board is not expert in the delivery of health-care services, but neither is it in pharmacology, chemical manufacturing, lumbering, shipping, or any of a host of varied and specialized business enterprises over which the Act confers jurisdiction. But the Board is expert in federal national labor relations policy, and it is in the Board, not petitioner, that the 1974 amendments vested responsibility for developing that policy in the health-care industry. It is not surprising or unnatural that petitioner’s assessment of the need for a particular practice might overcompensate its goals, and give too little weight to employee organizational interests. Here, as in many other contexts of labor policy, “[t]he ultimate problem is the balancing of the conflicting legitimate interests. The function of striking that balance to effectuate national labor policy is often a difficult and delicate responsibility, which the Congress committed primarily to the National Labor Relations Board, subject to limited judicial review.” NLRB v. Truck Drivers, 353 U. S. 87, 96 (1957). The judicial role is narrow: The rule which the Board adopts is judicially reviewable for consistency with the Act, and for rationality, but if it satisfies those criteria, the Board’s application of the rule, if supported by substantial evidence on the record as a whole, must be enforced.18 NLRB v. Erie Resistor Corp., 373 U. S. 221, 235-236 (1963); Phelps Dodge Corp. v. NLRB, 313 U. S. 177, 194 (1941). C Petitioner’s contention that the Board’s decision is unsupported by evidence and irrational is without merit. Not 18 See § 10 (e), NLRA, 29 IT. S. C. § 160 (e); Administrative Procedure Act, 5 U. S. C. §706 (2)(E) (1976 ed.); Universal Camera Corp. v. NLRB, 340 U. S. 474 (1951). 502 OCTOBER TERM, 1977 Opinion of the Court 437U.S. withstanding petitioner’s challenge, the Board’s conclusion that “the possibility of any disruption in patient care resulting from solicitation or distribution of literature is remote,” St. John's Hospital & School of Nursing, Inc., 222 N. L. R. B., at 1151, as applied to petitioner’s cafeteria, is fully supported by the record. The Board had before it evidence that patients’ meals are provided in their rooms. A patient is not allowed to visit the cafeteria unless his doctor certifies that he is well enough to do so. Thus, patient use of the cafeteria is voluntary, random, and infrequent. It is of critical significance that only 1.56% of the cafeteria’s patrons are patients. Patients who frequent the cafeteria would not expect to receive special attention or primary care there and any unusually sensitive to seeing union literature distributed or overhearing discussions about unionism, readily could avoid the cafeteria without interfering with the hospital’s program of care. Especially telling is the fact that petitioner, under compulsion of the Massachusetts Labor Commission, permitted limited union solicitation in the cafeteria for a significant period, apparently without untoward effects, and that petitioner, who logically is in the best position to offer evidence on the point, was unable to introduce any evidence to show that solicitation or distribution was or would be harmful.19 There was also cogent evidence that petitioner itself recognized that at least some solicitation and distribution would not upset patients and undermine its function of providing quality medical care. It thus appears that petitioner’s rule was more restrictive than necessary to avert that result.20 19 Cf. International Harvester Co. v. Ruckelshaus, 155 U. S. App. D. C. 411, 439, 478 F. 2d 615, 643 (1973). 20 Evidence that petitioner adopted a less restrictive approach to behavior in the cafeteria which would be at least as disquieting to patients as union solicitation further supports the Board’s conclusion that the risk of harm to patients is not so great as to justify an unlimited restriction. Petitioner advised its professional staff of complaints voiced by patients and visitors BETH ISRAEL HOSPITAL v. NLRB 503 483 Opinion of the Court Petitioner had permitted use of the cafeteria for other types of solicitation, including fund drives, which, if not to be equated with union solicitation in terms of potential for generating controversy, at least indicates that the hospital regarded the cafeteria as sufficiently commodious to admit solicitation and distribution without disruption.21 While in other contexts, it has been recognized that organizational activity can result in behavior which, as petitioner argues and we agree, would be undesirable in the hospital’s cafeteria,22 the Board has not foreclosed the hospital from imposing less restrictive means of regulating organizational activity more nearly directed toward the harm to be avoided.23 based on overheard clinical discussions about named patients in such places as the cafeteria line. Petitioner warned that the “effect [of this on patients] can be devastating . . . ,” App. 136, and that “[p]atients and visitors [have been] horrified to overhear—in . . . cafeteria lines . . .— what is to the engrossed clinician innocuous professional discussion.” Id., at 138. This kind of discussion, far more unsettling than talk of wages and working conditions, was not banned from the cafeteria; rather, petitioner merely required staff to “restrict the voicing of your clinical discussions to include none other than your intended audience.” Ibid. 21 Compare Goldblatt Bros., Inc., 11 N. L. R. B. 1262 (1948), in which, explaining its decision to uphold a ban on solicitation in a department store restaurant, the Board noted: “[I]n some of the stores the restaurant consists of a counter, in which restaurant employees on duty, other employees off duty, union organizers, and customers are in close contact with each other. Under these circum-stances, union solicitation in the restaurants is as apt to disrupt the Respondent’s business as is such solicitation carried on in any other portion of the store in which customers are present.” Id., at 1263-1264. 22 See, e. g., McDonald’s Corp., 205 N. L. R. B., at 407 n. 18 (opinion of Administrative Law Judge) (“Some solicitation might result in a pleasant and informative chat between the employees on their nonwork time in working areas. On the other hand, it might lead to a bitter exchange of insults or worse . . .”). 23 For example, a rule forbidding any distribution to or solicitation of nonemployees would do much to prevent potentially upsetting literature from being read by patients. Petitioner, in fact, has such a rule, see supra, 504 OCTOBER TERM, 1977 Opinion of the Court 437U.S. The Board was, of course, free to draw an inference from these facts in light of its experience, the validity of which “depends upon the rationality between what is proved and what is inferred.”24 Republic Aviation, 324 U. S., at 805 (footnote omitted). It cannot fairly be said that the inference drawn by the Board regarding the likelihood of disruption of patient care in light of this evidence was irrational. Similarly, it is the Board upon whom the duty falls in the first instance to determine the relative strength of the conflicting interests and to balance their weight. As the Court noted in Hudgens v. NLRB, 424 U. S. 507, 522 (1976), “[t]he locus of [the] accommodation [between the legitimate interests of both] may fall at differing points along the spectrum depending on the nature and strength of the respective § 7 rights and private property rights asserted in any given context.” Here, the employees’ interests are at their strongest, at 486-487, and it has not been shown that organizational activity by Schunior or anyone else actually resulted in distribution to nonemployees. This rule could be readily enforced at petitioner’s hospital, moreover, since employees are required to wear name tags—and many do—and since security guards monitor the cafeteria. Secondly, the Board may determine that a rule requiring face-to-face distribution rather than leaving literature on a table accessible to all is a justified accommodation of § 7 rights with petitioner’s legitimate desire to avoid having potentially upsetting literature read by patients. 24 The requirement that decisions be supported by evidence on the record “does not go beyond the necessity for the production of evidential facts, however, and compel evidence as to the results which may flow from such facts. ... An administrative agency with power after hearings to determine on the evidence in adversary proceedings whether violations of statutory commands have occurred may infer within the limits of the inquiry from the proven facts such conclusions as reasonably may be based upon the facts proven. One of the purposes which lead to the creation of such boards is to have decisions based upon evidential facts under the particular statute made by experienced officials with an adequate appreciation of the complexities of the subject which is entrusted to their administration.” Republic Aviation, 324 U. S., at 800. (Citations omitted.) BETH ISRAEL HOSPITAL v. NLRB 505 483 Opinion of the Court for unlike the interests involved in NLRB v. Babcock & Wilcox Co., 351 U. S., at 113, “[the] activity was carried on by employees already rightfully on the employer’s property.” Hudgens, 424 U. S., at 521-522, n. 10. “[T]he employer’s management interests rather than his property interests [are] involved. . . . This difference is ‘one of substance.’ ” Ibid. (citations omitted). On the other hand, in the context of health-care facilities, the importance of the employer’s interest in protecting patients from disturbance cannot be gainsaid. While outside of the health-care context, the availability of alternative means of communication is not, with respect to employee organizational activity, a necessary inquiry, see Babcock & Wilcox, supra, at 112-113, it may be that the importance of the employer’s interest here demands use of a more finely calibrated scale. For example, the availability of one part of a health-care facility for organizational activity might be regarded as a factor required to be considered in evaluating the permissibility of restrictions in other areas of the same facility. That consideration is inapposite here, however, where the only areas in which organizational rights are permitted is not conducive to their exercise. Moreover, the area in which organizational rights are sought here is a “natural gathering are [a]” for employees, 554 F. 2d, at 481, and one in which the risk of harm to patients is relatively low as compared to potential alternative locations within the facility. On the basis of the record before it, we cannot say that the Board, in evaluating the relative strength of the competing interests, failed to consider any factor appropriately to be taken into account. Cf. Babcock & Wilcox, supra. D Petitioner’s argument that it is irrational to hold, as the Board has, on the one hand, that a rule prohibiting solicitation in the dining area of a public restaurant is lawful because 506 OCTOBER TERM, 1977 Opinion of the Court 437U.S. solicitation has the tendency to upset patrons,25 while one prohibiting like activity in a hospital’s cafeteria is unlawful absent evidence that nonemployee patrons would be upset, on the other, has only superficial appeal. That argument wholly fails to consider that the Board concluded that these rules struck the appropriate balance between organizational and employer rights in the particular industry to which each is applicable. In the retail marketing and restaurant industries, the primary purpose of the operation is to serve customers, and this is done on the selling floor of a store or in the dining area of a restaurant. Employee solicitation in these areas, if disruptive, necessarily would directly and substantially interfere with the employer’s business. On the other hand, it would be an unusual store or restaurant which did not have stockrooms, kitchens, and other nonpublic areas, and in those areas employee solicitation of nonworking employees must be permitted. In that context, the Board concluded that, on balance, employees’ organizational interests do not outweigh the employer’s interests in prohibiting solicitation on the selling floor. In the hospital context the situation is quite different. The main function of the hospital is patient care and therapy and those functions are largely performed in areas such as operating rooms, patients’ rooms, and patients’ lounges. The Board does not prohibit rules forbidding organizational activity in these areas. On the other hand, a hospital cafeteria, 77% of whose patrons are employees, and which is a natural gathering place for employees, functions more as an employee-service area than a patient-care area. While it is true that the fact of access by visitors and patients renders the analogy to areas such as stockrooms in retail operations less than complete, it cannot be said that when the primary function and use of the cafeteria, the availability of alternative areas of the facility in which § 7 rights effectively could be exercised, and 25 See cases cited n. 11, supra. BETH ISRAEL HOSPITAL v. NLRB 507 483 Opinion of the Court the remoteness of interference with patient care are considered, it was irrational to strike the balance in favor of § 7 rights in the hospital cafeteria and against them in public restaurants. The Board’s explanation of the consistent principle underlying the different results in each situation cannot fairly be challenged. St. John’s Hospital & School of Nursing, Inc., 222 N. L. R. B., at 1150-1151, n. 3. IV In summary, we reject as without merit petitioner’s contention that, in enacting the 1974 health-care amendments, Congress intended the Board to apply different principles regarding no-solicitation and no-distribution rules to hospitals because of their patient-care functions. We therefore hold that the Board’s general approach of requiring health-care facilities to permit employee solicitation and distribution during nonworking time in nonworking areas, where the facility has not justified the prohibitions as necessary to avoid disruption of health-care operations or disturbance of patients, is consistent with the Act. We hold further that, with respect to the application of that principle to petitioner’s cafeteria, the Board was appropriately sensitive to the importance of petitioner’s interest in maintaining a tranquil environment for patients. Insofar as petitioner’s challenge is to the substantiality of the evidence supporting the Board’s conclusions, this Court’s review is, of course, limited. “Whether on the record as a whole there is substantial evidence to support agency findings is a question which Congress has placed in the keeping of the Courts of Appeals. This Court will intervene only in what ought to be the rare instance when the standard appears to have been misapprehended or grossly misapplied.” Universal Camera Corp. v. NLRB, 340 U. S. 474, 491 (1951). We cannot say that the Court of Appeals’ assessment of the record either “misapprehended” or “grossly misapplied” that standard. The Court of Appeals did note, however, that the 508 OCTOBER TERM, 1977 Blackmun, J., concurring in judgment 437 U. S. Board’s guidelines are still in flux and are far from self-defining, concluding, and we agree: “[T]he Board [bears] a heavy continuing responsibility to review its policies concerning organizational activities in various parts of hospitals. Hospitals carry on a public function of the utmost seriousness and importance. They give rise to unique considerations that do not apply in the industrial settings with which the Board is more familiar. The Board should stand ready to revise its rulings if future experience demonstrates that the wellbeing of patients is in fact jeopardized.” 554 F. 2d, at 481. The authority of the Board to modify its construction of the Act in light of its cumulative experience is, of course, clear. NLRB v. Iron Workers, 434 IL S., at 351; NLRB v. Weingarten, Inc., 420 U. S., at 265-267. , ’ ’ Affirmed. Mr. Justice Blackmun, with whom The Chief Justice and Mr. Justice Rehnquist join, concurring in the judgment. I concur only in the result the Court reaches here, for I, too, agree with much that Mr. Justice Powell says in his separate opinion. There is, of course, a certain irony when the Board grants protection from solicitation to the retail store and to the Burger Chef and the Hot Shoppe cafeteria, but at the same time denies it to the hospital restaurant facility where far more than mere commercial interests are at stake. Patients and their concerned families are not to be treated as impersonal categories or classes. They are individuals with problems that ought not be subject to aggravation. Nevertheless, on this record, as the Court’s opinion reveals, it would have been difficult for the Board to reach a different result, when it utilized, questionably in my view, the rule of Republic Aviation Corp. v. NLRB, 324 U. S. 793 (1945), even as perhaps modified for application in the hospital setting. BETH ISRAEL HOSPITAL v. NLRB 509 483 Powell, J., concurring in judgment The tenor of the Court’s opinion and of the Board’s approach concerns me. There are many hospital coffeeshops and cafeterias that are primarily patient and patient-relative oriented, despite the presence of employee patrons, far more so than this very restricted Beth Israel operation, that seems akin to a manufacturing plant’s emloyees’ cafeteria. I fear that this unusual case will be deemed to be an example for all hospital eating-facility cases, and that the Board and the courts now will go further down the open-solicitation road than they would have done, had a more usual hospital case been the one first to come here. Hospitals, after all, are not factories or mines or assembly plants. They are hospitals, where human ailments are treated, where patients and relatives alike often are under emotional strain and worry, where pleasing and comforting patients are principal facets of the day’s activity, and where the patient and his family—irrespective of whether that patient and that family are labor or management oriented—need a restful, uncluttered, relaxing, and helpful atmosphere, rather than one remindful of the tensions of the marketplace in addition to the tensions of the sickbed. I entertain distinct doubts about whether the Board, in its preoccupation with labor-management problems, has properly sensed and appreciated the true hospital operation and its atmosphere and the institution’s purpose and needs. I earnestly share the caveat pronounced by the Court of Appeals, and reproduced by the Court in the next-to-the-last paragraph of its opinion, ante, at 508, and I sincerely hope that the Board bears that heavy responsibility in mind when it considers other hospital cases that come before it for decision. Mr. Justice Powell, with whom The Chief Justice and Mr. Justice Rehnquist join, concurring in the judgment. In Republic Aviation Corp. v. NLRB, 324 U. S. 793 (1945), this Court approved the reasoning of the National Labor Relations Board in Peyton Packing Co., 49 N. L. R. B. 828 510 OCTOBER TERM, 1977 Powell, J., concurring in judgment 437 U. S. (1943), enf’d, 142 F. 2d 1009 (CA5), cert, denied, 323 U. S. 730 (1944), and the balance it struck in adjusting the respective rights of industrial employers and employees. The Court also endorsed the Board’s formulation: Because working time is for work, a rule prohibiting union solicitation during working time u ‘must be presumed to be valid in the absence of evidence that it was adopted for a discriminatory purpose’ ”; but during nonworking time, when an employee’s time is his own even though he is on company property, a rule prohibiting union solicitation “ ‘must be presumed to be an unreasonable impediment to self-organization and therefore discriminatory in the absence of evidence that special circumstances make the rule necessary in order to maintain production or discipline.’ ” 324 U. S., at 803-804, n. 10 (quoting Peyton Packing Co., supra, at 843-844). The Republic Aviation rule is inapplicable in the instant case, which arises from a setting entirely different from the one in which the rule was formulated. I concur in the judgment of the Court, however, because I regard the Board’s decision as based on substantial evidence even without the assistance of the Republic Aviation presumption. I The rule of Republic Aviation was adopted in the context of labor relations in industrial and manufacturing plants, where third parties unconnected with labor or management generally are not involved. In such a setting, it is relatively simple to divide the work environment into the two spheres defined in Peyton Packing. During working time an employer’s prohibition of solicitation and distribution may be presumed valid, because “[w]orking time is for work”; but during nonworking time or in nonworking areas, such rules are presumptively invalid. The latter part of the Board’s set of presumptions reflects the reasonable inference, based on the Board’s experience with the actual facts of industrial life, that BETH ISRAEL HOSPITAL v. NLRB 511 483 Powell, J., concurring in judgment such employers ordinarily will not have legitimate reasons to restrict employees’ activities on their own time, even if on company property. In sustaining the Board’s presumption, this Court recounted its development and said : “We perceive no error in the Board’s adoption of this presumption. The Board had previously considered similar rules in industrial establishments and the definitive form which the Peyton Packing Company decision gave to the presumption was the product of the Board’s appraisal of normal conditions about industrial establishments. Like a statutory presumption or one established by regulation, the validity, perhaps in a varying degree, depends upon the rationality between what is proved and what is inferred.” 324 U. S., at 804-805 (footnotes omitted; emphasis supplied). The rationality found to exist in Republic Aviation, and therefore the validity of the presumption, cannot be transferred automatically to other workplaces, for to do so would sever the connection between the inference and the underlying proof. The Court’s approval of the Republic Aviation rule was based explicitly on the Board’s considered appraisal of “normal conditions about industrial establishments.” 1 Conditions in industrial or manufacturing plants differ substantially from conditions in sales and service establishments where employees and members of the public mingle. When confronted with the problem of retail-establishment rules prohibiting solicitation and distribution, the Board wisely refrained from mechanically applying the Republic Aviation rule when its justification was absent. The Board recognized that in the setting of a retail establishment, an employer well 1 Even the formulation of the “special circumstances” rule is stated in terms of the specific environment of an industrial plant, speaking of circumstances making a restriction on employee activity “ 'necessary in order to maintain production or discipline.’ ” 324 U. S., at 803-804, n. 10. 512 OCTOBER TERM, 1977 Powell, J., concurring in judgment 437 U. S. might have legitimate reasons for prohibiting solicitation and distribution on the selling floor and in other areas where customers are likely to be present.2 In the retail-store cases, the Board weighed the respective interests of the employer and the employees and concluded that the employer’s rule was reasonable in view of the extent of the public’s presence on the premises, the relationship between the public and the employees, and the fact that the employer’s main business, consisting of direct selling to customers, would be disrupted. The same conclusion was reached with respect to a public restaurant on the premises of a retail store when on-duty and off-duty employees were “in close contact with each other” and with customers, on the theory that under such circumstances, union solicitation would be “as apt to disrupt the [employer’s] business as . . . solicitation carried on in any other portion of the store in which customers are present.” Goldblatt Bros., Inc., 77 N. L. R. B. 1262, 1264 (1948). See also McDonald’s Corp., 205 N. L. R. B. 404, 408 (1973).3 2 See Marriott Corp. (Children’s Inn), 223 N. L. R. B. 978 (1976); Bankers Club, Inc., 218 N. L. R. B. 22 (1975); McDonald’s Corp., 205 N. L. R. B. 404 (1973); Marshall Field & Co., 98 N. L. R. B. 88 (1952), enf’d, 200 F. 2d 375 (CA7 1953); Goldblatt Bros., Inc., 77 N. L. R. B. 1262 (1948); May Dept. Stores Co., 59 N. L. R. B. 976 (1944), enf’d as modified, 154 F. 2d 533 (CA8), cert, denied, 329 U. S. 725 (1946) . 3 The Board’s retail-establishment cases might be interpreted as instances in which the Board concluded that the Republic Aviation presumption had been rebutted by the employer’s proof of “special circumstances.” The special circumstances would be created by the “presence [of customers] and the likelihood of their being exposed to union activities.” Bankers Club, Inc., supra, at 27. But even if this were the correct formulation— that the Republic Aviation presumption applies to retail establishments but is rebutted by proof of the presence of members of the public in areas where solicitation takes place—that test would be satisfied in all retail-establishment cases as well as in the instant case. The result would be the same as if the presumption did not apply at all. After special circumstances had been shown, the Board then would have to determine the proper balance between employees’ rights and the employer’s interests. BETH ISRAEL HOSPITAL v. NLRB 513 483 Powell, J., concurring in judgment In my view, the presence of patients and members of the public in the hospital cafeteria removes the case from the framework established in Republic Aviation, just as the presence of customers has that effect in the Board’s retail-establishment cases. The hospital’s function in serving patients, their families, and visitors is much like the retail establishment’s function in serving its customers. That a nonprofit hospital does not share the profit motive of a retail establishment does not diminish the hospital employer’s professional concern for the welfare of those in its care, including not only patients but also their friends and relatives who come to visit. It is true that the hospital’s primary function is carried out in the immediate patient-care areas, just as the retail establishment’s main function is carried out on the selling floor. But the Board has applied its retail-store rules to public restaurants on the premises of the retail store, see supra, at 512, notwithstanding the fact that the primary selling function does not take place there. Public restaurants in retail stores are provided for some of the reasons that hospitals maintain public eating places—including the convenience of the establishment’s patrons. In addition, a hospital’s more general purpose extends to, and pervades, all areas of the hospital to which the public has access; it is not limited narrowly to the provision of technical medical treatment.4 Part of the hospital’s func 4 Thus, while the Board has distinguished between selling and certain nonselling areas of department stores, and has applied the presumption of invalidity to no-solicitation rules in some nonselling public areas, see Marshall Field & Co., supra, at 92-93, a similar line may not be drawn so easily between patient-care and nonpatient-care areas of a hospital. As the Court of Appeals for the Tenth Circuit observed in denying enforcement to the Board’s attempt to divide the areas of a hospital, “the ultimate factual inferences on which the Board’s distinction [is] based were drawn not from the record evidence but rather from the Board’s own perceptions of modem hospital care and the physical, mental, and emotional conditions of hospital patients—areas outside the Board’s acknowledged field of expertise in labor/management relations.” St. John’s Hospital & School of Nursing, Inc. v. NLRB, 557 F. 2d 1368, 1373 (1977). 514 OCTOBER TERM, 1977 Powell, J., concurring in judgment 437 U. S. tion is to provide a “total environment . . . where the medical needs of patients are served by maintaining a climate free of strife and controversy.” NLRB v. Baptist Hospital, Inc., 576 F. 2d 107, 110 (CA6 1978). In this respect, the Board should take greater account of the impact of solicitation in this sensitive area than it does with respect to retail establishments. A presumption developed in and geared to the context of industrial establishments, which the Board has declined to apply to retail stores, simply has no relevance to hospitals. II The Board contends that it has effected a proper accommodation of the competing interests in St. John’s Hospital & School of Nursing, Inc., 222 N. L. R. B. 1150 (1976), enf. granted in part and denied in part, 557 F. 2d 1368 (CAIO 1977), in which it applied the basic rule of Republic Aviation but found “sufficient justification” for curtailment of employee rights in certain areas of the hospital.5 Acknowledging that the “primary function of a hospital is patient care and that a tranquil atmosphere is essential to the carrying out of that function,” the Board concluded in St. John’s that “hospitals may be justified in imposing somewhat more stringent prohibitions on solicitation than are generally permitted.” Accordingly, a hospital might prohibit solicitation in “strictly patient care areas,” such as “patients’ rooms, operating rooms, and places where patients receive treatment”; but not in other areas of the hospital, even those to which patients and visitors have access. 222 N. L. R. B., at 1150-1151. In my view, the Board’s “accommodation” of the competing interests in St. John’s fails to give appropriate weight to the unique characteristics of a hospital. It amounts to no 5 Both the parties and the court in St. John’s started from the premise that the Republic Aviation rule applied. The Court of Appeals disagreed, however, with the Board’s assessment that special circumstances justified the hospital’s restriction only in "immediate” patient-care areas. BETH ISRAEL HOSPITAL v. NLRB 515 483 Powell, J., concurring in judgment more than an application of the Republic Aviation rule to certain areas of a hospital but not others, despite the fact that members of the public are present and potentially affected even in areas of a hospital not characterized as “strictly patient care” areas. I believe that the Tenth Circuit was correct in refusing to accord the St. John’s presumption the kind of deference that was accorded the Republic Aviation presumption when applied in the industrial setting. I would hold that the potential impact on patients and visitors of union solicitation and distribution of literature in hospitals requires the Board to make a far more sensitive inquiry into the actual circumstances of each case. Once the Board is deprived of the presumption of invalidity of an employer’s rule, it must establish by substantial evidence on the record as a whole that the employer has violated § § 8 (a)(1) and 8 (a) (3). On the facts of this case, I would hold that the Board has carried its burden. The Board must reach an accommodation between the respective rights of employer and employees “with as little destruction of one as is consistent with the maintenance of the other.” NLRB v. Babcock & Wilcox Co., 351 U. S. 105, 112 (1956); see Eastex, Inc. v. NLRB, post, p. 556; Hudgens v. NLRB, 424 U. S. 507, 521-523 (1976); Central Hardware Co. v. NLRB, 407 U. S. 539, 542-545 (1972). “The locus of that accommodation, however, may fall at differing points along the spectrum depending on the nature and strength of the respective § 7 rights and [the employer’s] rights asserted in any given context.” Hudgens, supra, at 522. In this case, the employer’s asserted concern is with the welfare of patients and their visitors, a particularly weighty “management” interest. In accommodating the interests of employer and employees in a hospital case, the Board must recognize the employer’s responsibility for the welfare of patients and other third parties present in the hospital.6 6 This, of course, is consistent with Congress’ concern, in enacting the 516 OCTOBER TERM, 1977 Powell, J., concurring in judgment 437 U. S. Yet in view of the facts in this case, which either are stipulated or largely undisputed, I think the Board has met its burden by substantial evidence. As found by the Administrative Law Judge, use of the hospital cafeteria by employees is substantial (77%), while use by patients is negligible (1.56%) and use by the general public is relatively low (under 10%). The cafeteria is predominantly the employees’ facility, and there hardly is any other area of the hospital in which employees may communicate with each other while at the hospital. The parties stipulated that the only areas where employees can gather are the locker areas and restrooms, and only 613 of the 2,200 employees’ lockers are accessible to all employees.7 In addition to the unavailability of other convenient places for employee communication, cf. Babcock & Wilcox, supra, at 112-113, the facts show that the hospital cafeteria is used by both the employer and employees for a variety of commercial and noncommercial notices and solicitations. And while the hospital was concerned about the disruptive effect on patients of employees’ conversations about the medical progress of particular patients, it implemented only a precatory rule, not an outright prohibition of all such conversations in the cafeteria. See ante, at 502-503, n. 20. The hospital failed to introduce any evidence of a reasonable possibility of harmful consequences to patients or visitors. 1974 health-care amendments, “for the need to avoid disruption of patient care wherever possible.” S. Rep. No. 93-766, p. 6 (1974). 7 The Administrative Law Judge also found that the urban location of the hospital and the widely dispersed residences of hospital employees ma.Up. communication outside the hospital difficult. In addition, petitioner would not provide the union with a list of employees’ names and addresses. “The place of work is a place uniquely appropriate for dissemination of views concerning the bargaining representative and the various options open to the employees,” NLRB v. Magnavox Co., 415 U. S. 322, 325 (1974); see Eastex, Inc. v. NLRB, post, at 574, and the hospital cafeteria was the most appropriate place for such communication on the facts of this case. BETH ISRAEL HOSPITAL v. NLRB 517 483 Powell, J., concurring in judgment It relied primarily on arguments with respect to hospitals in general. No testimony was introduced that the practice at Beth Israel is to seek early rehabilitation of patients by encouraging them to leave their rooms at the earliest time compatible with their condition, and to move about the hospital. The further weakness in petitioner’s case is that it introduced no medical testimony that related such practices and needs to its cafeteria.8 Putting it differently, the undisputed evidence portrays this cafeteria as being one essentially operated for employees as their primary gathering place, and as almost wholly unrelated to patient care. In sum, I view this case as essentially barren of the type of evidence that could be produced on behalf of many hospitals when confronted with a similar problem. See, e. g., NLRB v. Baptist Hospital, Inc., 576 F. 2d 107 (CA6 1978). My concurrence in the judgment is based entirely on the facts, as I disagree—for the reasons above stated—with the rationale of the Board, its reliance upon a wholly inappropriate presumption, and its unrealistic distinction between hospital and retail-store cafeterias. I also note that the Court emphasizes the facts of this case, and the “critical significance [of the fact] that only 1.56% of the cafeteria’s patrons are patients.” Ante, at 502.9 8 Rather, the employer rested on the allegedly inflammatory nature of a union newsletter distributed by one employee, without introducing any evidence that the newsletter had fallen or would fall into the hands of patients or visitors. Furthermore, proof of such a probability would not be relevant to the no-solicitation portion of the hospital’s rule. The hospital allowed one-to-one solicitation in the cafeteria until after the initiation of these proceedings; yet petitioner was “unable to show any instance of injury to patients” while that more permissive rule was in effect. 223 N. L. R.B. 1193, 1197 (1976). 9 Moreover, the Court’s opinion expresses no view as to the validity of prohibiting employee solicitation or distribution in other areas of a hospital which may not be devoted “strictly” or “immediately” to patient care but to which patients and visitors have access. This question was not presented in this case. 518 OCTOBER TERM, 1977 Syllabus 437 U. S. HICKLIN ET AL. V. ORBECK, COMMISSIONER, DEPARTMENT OF LABOR OF ALASKA, et al. APPEAL FROM SUPREME COURT OF ALASKA No. 77-324. Argued March 21, 1978—Decided June 22, 1978 Appellants, at least five of whom are not residents of Alaska, challenged in state court the constitutionality of the “Alaska Hire” statute (which was enacted professedly for the purpose of reducing unemployment within the State) that requires that all Alaskan oil and gas leases, easements or right-of-way permits for oil and gas pipelines, and unitization agreements contain a requirement that qualified Alaska residents be hired in preference to nonresidents. The trial court upheld the statute. The Alaska Supreme Court affirmed except for that part of the Act that contained a one-year durational residency requirement, which it held invalid. Held: 1. The invalidation of the one-year durational residency requirement does not moot the case, since a controversy still exists between the nonresident appellants, none of whom can qualify as “residents” under the statutory definition, and the appellees, state officials. Those appellants thus have a continuing interest in restraining the statutory discrimination favoring state residents. P. 523. 2. Alaska Hire violates the Privileges and Immunities Clause of Art. IV, § 2. Pp. 523-534. (a) Though the Clause “does not preclude disparity of treatment in the many situations where there are perfectly valid independent reasons for it,” it “does bar discrimination against citizens of other States where there is no reason for the discrimination beyond the mere fact that they are citizens of other States.” Toomer v. Witsell, 334 U. S. 385, 396. See also Mullaney v. Anderson, 342 U. S. 415. Pp. 524-526. (b) Even under the dubious assumption that a State may validly alleviate its unemployment problem by requiring private employers within the State to discriminate against nonresidents, Alaska Hire cannot be upheld, for the record indicates that Alaska’s unemployment was not attributable to the influx of nonresident jobseekers, but rather to the fact that a substantial number of Alaska’s jobless residents were unemployed either because of lack of education and job training or because of geographical remoteness from job opportunities. Employment of nonresidents threatened to deny jobs to residents only to the extent that jobs for which untrained residents were being prepared might be filled HICKLIN v. ORBECK 519 518 Syllabus by nonresidents before the residents’ training was completed. Moreover, even if a showing was made that nonresidents were “a peculiar source of the evil,” Toomer v. Witsell, supra, at 398, at which Alaska Hire was aimed, the statute would still be invalid, for its discrimination against nonresidents does not bear a substantial relationship to the “evil” that they are said to present, since statutory preference over nonresidents is given to all Alaskans, not just those who are unemployed. Pp. 526-528. (c) Alaska’s ownership of the oil and gas that are the subject matter of Alaska Hire constitutes insufficient justification for the statute’s pervasive discrimination against nonresidents. Alaska Hire’s reach includes employers who have no connection with the State’s oil and gas, perform no work on state land, have no contractual relationship with the State, and receive no payment from the State; and the Act’s coverage is not limited to activities connected with the extraction of Alaska’s oil and gas. Pp. 528-531. (d) The conclusion that Alaska Hire cannot withstand constitutional scrutiny is fortified by decisions under the Commerce Clause that circumscribe a State’s ability to prefer its own citizens in the utilization of natural resources found within its borders but destined for interstate commerce. West v. Kansas Natural Gas, 221 U. S. 229; Pennsylvania v. West Virginia, 262 U. S. 553; and Foster Packing Co. v. Haydel, 278 U. S. 1. The oil and gas upon which Alaska hinges its discrimination are bound for out-of-state consumption and are of profound national importance while the breadth of the discrimination mandated by Alaska Hire transcends the degree of resident bias that Alaska’s ownership of the oil and gas can justifiably support. Pp. 531-534. 565 P. 2d 159, reversed. Brennan, J., delivered the opinion for a unanimous Court. Robert H. Wagstaff argued the cause for appellants. With him on the briefs was Lee S. Glass. Ronald W. Lorensen, Assistant Attorney General of Alaska, argued the cause and filed a brief for appellees.* *Briefs of amici curiae urging reversal were filed by Edwin Vieira, Jr., for the National Right to Work Legal Defense Foundation; and by Peabody Testing—Bill Miller X-Ray, Inc. Ronald Y: Amemiya, Attorney General, and Lawrence D. Kumabe and Michael A. Lilly, Deputy Attorneys General, filed a brief for the State of Hawaii as amicus curiae urging affirmance. 520 OCTOBER TERM, 1977 Opinion of the Court 437U.S. Mr. Justice Brennan delivered the opinion of the Court. In 1972, professedly for the purpose of reducing unemployment in the State, the Alaska Legislature passed an Act entitled “Local Hire Under State Leases.” Alaska Stat. Ann. § § 38.40.010 to 38.40.090 (1977). The key provision of “Alaska Hire,” as the Act has come to be known, is the requirement that “all oil and gas leases, easements or right-of-way permits for oil or gas pipeline purposes, unitization agreements, or any renegotiation of any of the preceding to which the state is a party” contain a provision “requiring the employment of qualified Alaska residents” in preference to nonresidents.1 Alaska Stat. Ann. § 38.40.030 (a) (1977).2 This employment preference is administered by providing persons meeting the statutory requirements for Alaskan residency with certificates of residence—“resident cards”—that can be presented to an employer covered by the Act as proof of residency. 8 Alaska Admin. Code 35.015 (1977). Appellants, individuals desirous of securing jobs covered by the Act but unable to qualify for the necessary resident cards, challenge Alaska Hire as violative of 1 The regulations implementing the Act further require that all nonresidents be laid off before any resident “working in the same trade or craft” is terminated: “[T]he nonresident may be retained only if no resident employee is qualified to fill the position.” 8 Alaska Admin. Code 35.011 (1977). See also 8 Alaska Admin. Code 35.042 (4) (1977). 2 The complete text of § 38.40.030 (a) is as follows: “In order to create, protect and preserve the right of Alaska residents to employment, the commissioner of natural resources shall incorporate into all oil and gas leases, easements or right-of-way permits for oil or gas pipeline purposes, unitization agreements, or any renegotiation of any of the preceding to which the state is a party, provisions requiring the lessee to comply with applicable laws and regulations with regard to the employment of Alaska residents, a provision requiring the employment of qualified Alaska residents, a provision prohibiting discrimination against Alaska residents and, when in the determination of the commissioner of natural resources it is practicable, a provision requiring compliance with the Alaska Plan, all in accordance with the provisions of this chapter.” HICKLIN v. ORBECK 521 518 Opinion of the Court both the Privileges and Immunities Clause of Art. IV, § 2, and the Equal Protection Clause of the Fourteenth Amendment. I Although enacted in 1972, Alaska Hire was not seriously-enforced until 1975, when construction on the Trans-Alaska Pipeline3 was reaching its peak. At that time, the State Department of Labor began issuing residency cards and limiting to resident cardholders the dispatchment to oil pipeline jobs. On March 1, 1976, in response to “numerous complaints alleging that persons who are not Alaska residents have been dispatched on pipeline jobs when qualified Alaska residents were available to fill the jobs,” Executive Order #76-1, Alaska Dept, of Labor (Mar. 1, 1976) (emphasis in original), Edmund Orbeck, the Commissioner of Labor and one of the appellees here, issued a cease-and-desist order to all unions supplying pipeline workers4 enjoining them “to respond to all open job calls by dispatching all qualified Alaska residents before any non-residents are dispatched.” Ibid, (emphasis in original). As a result, the appellants, all but one of whom had previously worked on the pipeline, were prevented from obtaining pipeline-related work. Consequently, on April 28, 1976, appellants filed a complaint in the Superior Court in Anchorage seeking declaratory and injunctive relief against enforcement of Alaska Hire. At the time the suit was filed, the provision setting forth the qualifications for Alaskan residency for purposes of Alaska 3 See Trans Alaska Pipeline Rate Cases, 436 U. S. 631 (1978); TransAlaska Pipeline Authorization Act, 87 Stat. 584, 43 U. S. C. § 1651 et seq. (1970 ed., Supp. V). 4App. 13-14. The vast majority of pipeline jobs were filled through union dispatchment. Deposition of David Finrow, Deputy Director of the Wage and Hour Division of the Alaska Dept, of Labor, in No. 3025 (Sup. Ct. Alaska), pp. 18-19, 28, 48. 522 OCTOBER TERM, 1977 Opinion of the Court 437 U. S. Hire, Alaska Stat. Ann. § 38.40.090,5 included a one-year durational residency requirement. Appellants attacked that requirement as well as the flat employment preference given by Alaska Hire to state residents. By agreement of the parties, consideration of a motion for a preliminary injunction was consolidated with the determination of the suit on its merits. The case was submitted on affidavits, depositions, and memoranda of law; no oral testimony was taken. On July 21,1976, the Superior Court upheld Alaska Hire in its entirety and denied appellants all relief. On appeal, the Alaska Supreme Court unanimously held that Alaska Hire’s one-year durational residency requirement was unconstitutional under both the state and federal Equal Protection Clauses, 565 P. 2d 159, 165 (1977), and held further that a durational residency requirement in excess of 30 days was constitutionally infirm. Id., at 171.6 By a vote of 3 to 2, however, the court held that the Act’s general preference for Alaska residents was constitutionally permissible. Appellants appealed the State Supreme Court’s judgment insofar as it embodied the latter holding, and we noted probable jurisdiction. 434 U. S. 919 (1977). We reverse. 5 Section 38.40.090 provides: “In this chapter “(1) ‘resident’ means a person who “(A) except for brief intervals, military service, attendance at an educational or training institution, or for absences for good cause, is physically present in the state for a period of one year immediately before the time his status is determined; “(B) maintains a place of residence in the state; “(C) has established residency for voting purposes in the state; “(D) has not, within the period of required residency, claimed residency in another state; and “(E) shows by all attending circumstances that his intent is to make Alaska his permanent residence.” 6 Appellees have not cross-appealed this portion of the Alaska Supreme Court’s decision, which rests upon an independent and adequate state ground. Murdock v. Memphis, 20 Wall. 590 (1875). HICKLIN v. ORBECK 523 518 Opinion of the Court II Preliminarily, we hold that this case is not moot. Despite the Alaska Supreme (Court’s invalidation of the one-year durational residency requirement, a controversy still exists between at least five of the appellants—Tommy Ray Woodruff, Frederick A. Mathers, Emmett Ray, Betty Cloud, and Joseph G. O’Brien—and the state appellees. These five appellants have all sworn that they are not residents of Alaska, Record 43, 47, 49, 96, 124. Therefore, none of them can satisfy the element of the definition of “resident” under § 38.40.090 (1) (D) that requires that an individual “has not, within the period of required residency, claimed residency in another state.” They thus have a continuing interest in restraining the enforcement of Alaska Hire’s discrimination in favor of residents of that State.7 Appellants’ principal challenge to Alaska Hire is made under the Privileges and Immunities Clause of Art. IV, § 2: “The Citizens of each State shall be entitled to all Privileges and Immunities of Citizens in the several States.” That provision, which “appears in the so-called States’ Relations Article, the same Article that embraces the Full Faith and Credit Clause, the Extradition Clause . . . , the provisions for the admission of new States, the Territory and Property Clause, and the Guarantee Clause,” Baldwin v. Montana Fish and Game Comm’n, 436 U. S. 371, 379 (1978), “establishes a norm of comity,” Austin v. New Hampshire, 420 U. S. 656, 660 (1975), that is to prevail among the States with respect to their treat 7 As to the remaining three appellants—Sidney S. Hicklin, Ruby E. Dorman, and Harry A. Browning—the case does appear moot. At the time this suit was instituted, all three claimed to be Alaskan residents, but none had lived in the State continuously for one year. Record 45, 51-52, 126-127. Consequently, the only aspect of Alaska Hire they challenged was the Act’s one-year durational residency requirement. When this requirement was held invalid by the Alaska Supreme Court, their controversy with the appellees seems to have terminated. 524 OCTOBER TERM, 1977 Opinion of the Court 437 U. S. ment of each other’s residents.8 The purpose of the Clause, as described in Paul v. Virginia, 8 Wall. 168, 180 (1869), is “to place the citizens of each State upon the same footing with citizens of other States, so far as the advantages resulting from citizenship in those States are concerned. It relieves them from the disabilities of alienage in other States; it inhibits discriminating legislation against them by other States; it gives them the right of free ingress into other States, and egress from them; it insures to them in other States the same freedom possessed by the citizens of those States in the acquisition and enjoyment of property and in the pursuit of happiness; and it secures to them in other States the equal protection of their laws. It has been justly said that no provision in the Constitution has tended so strongly to constitute the citizens of the United States one people as this.” Appellants’ appeal to the protection of the Clause is strongly supported by this Court’s decisions holding violative of the Clause state discrimination against nonresidents seeking to ply their trade, practice their occupation, or pursue a common calling within the State. For example, in Ward v. Maryland, 12 Wall. 418 (1871), a Maryland statute regulating the sale of most goods in the city of Baltimore fell to the privileges and immunities challenge of a New Jersey resident against whom the law discriminated. The statute discrimi 8 Although this Court has not always equated state residency with state citizenship, compare Travis v. Yale & Towne Mig. Co., 252 U. S. 60, 78-79 (1920), and Blake v. McClung, 172 U. S. 239, 246-247 (1898), with Southern R. Co. v. Mayfield, 340 U. S. 1, 3-4 (1950); Douglas v. New Haven R. Co., 279 U. S. 377, 386-387 (1929); and La Tourette v. McMaster, 248 U. S. 465, 469-470 (1919), it is now established that the terms “citizen” and “resident” are “essentially interchangeable,” Austin v. New Hampshire, 420 U. S. 656, 662 n. 8 (1975), for purposes of analysis of most cases under the Privileges and Immunities Clause of Art. IV, § 2. See Toomer v. Witsell, 334 U. S. 385, 397 (1948). HICKLIN v. ORBECK 525 518 Opinion of the Court nated against nonresidents of Maryland in several ways: It required nonresident merchants to obtain licenses in order to practice their trade without requiring the same of certain similarly situated Maryland merchants; it charged nonresidents a higher license fee than those Maryland residents who were required to secure licenses; and it prohibited both resident and nonresident merchants from using nonresident salesmen, other than their regular employees, to sell their goods in the city. In holding that the statute violated the Privileges and Immunities Clause, the Court observed that “the clause plainly and unmistakably secures and protects the right of a citizen of one State to pass into any other State of the Union for the purpose of engaging in lawful commerce, trade, or business without molestation.” Id., at 430. Ward thus recognized that a resident of one State is constitutionally entitled to travel to another State for purposes of employment free from discriminatory restrictions in favor of state residents imposed by the other State. Again, Toomer v. Witsell, 334 U. S. 385 (1948), the leading modern exposition of the limitations the Clause places on a State’s power to bias employment opportunities in favor of its own residents, invalidated a South Carolina statute that required nonresidents to pay a fee 100 times greater than that paid by residents for a license to shrimp commercially in the three-mile maritime belt off the coast of that State. The Court reasoned that although the Privileges and Immunities Clause “does not preclude disparity of treatment in the many situations where there are perfectly valid independent reasons for it,” id., at 396, “ [i] t does bar discrimination against citizens of other States where there is no substantial reason for the discrimination beyond the mere fact that they are citizens of other States.” Ibid. A “substantial reason for the discrimination” would not exist, the Court explained, “unless there is something to indicate that noncitizens constitute a peculiar source of the evil at which the 526 OCTOBER TERM, 1977 Opinion of the Court 437U.S. [discriminatory] statute is aimed.” Id., at 398. Moreover, even where the presence or activity of nonresidents causes or exacerbates the problem the State seeks to remedy, there must be a “reasonable relationship between the danger represented by non-citizens, as a class, and the . . . discrimination practiced upon them.” Id., at 399. Toomer’s analytical framework was confirmed in Mullaney v. Anderson, 342 U. S. 415 (1952), where it was applied to invalidate a scheme used by the Territory of Alaska for the licensing of commercial fishermen in territorial waters; under that scheme residents paid a license fee of only $5 while nonresidents were charged $50. Even assuming that a State may validly attempt to alleviate its unemployment problem by requiring private employers within the State to discriminate against nonresidents—an assumption made at least dubious by Ward9— it is clear that under the Toomer analysis reaffirmed in Mullaney, Alaska Hire’s discrimination against nonresidents cannot withstand scrutiny under the Privileges and Immunities Clause. For although the statute may not violate the Clause if the State shows “something to indicate that noncitizens constitute a peculiar source of the evil at which the statute is aimed,” Toomer v. Witsell, supra, at 398, and, beyond this, the State “has no burden to prove that its laws are not violative of the , . . Clause,” Baldwin v. Montana Fish and Game Comm’n, 436 U. S., at 402 (Brennan, J., dissenting), certainly no showing was made on this record that nonresidents were “a peculiar source of the evil” Alaska Hire was enacted to remedy, namely, Alaska’s “uniquely high unemployment.” Alaska Stat. Ann. § 38.40.020 (1977). What evidence the record does contain indicates that the major cause of Alaska’s high unemployment was not the influx of nonresidents seeking employment, but rather the fact that a substantial number of Alaska’s jobless residents—especially the unemployed Eskimo and Indian residents—were unable to 9Cf. Edwards n. California, 314 U. S. 160 (1941). HICKLIN v. ORBECK 527 518 Opinion of the Court secure employment either because of their lack of education and job training or because of their geographical remoteness from job opportunities;10 and that the employment of nonresidents threatened to deny jobs to Alaska residents only to the extent that jobs for which untrained residents were being prepared might be filled by nonresidents before the residents’ training was completed. Moreover, even if the State’s showing is accepted as sufficient to indicate that nonresidents were “a peculiar source of evil,” Toomer and Mullaney compel the conclusion that Alaska Hire nevertheless fails to pass constitutional muster. For the discrimination the Act works against nonresidents does not bear a substantial relationship to the particular “evil” they are said to present. Alaska Hire simply grants all Alaskans, regardless of their employment status, education, or training, a flat employment preference for all jobs covered by the Act. A highly skilled and educated resident who has never been unemployed is entitled to precisely the same preferential treatment as the unskilled, habitually unemployed Arctic Eskimo enrolled in a job-training program. If 10 For example, a report quoted in the State’s Memorandum in Opposition to Plaintiffs’ Motion for Partial Preliminary Injunction and Second Motion for Preliminary Injunction, Record 58, observed: “The skill levels of in-migrants and seasonal workers are generally higher than those of the unemployed or under-employed resident workers. Their ability to command jobs in Alaska is a sympton of, rather than the cause of conditions resulting in high unemployment rates, particularly among Alaska Natives. Those who need the jobs the most tend to be undereducated, untrained, or living in areas of the state remote from job opportunities. Unless unemployed residents—most of whom are Eskimos and Indians—have access to job markets and receive the education and training required to fit them into Alaska’s increasingly technological economy and unless there is a restructuring of labor demands, new jobs will continue to be filled by persons from other states who have the necessary qualifications.” Federal Field Committee for Development Planning in Alaska, Economic Outlook for Alaska 311-312 (1971) (emphasis added; footnote omitted). 528 OCTOBER TERM, 1977 Opinion of the Court 437U.S. Alaska is to attempt to ease its unemployment problem by forcing employers within the State to discriminate against nonresidents—again, a policy which may present serious constitutional questions—the means by which it does so must be more closely tailored to aid the unemployed the Act is intended to benefit. Even if a statute granting an employment preference to unemployed residents or to residents enrolled in job-training programs might be permissible, Alaska Hire’s across-the-board grant of a job preference to all Alaskan residents clearly is not. Relying on McCready v. Virginia, 94 U. S. 391 (1877), however, Alaska contends that because the oil and gas that are the subject of Alaska Hire are owned by the State,11 this ownership, of itself, is sufficient justification for the Act’s discrimination against nonresidents, and takes the Act totally without the scope of the Privileges and Immunities Clause. As the State sees it “the privileges and immunities clause [does] not apply, and was never meant to apply, to decisions by the states as to how they would permit, if at all, the use and distribution of the natural resources which they own . . . .” Brief for Appellees 20 n. 14. We do not agree that the fact that a State owns a resource, of itself, completely removes a law concerning that resource from the prohibitions of the Clause. Although some courts, including the court below, have read McCready as creating an “exception” to the Privileges and Immunities Clause, we have just recently confirmed that “[i]n more recent years . . . the Court has recognized 11 At the time Alaska was admitted into the Union on January 3, 1959, 99% of all land within Alaska’s borders was owned by the Federal Government. In becoming a State, Alaska was granted and became entitled to select approximately 103 million acres of those federal lands. Alaska Statehood Law, 72 Stat. 340, § 6, note preceding 48 U. S. C. § 21. The selection process is not yet complete, but since 1959 large portions of land have been conveyed to the State, in fee, by the Federal Government. Full title to those lands and to the minerals on and below them is vested in the State. 72 Stat. 342, § 6 (i), note preceding 48 U. S. C. § 21. HICKLIN v. ORBECK 529 518 Opinion of the Court that the States’ interest in regulating and controlling those things they claim to 'own’ ... is by no means absolute.” Baldwin v. Montana Fish and Game Comm’n, 436 U. S., at 385. Rather than placing a statute completely beyond the Clause, a State’s ownership of the property with which the statute is concerned is a factor—although often the crucial factor—to be considered in evaluating whether the statute’s discrimination against noncitizens violates the Clause. Dispositive though this factor may be in many cases in which a State discriminates against nonresidents, it is not dispositive here. The reason is that Alaska has little or no proprietary interest in much of the activity swept within the ambit of Alaska Hire; and the connection of the State’s oil and gas with much of the covered activity is sufficiently attenuated so that it cannot justifiably be the basis for requiring private employers to discriminate against nonresidents. The extensive reach of Alaska Hire is set out in Alaska Stat. Ann. § 38.40.050 (a) (1977). That section provides: "The provisions of this chapter apply to all employment which is a result of oil and gas leases, easements, leases or right-of-way permits for oil or gas pipeline purposes, unitization agreements [12] or any renegotiation of any of the preceding to which the state is a party after July 7, 1972; however, the activity which generates the employment must take place inside the state and it must 12 The term "unitization agreement” is not defined in the Act. Alaska’s Commissioner of Natural Resources gave the following definition of the term: “Well, unitization agreement is an agreement between the operators and any given oil field as to the equity that each of them would have with respect to the oil and gas resources in that field. And in some cases that word is used to also include something called the ‘Plan of Operations’, which sets out the way in which an oil field or gas field would be operated pursuant to the State’s conservation laws.” Deposition of Guy R. Martin in No. 3025 (Sup. Ct. Alaska), p. 5. 530 OCTOBER TERM, 1977 Opinion of the Court 437 U. S. take place either on the property under the control of the person subject to this chapter or be directly related to activity taking place on the property under his control and the activity must be performed directly for the person subject to this chapter or his contractor or a subcontractor of his contractor or a supplier of his contractor or subcontractor.” (Emphasis added.) Under this provision, Alaska Hire extends to employers who have no connection whatsoever with the State’s oil and gas, perform no work on state land, have no contractual relationship with the State, and receive no payment from the State. The Act goes so far as to reach suppliers who provide goods or services to subcontractors who, in turn, perform work for contractors despite the fact that none of these employers may themselves have direct dealings with the State’s oil and gas or ever set foot on state land.13 Moreover, the Act’s coverage is not limited to activities connected with the extraction of Alaska’s oil and gas.14 It encompasses, as emphasized by the dissent below, “employment opportunities at refineries and in distribution systems utilizing oil and gas obtained under Alaska leases.” 565 P. 2d, at 171. The only limit of any consequence on the Act’s reach is the requirement that “the 13 According to one of the administrative regulations implementing Alaska Hire, “[s]uppliers shall have the same hiring requirements as an employer covered by this chapter, as to that portion of their supply business that is the result of a project or activity of a lessee, contractor or subcontractor.” 8 Alaska Admin. Code 35.080 (a) (1977). 14 The Commissioner of Natural Resources expressed this understanding of the scope of the Act: Mr. Martin: “. . . I think it would cover relationships such as anything on a work pad or an associated construction road or possibly a site for a support camp or construction camp.” Mr. Wagstaff (attorney for appellants): “What about things such as docks if shipping is being used?” Mr. Martin: “I would think that it could possibly include that.” Deposition of Guy R. Martin, supra, at 4. HICKLIN v. ORBECK 531 518 Opinion of the Court activity which generates the employment must take place inside the state.” Although the absence of this limitation would be noteworthy, its presence hardly is; for it simply prevents Alaska Hire from having what would be the surprising effect of requiring potentially covered out-of-state employers to discriminate against residents of their own State in favor of nonresident Alaskans. In sum, the Act is an attempt to force virtually all businesses that benefit in some way from the economic ripple effect of Alaska’s decision to develop its oil and gas resources to bias their employment practices in favor of the State’s residents. We believe that Alaska’s ownership of the oil and gas that is the subject matter of Alaska Hire simply constitutes insufficient justification for the pervasive discrimination against nonresidents that the Act mandates.15 Although appellants raise no Commerce Clause challenge to the Act, the mutually reinforcing relationship between the Privileges and Immunities Clause of Art. IV, § 2, and the Commerce Clause—a relationship that stems from their common 15 Heim v. McCall, 239 U. S. 175 (1915) and Crane v. New York, 239 U. S. 195 (1915)—if they have any remaining vitality, see Sugarman v. Dougall, 413 U. S. 634, 643-645 (1973); C. D. R. Enterprises, Ltd. V. Board of Education, 412 F. Supp. 1164 (EDNY 1976), summarily aff’d sub nom. Lefkowitz v. C. D. R. Enterprises, Ltd., 429 U. S. 1031 (1977)— do not suggest otherwise. In those cases, a New York statute that limited employment “in the construction of public works” to United States citizens and also required that an employment preference be given to New York citizens in such projects was upheld against challenges under both the Constitution and the Treaty of 1871 with Italy. Although the Art. IV, § 2, Privileges and Immunities Clause, along with the Due Process, Equal Protection, and Privileges and Immunities Clauses of the Fourteenth Amendment, was listed as one of the constitutional bases for attacking the statute, no out-of-state United States citizen challenged the law. As a consequence, both the appellants and the Court were concerned almost exclusively with the statute’s discrimination against resident aliens. This was reflected in the Court’s holding, which was limited to the Fourteenth Amendment and Treaty challenges and expressed no view on appellants’ passing Art. IV, § 2, privileges and immunities claim. 532 OCTOBER TERM, 1977 Opinion of the Court 437U.S. origin in the Fourth Article of the Articles of Confederation16 and their shared vision of federalism, see Baldwin v. Montana Fish and Game Comm’n, 436 U. S., at 379-380—renders several Commerce Clause decisions appropriate support for our conclusion. West v. Kansas Natural Gas, 221 U. S. 229 (1911), struck down an Oklahoma statutory scheme that completely prohibited the out-of-state shipment of natural gas found within the State. The Court reasoned that if a State could so prefer its own economic well-being to that of the Nation as a whole, “Pennsylvania might keep its coal, the Northwest its timber, [and] the mining States their minerals,” so that “embargo may be retaliated by embargo” with the result that “commerce [would] be halted at state lines.” Id., at 255. West was held to be controlling in Pennsylvania v. West Virginia, 262 U. S. 553 (1923), where a West Virginia statute that effectively required natural gas companies within the State to satisfy all fuel needs of West Virginia residents before transporting any natural gas out of the State was held to violate the Commerce Clause. West and Pennsylvania v. West Virginia thus established that the location in a given State of a resource bound for interstate commerce is an insufficient basis for preserving the benefits of the resource exclusively or even 16 That Article provided: “The better to secure and perpetuate mutual friendship and intercourse among the people of the different states in this union, the free inhabitants of each of these states, paupers, vagabonds and fugitives from justice excepted, shall be entitled to all privileges and immunities of free citizens in the several states; and the people of each State shall have free ingress and regress to and from any other State, and shall enjoy therein all the privileges of trade and commerce, subject to the same duties, impositions, and restrictions, as the inhabitants thereof respectively; provided, that such restrictions shall not extend so far as to prevent the removal of property, imported into any State, to any other State of which the owner is an inhabitant; provided, also that no imposition, duties or restriction, shall be laid by any State on the property of the United States, or either of them.” 9 Journal of the Continental Congress 908-909 (1777) (Library of Congress ed., 1907). HICKLIN v. ORBECK 533 518 Opinion of the Court principally for that State’s residents. Foster Packing Co. v. Hay del, 278 U. S. 1 (1928), went one step further; it limited the extent to which a State’s purported ownership of certain resources could serve as a justification for the State’s economic discrimination in favor of residents. There, in the face of Louisiana’s claim that the State owned all shrimp within state waters, the Court invalidated a Louisiana law that required the local processing of shrimp taken from Louisiana marshes as a prerequisite to their out-of-state shipment. The Court observed that “by permitting its shrimp to be taken and all the products thereof to be shipped and sold in interstate commerce, the State necessarily releases its hold and, as to the shrimp so taken, definitely terminates its control.” Id., at 13. West, Pennsylvania v. West Virginia, and Foster Packing thus establish that the Commerce Clause circumscribes a State’s ability to prefer its own citizens in the utilization of natural resources found within its borders, but destined for interstate commerce. Like Louisiana’s shrimp in Foster Packing, Alaska’s oil and gas here are bound for out-of-state consumption. Indeed, the construction of the Trans-Alaska Pipeline, on which project appellants’ nonresidency has prevented them from working, was undertaken expressly to accomplish this end.17 Although the fact that a state-owned resource is destined for interstate commerce does not, of itself, disable the State from preferring its own citizens in the utilization of that resource, it does inform analysis under the Privileges and Immunities Clause as to the permissibility of the discrimination the State visits upon nonresidents based on its ownership of the resource. Here, the oil and gas upon 17 In authorizing the construction of the Trans-Alaska Pipeline, Congress expressly found that “[t]he early development and delivery of oil and gas from Alaska’s North Slope to domestic markets is in the national interest because of growing domestic shortages and increasing dependence upon insecure foreign sources.” 43 U. S. C. § 1651 (a) (1970 ed., Supp. V) (emphasis added). 534 OCTOBER TERM, 1977 Opinion of the Court 437 U. S. which Alaska hinges its discrimination against nonresidents are of profound national importance.18 On the other hand, the breadth of the discrimination mandated by Alaska Hire goes far beyond the degree of resident bias Alaska’s ownership of the oil and gas can justifiably support. The confluence of these realities points to but one conclusion: Alaska Hire cannot withstand constitutional scrutiny. As Mr. Justice Cardozo observed in Baldwin v. G. A. F. Seelig, Inc., 294 U. S. 511, 523 (1935), the Constitution “was framed upon the theory that the peoples of the several states must sink or swim together, and that in the long run prosperity and salvation are in union and not division.”19 Reversed. 18 In enacting the Alaska Natural Gas Transportation Act of 1976, 15 U. S. C. § 719 et seq. (1976 ed.) Congress declared: “(1) a natural gas supply shortage exists in the contiguous States of the United States; “(2) large reserves of natural gas in the State of Alaska could help significantly to alleviate this supply shortage; “(3) the expeditious construction of a viable natural gas transportation system for delivery of Alaska natural gas to United States markets is in the national interest; and “(4) the determinations whether to authorize a transportation system for delivery of Alaska natural gas to the contiguous States and, if so, which system to select, involve questions of the utmost importance respecting national energy policy, international relations, national security, and economic and environmental impact, and therefore should appropriately be addressed by the Congress and the President in addition to those Federal officers and agencies assigned functiops under law pertaining to the selection, construction, and initial operation of such a system.” 15 U. S. C. § 719 (1976 ed.). See n. 17, supra. 19 In light of our conclusion that Alaska Hire is invalid under the Privileges and Immunities Clause of Art. IV, § 2, we have no occasion to address appellants’ challenges to the Act under the Equal Protection Clause of the Fourteenth Amendment. WISE v. LIPSCOMB 535 Syllabus WISE, MAYOR OF DALLAS, et al. v. LIPSCOMB et al. CERTIORARI TO THE UNITED STATES COURT OF APPEALS FOR THE FIFTH CIRCUIT No. 77-529. Argued April 26, 1978—Decided June 22, 1978 Respondents, Negro and Mexican-American residents of Dallas, Tex., brought this action for injunctive and declaratory relief against petitioners, the Mayor and members of the Dallas City Council, alleging that the City Charter’s at-large system of electing council members unconstitutionally diluted the vote of racial minorities. After an evidentiary hearing, the District Court orally declared that system unconstitutional and then “afforded the city an opportunity as a legislative body for the City of Dallas to prepare a plan which would be constitutional.” The City Council then passed a resolution expressing its intention to enact an ordinance that would provide for eight council members to be elected from single-member districts and for the three remaining members, including the Mayor, to be elected at large. After an extensive remedy hearing, the District Court approved the plan, which the City Council thereafter formally enacted as an ordinance. The District Court later issued a memorandum opinion that sustained the plan as a valid legislative Act. The Court of Appeals reversed, holding that the District Court had erred in evaluating the plan only under constitutional standards without also applying the teaching of East Carroll Parish School Bd. v. Marshall, 424 U. S. 636, which held that, absent exceptional circumstances, judicially imposed reapportionment plans should use only single-member districts. Held: The judgment is reversed and the case is remanded. Pp. 539-547; 547-549. 551 F. 2d 1043, reversed and remanded. Mr. Justice White, joined by Mr. Justice Stewart, concluded: 1. Federal courts, absent special circumstances, must employ single-member districts when they impose remedial reapportionment plans. That standard, however, is more stringent than the constitutional standard that is applicable when the reapportionment is accomplished by the legislature. Here, after the District Court had invalidated the Dallas at-large election scheme in the City Charter, the city discharged its duty to devise a substitute by enacting the eight/three ordinance, which the District Court reviewed as a legislatively enacted plan and held constitutional despite the use of at-large voting for three council seats. Pp. 539-543. 536 OCTOBER TERM, 1977 Syllabus 437 U. 8. 2. The eight/three ordinance was properly considered to be a legislative plan and the Court of Appeals erred in evaluating it under principles applicable to judicially devised reapportionment plans. Pp. 543-546. (a) No special reason for not applying the standard applicable to a legislatively devised plan can be found in the provisions of Texas law that specify that a city charter can be amended only by a vote of the people, for the City Council in enacting the plan did not purport to amend the Charter but only to exercise its legislative powers after the Charter provision had been declared unconstitutional. P. 544. (b) East Carroll Parish School Bd., supra, does not support the conclusion of the Court of Appeals that the plan presented by the city must be viewed as judicial and therefore as subject to a level of scrutiny more stringent than that required by the Constitution, rather than legislative. In reaching the conclusion that singe-member districts are to be preferred, the Court emphasized that the bodies that submitted the plans did not purport to reapportion themselves and could not legally do so under federal law because state legislation providing them with such powers had been disapproved under § 5 of the Voting Rights Act of 1965. On the facts of the instant case, however, unlike the situation in East Carroll Parish School Bd., the Dallas City Council validly met its responsibility of replacing the invalid apportionment provision with one that could withstand constitutional scrutiny. Pp. 545-546. 3. Though it has been urged that § 5 of the Voting Rights Act of 1965, which became applicable to Texas while this case was pending on appeal, barred effectuation of the challenged ordinance absent the clearance mandated by § 5, that issue was not dealt with by the Court of Appeals and should more appropriately be considered by that court on remand. Pp. 546-547. Mr. Justice Powell, joined by The Chief Justice, Mr. Justice Blackmun, and Mr. Justice Rehnquist, while agreeing that the eight/three ordinance was a “legislative plan” for purposes of federal-court review, concluded that the instant case is controlled by Bums v. Richardson, 384 U. 8. 73. By analogy to the reasoning of that case the eight/three plan must be considered legislative, even if the Council had no power to apportion itself, a Charter amendment being necessary to that end. Under the Bums rule whereby “a State’s freedom of choice to devise substitutes for an .apportionment plan found unconstitutional . . . should not be restricted beyond the clear commands of the Equal Protection Clause,” plans proposed by the local body must be regarded as “legislative” even if, as in that case, the Court’s examination of state law suggests that the local body lacks authority to reapportion WISE v. LIPSCOMB 537 535 Opinion of White, J. itself. To the extent that East Carroll Parish School Bd. implies anything further about the principle established in Bums, the latter must be held to control. Pp. 547-549. White, J., announced the Court’s judgment and delivered an opinion, in which Stewart, J., joined. Powell, J., filed an opinion concurring in part and concurring in the judgment, in which Burger, C. J., and Blackmun and Rehnquist, JJ., joined, post, p. 547. Rehnquist, J., filed a separate opinion, in which Burger, C. J., and Stewart and Powell, J J., joined, post, p. 549. Marshall, J., filed a dissenting opinion, in which Brennan and Stevens, JJ., joined, post, p. 550. Joseph G. Werner argued the cause for petitioners. With him on the brief was Lee E. Holt. James A. Johnston argued the cause for respondents. With him on the brief were Edward B. Cloutman III and Walter L. Irvin. Joaquin G. Avila, Vilma S. Martinez, and Morris J. Baller filed a brief for respondents Callejo et al. Peter Buscemi argued the cause pro hac vice for the United States as amicus curiae urging affirmance. With him on the brief were Solicitor General McCree, Assistant Attorney General Days, Brian K. Landsberg, and. Robert J. Reinstein.* Mr. Justice White announced the judgment of the Court and delivered an opinion in which Mr. Justice Stewart joined. This case involves the recurring issue of distinguishing between legislatively enacted and judicially imposed reapportionments of state legislative bodies. I In 1971 respondents, Negro and Mexican-American residents of Dallas, Tex., filed suit in the United States District *Charles A. Bane, Thomas D. Barr, Armand Derjner, Norman Redlich, Frank R. Parker, Thomas J. Ginger, Robert A. Murphy, Norman J. Chachkin, and William E. Caldwell filed a brief for the Lawyers Committee for Civil Rights Under Law as amicus curiae urging affirmance. 538 OCTOBER TERM, 1977 Opinion of White, J. 437U.S. Court for the Northern District of Texas against petitioners, the Mayor and members of the City Council of Dallas, the city’s legislative body, alleging that the at-large system of electing council members unconstitutionally diluted the vote of racial minorities. They sought a declaratory judgment to this effect and an injunction requiring the election of councilmen from single-member districts. The complaint was dismissed for failure to state a claim, but the Court of Appeals for the Fifth Circuit disagreed and remanded. Lipscomb v. Jonsson, 459 F. 2d 335 (1972). On January 17, 1975, after certifying a plaintiff class consisting of all Negro citizens of the city of Dallas1 and following an evidentiary hearing, the District Court orally declared that the system of at-large elections to the Dallas City Council unconstitutionally diluted the voting strength of Negro citizens.2 The District Court then “afforded the city an opportunity as a legislative body for the City of Dallas to prepare a plan which would be constitutional.” App. 29. On January 20, 1975, the City Council passed a resolution which stated that the Council intended to enact an ordinance which would provide for eight Council members to be elected from single-member districts and for the three remaining members, including the Mayor, to be elected at-large. This plan was submitted to the District Court on January 24, 1975. The court then conducted a remedy hearing “to determine the constitutionality of the new proposed plan by the City of Dallas.” Ibid. After an extensive hearing, the court announced in an oral opinion delivered on February 8, 1975, that the city’s plan met constitutional guidelines and was ac- 1 Several plaintiffs, including all of the Mexican-American plaintiffs, were dismissed from the case for failure to respond to interrogatories. Two Mexican-Americans subsequently attempted to intervene. The District Court denied their application but later permitted several Mexican-Americans to participate in the remedy hearing held after the at-large election system was declared unconstitutional. 2 Petitioners did not appeal this ruling and do not question it here. WISE v. LIPSCOMB 539 535 Opinion of White, J. ceptable and that it would issue a written opinion in the near future. Two days later, the City Council formally enacted the promised ordinance, and on March 25, the court issued a memorandum opinion containing its findings of fact and conclusions of law and again sustaining the city plan as a valid legislative Act. 399 F. Supp. 782 (1975).3 The Court of Appeals reversed. 551 F. 2d 1043 (1977). It held that the District Court erred by evaluating the city’s actions only under constitutional standards rather than also applying the teaching of East Carroll Parish School Bd. n. Marshall, 424 U. S. 636 (1976), that, absent exceptional circumstances, judicially imposed reapportionment plans should employ only single-member districts. It concluded that no considerations existed in this case which justified a departure from this preference and remanded with instructions that the District Court require the city to reapportion itself into an appropriate number of single-member districts.4 We granted certiorari, 434 U. S. 1008 (1978), and reverse on the grounds that the Court of Appeals misapprehended East Carroll Parish School Bd. and its predecessors. II The Court has repeatedly held that redistricting and reapportioning legislative bodies is a legislative task which the federal courts should make every effort not to pre-empt. Connor v. Finch, 431 U. S. 407, 414-415 (1977); Chapman v. Meier, 420 U. S. 1, 27 (1975); Gaffney v. Cummings, 412 U. S. 735, 749 (1973); Burns v. Richardson, 384 U. S. 73, 84-85 3 On April 1, 1975, the Dallas City Council election was held under the eight/three plan. During the pendency of the appeal the electorate approved this plan in a referendum conducted in April 1976, thus incorporating it into the City Charter. 4 The court stated that the city may provide for the election of the Mayor by general citywide election if it desired. Mr. Justice Powell stayed the Court of Appeals’ judgment pending disposition by this Court. 434 U. S. 1329 (1977). 540 OCTOBER TERM, 1977 Opinion of White, J. 437U.S. (1966). When a federal court declares an existing apportionment scheme unconstitutional, it is therefore, appropriate, whenever practicable, to afford a reasonable opportunity for the legislature to meet constitutional requirements by adopting a substitute measure rather than for the federal court to devise and order into effect its own plan. The new legislative plan, if forthcoming, will then be the governing law unless it, too, is challenged and found to violate the Constitution. “[A] State’s freedom of choice to devise substitutes for an apportionment plan found unconstitutional, either as a whole or in part, should not be restricted beyond the clear commands of the Equal Protection Clause.” Id., at 85. Legislative bodies should not leave their reapportionment tasks to the federal courts; but when those with legislative responsibilities do not respond, or the imminence of a state election makes it impractical for them to do so, it becomes the “unwelcome obligation,” Connor v. Finch, supra, at 415, of the federal court to devise and impose a reapportionment plan pending later legislative action. In discharging this duty, the district courts “will be held to stricter standards . . . than will a state legislature . . . .” 431 U. S., at 414. Among other requirements, a court-drawn plan should prefer singlemember districts over multimember districts, absent persuasive justification to the contrary. Connor v. Johnson, 402 U. S. 690, 692 (1971). We have repeatedly reaffirmed this remedial principle. Connor v. Williams, 404 U. S. 549, 551 (1972); Mahan n. Howell, 410 U. S. 315, 333 (1973); Chapman v. Meier, supra, at 18; East Carroll Parish School Bd. n. Marshall, supra, at 639. The requirement that federal courts, absent special circumstances, employ single-member districts when they impose remedial plans, reflects recognition of the fact that “the practice of multimember districting can contribute to voter confusion, make legislative representatives more remote from their constituents, and tend to submerge electoral minorities and overrepresent electoral majorities . . . .” Connor n. WISE v. LIPSCOMB 541 535 Opinion of White, J. Finch, supra, at 415. See also Chapman v. Meier, supra, at 15-16. Despite these dangers, this Court has declined to hold that state multimember districts are per se unconstitutional. See, for example, Whitcomb v. Chavis, 403 U. S. 124 (1971); Fortson v. Dorsey, 379 U. S. 433 (1965); Bums v. Richardson, supra; Chapman v. Meier, supra, at 15. A more stringent standard is applied to judicial reapportionments, however, because a federal court, “lacking the political authoritativeness that the legislature can bring to the task,” must act “circumspectly, and in a manner ‘free from any taint of arbitrariness or discrimination.’ ” Connor n. Finch, supra, at 415, quoting from Roman v. Sincock, 377 U. S. 695, 710 (1964).5 The foregoing principles, worked out in the course of reconciling the requirements of the Constitution with the goals of state political policy, are useful guidelines and serve to decide many cases. But, as is true in this case, their application to the facts presented is not always immediately obvious. Furthermore, the distinctive impact of § 5 of the Voting Rights Act of 1965, as amended, 89 Stat. 404, 42 U. S. C. § 1973c (1970 ed., Supp. V), upon the power of the 5 The numerous cases in which this Court has required the use of singlemember districts in court-ordered reapportionment plans have all involved apportionment schemes which, unlike the one in this case, were held unconstitutional because they departed from the one-person, one-vote rule of Reynolds n. Sims, 377 U. S. 533 (1964), and its progeny. We are fully persuaded, however, that the same considerations which have induced this Court to express a preference for single-member districts in court-ordered reapportiomnent plans designed to remedy violations of the one-person, one-vote rule compel a similar rule with regard to court-imposed reapportionments designed to cure the dilution of the voting strength of racial minorities resulting from unconstitutional racial discrimination. Indeed, the Court has justified the preference for single-member districts in judicially imposed reapportionments on the ground that multimember districts “tend to submerge electoral minorities and overrepresent electoral majorities . . . ,” which is the source of the very violation which the court is seeking to eliminate in racial dilution cases. Connor v. Finch, 431 U. S. 407, 415 (1977). See White v. Regester, 412 U. S. 755, 765-770 (1973). 542 OCTOBER TERM, 1977 Opinion of White, J. 437U.S. States to reapportion themselves must be observed. Plans imposed by court order are not subject to the requirements of § 5,6 but under that provision, a State or political subdivision subject to the Act may not “enact or seek to administer” any “different” voting qualification or procedure with respect to voting without either obtaining a declaratory judgment from the United States District Court for the District of Columbia that the proposed change “does not have the purpose and will not have the effect of denying or abridging the right to vote on account of race or color” or submitting the change to the Attorney General and affording him an appropriate opportunity to object thereto. A new reapportionment plan enacted by a State, including one purportedly adopted in response to invalidation of the prior plan by a federal court, will not be considered “effective as law,” Connor v. Finch, 431 U. S., at 412; Connor v. Waller, 421 U. S. 656 (1975), until it has been submitted and has received clearance under § 5. Neither, in those circumstances, until clearance has been obtained, should a court address the constitutionality of the new measure. Connor v. Finch, supra; Connor v. Waller, supra. Pending such submission and clearance, if a State’s electoral processes are not to be completely frustrated, federal courts will at times necessarily be drawn further into the reapportionment process and required to devise and implement their own plans. Ill Texas was not subject to the Voting Rights Act when this case was pending in the District Court. Hence, insofar as federal law was concerned, when the District Court invalidated the provisions of the Dallas City Charter mandating at-large Council elections, the city was not only free but was expected to devise a substitute rather than to leave the matter 6 “A decree of the United States District Court is not within reach of Section 5 of the Voting Rights Act.” Connor v. Johnson, 402 U. S. 690, 691 (1971). WISE v. LIPSCOMB 543 535 Opinion of White, J. to the District Court. This duty, the District Court found, was discharged when the city enacted the eight/three plan of electing Council members. Noting that only if “the legislature failed in [its reapportionment] task, would the responsibility fall to the federal courts” and declaring that the plan adopted by the Council was not one “hastily conceived merely for the purposes of this litigation,” 399 F. Supp., at 797, the District Court proceeded to declare the plan constitutional despite the use of at-large voting for three Council seats. Although there are some indications in the District Court’s opinion that it was striving to satisfy those rules governing federal courts when they devise their own reapportionment plans, it seems to us that on balance, the District Court, as the United States observes in its amicus brief, reviewed the apportionment plan proposed by the Council as a legislatively enacted plan.7 The Court of Appeals was not in disagreement in this respect. It observed that “[t]he district court approved the City’s plan for relief, which was enacted as a city ordinance following the court’s decision that the prior system was unconstitutional.” 551 F. 2d, at 1045. It further noted that “the election plan [was] formally adopted by the City Council.” Id., at 1046. Neither did the Court of Appeals disturb the ruling of the District Court that the ordinance was constitutional. It did, however, insist that the plan also satisfy the special preference for single-member districts applicable where district courts are themselves put to the task of devising reapportionment plans and reversed the judgment of the District Court because in its view the record did not disclose the presence of those special circumstances that would warrant departure from the 7 In his oral announcement, the judge remarked: “I’m not saying it’s the best plan. It’s not even the plan that this Court would have drawn. But this Court’s not in the plan-drawing business. That’s the legislative duty.” Record 195. 544 OCTOBER TERM, 1977 Opinion of White, J. 437U.S. rule. This was clearly error unless there was some convincing reason why the District Court was not entitled to consider the substitute plan under the principles applicable to legislatively adopted reapportionment plans. As we see it, no such reason has been presented. It is suggested that the city was without power to enact the ordinance because the at-large system declared unconstitutional was established by the City Charter and because, under the Texas Constitution, Art. XI, § 5, and Texas statutory law, Tex. Rev. Civ. Stat. Ann., Art. 1170 (Vernon Supp. 1978), the Charter cannot be amended without a vote of the people. But the District Court was of a different view. Although the Council itself had no power to change the at-large system as long as the Charter provision remained intact, once the Charter provision was declared unconstitutional, and, in effect, null and void, the Council was free to exercise its legislative powers which it did by enacting the eight/three plan. 399 F. Supp., at 800; Tr. of Oral Arg. 6. When the City Council reapportioned itself by means of resolution and ordinance, it was not purporting to amend the City Charter but only to exercise its legislative powers as Dallas’ governing body. The Court of Appeals did not disagree with the District Court in this respect, and we are in no position to overturn the District Court’s acceptance of the city ordinance as a valid legislative response to the court’s declaration of unconstitutionality.8 8 The record suggests no statutory, state constitutional, or judicial prohibition upon the authority of the City Council to enact a municipal election plan under circumstances such as this and respondents have been unable to cite any support for its contention that the City Council exceeded its authority. It must be noted that since there is no provision under Texas law for reapportionment of Home Rule cities such as Dallas by the state legislature, or other state agency, acceptance of respondents’ position would leave Dallas utterly powerless to reapportion itself in those instances where the time remaining before the next scheduled election is too brief to permit the approval of a new plan by referendum. We are WISE v. LIPSCOMB 545 535 Opinion of White, J. East Carroll Parish School Bd. v. Marshall does not support the conclusion of the Court of Appeals in this case that the plan presented by the city must be viewed as judicial rather than legislative. In that case the District Court instructed the East Carroll police jury and school boards to file reapportionment plans. They both submitted a multimember arrangement which the court adopted. We held that the District Court erred in approving a multimember plan because “when United States district courts are put to the task of fashioning reapportionment plans to supplant concededly invalid state legislation, single-member districts are to be preferred absent unusual circumstances.” 424 U. S., at 639. In reaching this conclusion, however, we emphasized that the bodies which submitted the plans did not purport to reapportion themselves and, furthermore, could not even legally do so under federal law because state legislation providing them with such powers had been disapproved by the Attorney General of the United States under § 5 of the Voting Rights Act of 1965. 424 U. S., at 638 n. 6, 637 n. 2. Under these circumstances, it was concluded that the mere act of submitting a plan was not the equivalent of a legislative Act of reapportionment performed in accordance with the political processes of the community in question. Even if one disagreed with that conclusion, this case is markedly different from East Carroll Parish School Bd. After the District Court found that the existing method of electing the City Council was constitutionally defective on January 17, 1975, it “gave the City of Dallas an opportunity to perform its duty to enact a constitutionally acceptable plan.” 399 F. Supp., at 792. The City Council, the legislative body governing Dallas, promptly took advantage of this opportunity and on January 24, 1975, passed a resolution which stated “that it unwilling to adopt such an interpretation of Texas and Dallas law in the absence of any indication whatsoever that it would be accepted by Texas courts. 546 OCTOBER TERM, 1977 Opinion of White, J. 437U.S. is the intention of the majority of this City Council to pass an ordinance [enacting a plan of eight single-member districts with three individuals, including the Mayor, to be elected at-large].” App. 188. On February 8, 1975, the District Court announced in an oral opinion following a hearing held to consider the constitutionality of the city’s plan that it was accepting the city’s plan but retained jurisdiction. Two days later, on February 10, the City Council, as promised, enacted an ordinance incorporating the eight/three plan. Id., at 189. In a written opinion filed subsequently, the District Court specifically found “that [the city of Dallas] has met [its constitutional] duty in enacting the eight/three plan of electing council members.” 399 F. Supp., at 792. Here, unlike the situation in East Carroll Parish School Bd., as the Court there viewed it, the body governing Dallas validly met its responsibility of replacing the apportionment provision invalidated by the District Court with one which could survive constitutional scrutiny. The Court of Appeals therefore erred in regarding the plan as court imposed and in subjecting it to a level of scrutiny more stringent than that required by the Constitution.9 Finally, it is urged that the Court of Appeals be affirmed because Texas became subject to § 5 of the Voting Rights Act while the case was pending on appeal and because under § 5, as amended, Dallas could neither enact nor seek to administer any reapportionment plan different from that in effect on November 1, 1972, without securing the clearance called for by that section. It is urged that the city ordinance of February 1975, relied upon by the District Court and 9 In light of our disposition, we do not consider petitioners’ claim that the Court of Appeals also erred in holding that the alleged effect of all single-member districts on the representation of Mexican-American voters and the desirability of permitting some citywide representation did not constitute special circumstances justifying departure from the preference for single-member districts in remedial reapportionments conducted by federal courts. WISE v. LIPSCOMB 547 535 Opinion of Powell, J. validly enacted prior to § 5’s becoming applicable to Texas, cannot be considered as effective law until it has secured the necessary approval. The same is said with respect to the Charter amendment approved by the people of Dallas in 1976. See n. 3, supra. We think it inappropriate, however, to address the § 5 issue. Respondents may, of course, seek to sustain the judgment below on grounds not employed by the Court of Appeals; but there is a preliminary question as to whether the § 5 issue is open in this Court. Respondents did not crosspetition, and sustaining the § 5 submission, even if it would not expand the relief in respondents’ favor, would alter the nature of the judgment issued by the Court of Appeals. See United States v. New York Telephone Co., 434 U. S. 159, 166 n. 8 (1977). In any event, however, we are not obligated to address the issue here, particularly where the Court of Appeals did not deal with it one way or another—apparently because it considered the plan to be a judicial product beyond the reach of the section. The impact of the Voting Rights Act on the city ordinance and on the Charter amendment approved by referendum will be open on remand, and we deem it appropriate for the Court of Appeals to deal with these questions. The judgment of the Court of Appeals is reversed, and the case is remanded to that court for further proceedings. So ordered. Mr. Justice Powell, with whom The Chief Justice, Mr. Justice Blackmun, and Mr. Justice Rehnquist join, concurring in part and concurring in the judgment. I agree with Mr. Justice White’s conclusion that the reapportionment plan adopted by the Dallas City Council was a “legislative plan” for purposes of review by a federal court. In my view, however, his reasoning in reaching that conclusion casts doubt on Burns n. Richardson, 384 U. S. 73 (1966). 548 OCTOBER TERM, 1977 Opinion of Powell, J. 437U.S. Mr. Justice White reads East Carroll Parish School Bd. v. Marshall, 424 U. S. 636 (1976), as establishing the principle that a proposed reapportionment plan cannot be considered a legislative plan if the political body suggesting it lacks legal power to reapportion itself. Ante, at 545. Because the City Council ordinarily would have had no power to reapportion itself—a Charter amendment being necessary to that end— Mr. Justice White is constrained to assume that the Council became imbued with such power after the District Court struck down the apportionment provisions of the City Charter. Aside from the fact that this aspect of Texas law was neither fully briefed nor argued, the assumption seems unnecessary. In Burns N. Richardson, supra, the Hawaii Legislature was without power to reapportion itself, a constitutional amendment being required for that purpose. Nevertheless, this Court treated the plan that the legislature proposed to submit to the voters as a legislative plan. By parity of reasoning, the plan proposed by the Dallas City Council in this case must be considered legislative, even if the Council had no power to reapportion itself. The Council plan was then implemented by court order, 399 F. Supp. 782, 798 (ND Tex. 1975), just as the legislature’s plan in Bums ultimately was imposed pending the outcome of the constitutional amendment process, 384 U. S., at 98. The essential point is that the Dallas City Council exercised a legislative judgment, reflecting the policy choices of the elected representatives of the people, rather than the remedial directive of a federal court. As we held in Burns, supra, at 85, “a State’s freedom of choice to devise substitutes for an apportionment plan found unconstitutional, either as a whole or in part, should not be restricted beyond the clear commands of the Equal Protection Clause.” This rule of deference to local legislative judgments remains in force even if, as in Burns, our examination of state law suggests that the local body lacks authority to reapportion itself. WISE v. LIPSCOMB 549 535 Opinion of Rehnquist, J. Thus, Mr. Justice White’s statement that East Carroll School Bd. stands for the proposition that a plan submitted by a political body without power to reapportion itself cannot be considered a legislative plan appears to be in direct conflict with Burns. Because the brief per curiam in East Carroll did not even cite Bums, I would read it as turning on its peculiar facts. In response to the litigation in East Carroll, the legislature enacted a statute enabling police juries and school boards to reapportion themselves by employing at-large elections. That enabling legislation was disapproved by the Attorney General of the United States under § 5 of the Voting Rights Act of 1965, as amended, 42 U. S. C. § 1973c (1970 ed., Supp. V), because of its impermissible impact on Negro voters. This determination meant that the specific plans proposed by the school board and police jury in that case would have had unlawful effects. Because their legislative judgment had been found tainted in that respect, it followed that the normal presumption of legitimacy afforded the balances reflected in legislative plans, see Bums, supra, at 84-85, could not be indulged. To the extent that East Carroll implies anything further about the principle established in Bums, the latter must be held to control. Having determined on the basis of Bums that the City Council plan was legislative, I agree with Mr. Justice White’s conclusion that the judgment of the Court of Appeals must be reversed. I also agree that there is no reason for this Court to explore difficult questions concerning § 5 of the Voting Rights Act in the absence of consideration by the courts below. Opinion of Mr. Justice Rehnquist, with whom The Chief Justice, Mr. Justice Stewart, and Mr. Justice Powell join. I write separately to emphasize that the Court today is not presented with the question of whether the District Court erred in concluding that the form of government of the city of 550 OCTOBER TERM, 1977 Marshall, J., dissenting 437 U. S. Dallas unconstitutionally diluted the voting power of black citizens. While this Court has found that the use of multimember districts in a state legislative apportionment plan may be invalid if “used invidiously to cancel out or minimize the voting strength of racial groups,” White v. Regester, 412 U. S. 755, 765 (1973), we have never had occasion to consider whether an analogue of this highly amorphous theory may be applied to municipal governments. Since petitioners did not preserve this issue on appeal, we need not today consider whether relevant constitutional distinctions may be drawn in this area between a state legislature and a municipal government. I write only to point out that the possibility of such distinctions has not been foreclosed by today’s decision. Mr. Justice Marshall, with whom Mr. Justice Brennan and Mr. Justice Stevens join, dissenting. I agree with the majority’s decision not to reach the Voting Rights Act question, since it was not presented to either of the courts below. I also agree with the analysis of our past decisions found in Part II of Mr. Justice White’s opinion. I cannot agree, however, that the actions of the Dallas City Council are distinguishable from those of the local governing body in East Carroll Parish School Bd. v. Marshall, 424 U. S. 636 (1976). I therefore conclude that the plan ordered by the District Court here must be evaluated in accordance with the federal common law of remedies applicable to judicially devised reapportionment plans. I In East Carroll Parish School Bd. v. Marshall, supra, suit against the parish (county) was initially brought by a white resident who claimed that population disparities among the wards of the parish unconstitutionally denied him an equal vote in elections for members of the school board and the police jury, the governing body of the parish. Following a WISE v. LIPSCOMB 551 535 Marshall, J., dissenting finding of unconstitutionality, the District Court adopted a plan submitted by the police jury, which called for at-large elections of both bodies. Two years later (after the 1970 census), in response to the court’s direction, the at-large plan was resubmitted by the police jury. Respondent Marshall then intervened, arguing that the at-large elections would dilute the Negro vote in violation of the Fourteenth and Fifteenth Amendments. The District Court again accepted the police jury plan, but the Court of Appeals reversed, holding that multimember districts were unconstitutional. Although we did not reach the constitutional ground relied on by the Court of Appeals, we sustained its judgment. We concluded that the District Court had abused its equitable discretion in not requiring the division of the parish into singlemember wards: “We have frequently reaffirmed the rule that when United States district courts are put to the task of fashioning reapportionment plans to supplant concededly invalid state legislation, single-member districts are to be preferred absent unusual circumstances.” 424 U. S., at 639. It is plain from the foregoing that we treated the plan submitted by the local legislative body in East Carroll as a judicially devised plan, to which the federal common law of remedies developed in reapportionment cases was applicable. It is equally plain that we did not treat the police jury’s submission as a “legislatively enacted” plan, which would only have had to meet the strictures of the Constitution and would not necessarily have been subject to evaluation under the more stringent standards applicable to court-devised plans. See Connor v. Finch, 431 U. S. 407, 41-4-415 (1977). Indeed, in rejecting the argument of the United States (appearing as amicus curiae) that the East Carroll plan was subject to the preclearance procedure of § 5 of the Voting Rights Act of 1965, we expressly noted that the police jury “did not have the authority to reapportion itself,” and that the plan, though sub 552 OCTOBER TERM, 1977 Marshall, J., dissenting 437U.S. mitted by the police jury, was a “court-ordered pla[n] resulting from equitable jurisdiction over the adversary proceedings.” 424 U. S., at 638-639, n. 6. There is no meaningful distinction between the facts here and the facts in East Carroll. Like the police jury in East Carroll, the City Council of Dallas did not act pursuant to any state enabling legislation governing the procedures for reapportioning itself when it first proposed the eight/three plan to the District Court in January 1975. Nor did it act pursuant to any state-derived authority when it “enacted” the plan following the District Court’s first approval of it in March 1975. Under the terms of its Charter, the Dallas City Council could reapportion itself only by a popular referendum. See Tex. Const., Art. XI, § 5; Tex. Rev. Civ. Stat. Ann., Art. 1170 (Vernon Supp. 1978). The Council unquestionably failed to comply with the existing state procedures for enacting a reapportionment plan; indeed, the District Court itself noted that, were the Dallas City Council not responding to a judicial finding of unconstitutionality, it would have been acting unlawfully in unilaterally reapportioning itself. 399 F. Supp. 782, 800 (ND Tex. 1975). That this plan was not devised by the City Council in the usual course of its legislative responsibilities is further evidenced by the fact that the Council told a group of Mexican-American citizens, who wished to present for the Council’s deliberations an alternative, single-member district plan, that they were in the “wrong forum” and should go to federal court. App. 43-44. It seems clear that the eight/three plan was proposed less as a matter of legislative judgment than as a response by a party litigant to the court’s invitation to aid in devising a plan. Indeed, the District Court itself appeared at times to regard the eight/three plan as a court-devised plan in which at-large voting had to be justified by special and unique circumstances. See ante, at 543 (opinion of White, J.). WISE v. LIPSCOMB 553 535 Marshall, J., dissenting It is suggested that the City Council here, unlike the police jury in East Carroll, purported to reapportion itself when it first submitted the eight/three plan. See ante, at 545 (opinion of White, J.). But that simply is not the case. This plan was initially proposed not in the form of a formal, binding enactment but merely as an expression of the Council’s “intention.” App. 188. The Council did not even bother to go through the formality of enacting a supposedly binding ordinance until after the District Court, following a full hearing, indicated that it approved of the plan as a remedy for the constitutional violations; the procedures followed prior to the time when the District Court ordered implementation of the eight/three plan, moreover, were insufficient under state law validly to change the structure of the Council. While our past decisions have held that a legislatively enacted reapportionment plan is the preferred response to a judicial finding of unconstitutional apportionment, I do not believe that these cases contemplated that a legislature could meet this responsibility—and thereby avoid the requirements applicable to court-devised plans—by making a submission not in accordance with valid state procedures governing legislative enactments.1 If the plan submitted in East Carroll was properly regarded as a judicially devised plan, 11 do not agree with my Brother Powell that Burns v. Richardson, 384 U. S. 73 (1966), stands for the proposition that any legislative submission whatsoever should be treated as a “legislative plan.” In Burns, the very mechanism by which changes in apportionment could be made under state law had been found by the District Court to be designed to freeze existing unconstitutional apportionments and had thus been held unconstitutional in its own right. 238 F. Supp. 468, 472 (Haw. 1965). Here, by contrast, there was a lawful mechanism available for modifying the apportionment under the Dallas City Charter: the drafting of a proposal by the Council and its submission to the voters of the city at a popular referendum. If this process could not be completed in time for the next election, then the District Court would be justified in devising a temporary, court-ordered plan. See ante, at 540 (opinion of White, J.). See also Connor v. Williams, 404 U. S. 549, 552, and n. 4 (1972). 554 OCTOBER TERM, 1977 Marshall, J., dissenting 437 U. S. then the plan before us today must also be so regarded, and I see no reason to depart from the clear implications of this unanimous decision of the Court rendered only two Terms ago. I therefore conclude that the Court of Appeals properly evaluated this plan under the standards of the federal common law, which has for years recognized that multimember districts and at-large voting are presumptively disfavored. II Even if this plan were properly to be viewed as a “legislatively enacted” plan, however, the majority’s apparent assumption that it represents a proper remedy would nonetheless be troubling. Where the very nature of the underlying violation is dilution of the voting power of a racial minority resulting from the effects of at-large voting in a particular political community, I believe that it is inappropriate either for the local legislative body or a court to respond with more of the same. Although we have refrained from holding that multimember districts are unconstitutional per se, the presumption in favor of single-member districts as a matter of federal remedial law is a strong one. See, e. g., Connor n. Johnson, 402 U. S. 690 (1971); Connor n. Williams, 404 U. S. 549, 551 (1972); Chapman n. Meier, 420 U. S. 1, 16-19 (1975). We have repeatedly explained this preference by virtue of the fact that multimember districts “tend to submerge electoral minorities and overrepresent electoral majorities.” Connor n. Finch, 431 U. S., at 415; accord, Whitcomb n. Chavis, 403 U. S. 124, 158-159 (1971). See also Chapman v. Meier, supra, at 16. In the instant case, it is essentially undisputed that the use of a multimember district (the city of Dallas) for the at-large election of all City Council members had “submerged” an electoral minority, the Negro voters of Dallas. In this respect the case is unlike East Carroll, where the original electoral scheme was invalidated solely on the ground of mal WISE v. LIPSCOMB 555 535 Marshall, J., dissenting apportionment and where the “racial dilution” challenge was raised only in objection to the proposed remedy. Multimember districts, which are disfavored as court-devised remedies because of their “tendency” or potential to create racial dilution, should a fortiori be disfavored when they are proposed to cure a proved use of a “multi-member . . . scheme ... to minimize or cancel out the voting strength of racial . . . elements of the voting population.” Fortson v. Dorsey, 379 U. S. 433,439 (1965).2 Based on respondents’ proof of a diluting effect on Negro voting strength in Dallas—and of the long history of de jure discrimination contributing to it—the District Court held the Dallas scheme to be unconstitutional. Although the Council did not challenge the finding that the at-large election of all its members was unconstitutional, the plan it submitted to the District Court replicated the offending feature of its original scheme by providing for the at-large election of three Council members. To put the burden on respondents to prove that the submission, insofar as it perpetuates at-large voting for Council members, is as unconstitutional as the original plan seems contrary to logic and common sense. I cannot agree that either the Constitution or the remedial principles of equity require such a result. For both of these reasons, I believe that the Court of Appeals correctly held that the use of at-large voting for City Council members in the city of Dallas should not have been approved as part of the remedy in this case by the District Court. I therefore dissent. 2 In White v. Regester, 412 U. S. 755, 765-770 (1973), this Court affirmed a District Court order directing that an unconstitutional multimember district be reapportioned into single-member districts designated by the court. The District Court had found the multimember district to be unconstitutional because of its dilutive effect on Negro voting strength, and had ordered implementation of its remedy without awaiting a legislative response to its finding of unconstitutionality. See Graves v. Barnes, 343 F. Supp. 704 (WD Tex. 1972) (three-judge court). 556 OCTOBER TERM, 1977 Syllabus 437 U. S. EASTEX, INC. v. NATIONAL LABOR RELATIONS BOARD CERTIORARI TO THE UNITED STATES COURT OF APPEALS FOR THE FIFTH CIRCUIT No. 77-453. Argued April 25, 1978—Decided June 22, 1978 Employees of petitioner corporation sought to distribute a four-part union newsletter in nonworking areas of petitioner’s plant during nonworking time. The first and fourth sections urged employees to support the union and extolled union solidarity. The second section encouraged employees to write their legislators to oppose incorporation of the state “right-to-work” statute into a revised state constitution. The third section criticized a Presidential veto of an increase in the federal minimum wage and urged employees to register to vote to “defeat our enemies and elect our friends.” After representatives of petitioner refused to permit the requested distribution, the union filed an unfair labor practice charge with the National Labor Relations Board (NLRB), alleging that petitioner’s refusal interfered with the employees’ exercise of their rights under § 7 of the National Labor Relations Act (Act), which provides that “[e]mployees shall have the right ... to engage in . . . concerted activities for the purpose of collective bargaining or other mutual aid or protection . . . ,” and thus violated §8 (a)(1). Following a hearing, at which petitioner contended that the second and third sections of the letter were not protected by § 7 because they did not relate to petitioner’s association with the union, the NLRB ordered petitioner to cease and desist from the violation, having determined that both those sections of the newsletter came within the ambit of § 7’s protection. The second section of the newsletter was held to be protected because union security is “central to the union concept of strength through solidarity” and “a mandatory subject of bargaining in other than right-to-work states,” and the fact that Texas already has a “right-to-work” statute was held not to diminish employees’ interest in the matter. The third section was held to be protected even though petitioner’s employees were paid more than the vetoed minimum wage, on the ground that the “minimum wage inevitably influences wage levels derived from collective bargaining, even those far above the minimum,” and that the petitioner’s employees’ concern “for the plight of other employees might gain support for them at some future time when they might have a dispute with their employer.” The Court of Appeals en EASTEX, INC. v. NLRB 557 556 Syllabus forced the NLRB’s order, rejecting petitioner’s contention that § 7’s “mutual aid or protection” clause protects only concerted activity by employees that is directed at conditions that their employer has the authority or power to change or control, and that the second and third sections of the newsletter did not constitute such activity. The court concluded that “whatever is reasonably related to the employees’ jobs or to their status or condition as employees in the plant may be the subject of such handouts as we treat of here, distributed on the plant premises in such a manner as not to interfere with the work . . . ,” and that the material in the newsletter met that test. Held: 1. Distribution of the challenged second and third sections of the newsletter is protected under the “mutual aid or protection” clause of § 7. Pp. 563-570. (a) The Act’s definition of “employee” in § 2 (3) was intended to protect employees when they engage in otherwise proper concerted activities in support of employees of employers other than their own, and it has long been held that “mutual aid or, protection” encompasses such activity. Pp. 564-565. (b) Employees do not lose their protection under the “mutual aid or protection” clause when they seek to improve terms and conditions of employment or otherwise improve their lot as employees through channels outside the immediate employee-employer relationship, and the NLRB did not err in holding that distribution of the challenged parts of the newsletter was for the purpose of “mutual aid or protection.” Pp. 565-570. 2. The NLRB did not err in holding that petitioner’s employees may distribute the newsletter in nonworking areas of petitioner’s property during nonworking time. The fact that the distribution is to take place on petitioner’s property does not give rise to a countervailing interest that petitioner can assert outweighing the exercise of § 7 rights by its employees in that location. Under the circumstances of this case, the NLRB was not required to apply a rule different from the one it applied in Republic Aviation Corp. v. NLRB, 324 U. S. 793, to the effect that an employer may not prohibit his employees from distributing union literature (in that case organizational material) in nonworking areas of industrial property during nonworking time, absent a showing by the employer that a ban is necessary to maintain plant discipline or production. Here, as in Republic Aviation, petitioner’s employees were “already rightfully on the employer’s property,” so that in the context of this case it is the employer’s management interests rather than its property interests that primarily are implicated. Petitioner, however, made no attempt to show that its management interests would be prejudiced 558 OCTOBER TERM, 1977 Opinion of the Court 437U.S. by distribution of the sections to which it objected, and any incremental intrusion on its property rights from their distribution together with the other sections would be minimal. In addition, viewed in context, the distribution was closely tied to vital concerns of the Act. Pp. 570-576. 550 F. 2d 198, affirmed. Powell, J., delivered the opinion of the Court, in which Brennan, Stewart, White, Marshall, Blackmun, and Stevens, JJ., joined. White, J., filed a concurring opinion, post, p. 578. Rehnquist, J., filed a dissenting opinion, in which Burger, C. J., joined, post, p. 579. John B. Abercrombie argued the cause for petitioner. With him on the brief was Tom Martin Davis. Richard A. Allen argued the cause for respondent. With him on the brief were Solicitor General McCree, John S. Irving, Carl L. Taylor, Norton J. Come, Linda Sher, and David S. Fishback* Mr. Justice Powell delivered the opinion of the Court. Employees of petitioner sought to distribute a union newsletter in nonworking areas of petitioner’s property during nonworking time urging employees to support the union and discussing a proposal to incorporate the state “right-to-work” statute into the state constitution and a Presidential veto of an increase in the federal minimum wage. The newsletter also called on employees to take action to protect their interests as employees with respect to these two issues. The question presented is whether petitioner’s refusal to allow the distribution violated §8 (a)(1) of the National Labor Relations Act, as amended, 61 Stat. 140, 29 U. S. C. § 158 (a)(1), by interfering with, restraining, or coercing employees’ exercise of their right under § 7 of the Act, 29 U. S. C. § 157, to engage in “concerted activities for the purpose of . . . mutual aid or protection.” * William L. Keller and Stephen A. Bokat filed a brief for the Chamber of Commerce of the United States as amicus curiae urging reversal. EASTEX, INC. v. NLRB Opinion of the Court 559 556 I Petitioner is a company that manufactures paper products in Silsbee, Tex. Since 1954, petitioner’s production employees have been represented by Local 801 of the United Paperworkers International Union. It appears that many, although not all, of petitioner’s approximately 800 production employees are members of Local 801. Since Texas is a “right-to-work State by statute,1 Local 801 is barred from obtaining an agreement with petitioner requiring all production employees to become union members. In March 1974, officers of Local 801, seeking to strengthen employee support for the union and perhaps recruit new members in anticipation of upcoming contract negotiations with petitioner, decided to distribute a union newsletter to petitioner’s production employees.2 The newsletter was divided into four sections. The first and fourth sections urged employees to support and participate in the union and, more generally, extolled the benefits of union solidarity. The second section encouraged employees to write their legislators to oppose incorporation of the state “right-to-work” statute into a revised state constitution then under consideration, warning that incorporation would “weakefn] Unions and improv[e] the edge business has at the bargaining table.” The third section noted that the President recently had vetoed a bill to increase the federal minimum wage from $1.60 to $2.00 per hour, compared this action to the increase of prices and profits in the oil industry under administration policies, and admonished: “As working men and women we must defeat our enemies and iTex. Rev. Civ. Stat. Ann., Art. 5154g, §1; Art. 5207a, §2 (Vernon 1971). 2 The president of Local 801 testified: “We were going into negotiations, and ... we was [sic] trying to reorganize our group into a stronger group. We were trying to get members, people that were working there who were non-members, and try to motivate or strengthen the conviction of our members, and it was to organize a little.” App. 11. 560 OCTOBER TERM, 1977 Opinion of the Court 437 U. S. elect our friends. If you haven’t registered to vote, please do so today.” 3 On March 26, 1974, Hugh Terry, an employee of petitioner and vice president of Local 801, asked Herbert George, petitioner’s assistant personnel director, for permission to distribute the newsletter to employees in the “clock alley” that leads to petitioner’s time clocks.4 George doubted whether management would allow employees to “hand out propaganda like that,” but agreed to check with his superiors. Leonard Menius, petitioner’s personnel director, confirmed that petitioner would not allow employees to distribute the newsletter in clock alley. A few days later George communicated this decision to Terry, but gave no reasons for it. On April 22, 1974, Boyd Young, president of Local 801,5 together with Terry and another employee, asked George whether employees could distribute the newsletter in any nonworking areas of petitioner’s property other than clock alley.6 After conferring again with Menius, George reported 3 The newsletter is reprinted in full as an appendix to this opinion. 4 The Administrative Law Judge described “clock alley” as “a passageway 6 or 7 feet wide, flanked on either side by administrative offices. In addition to time clocks, the area contains an employee bulletin board and benches and chairs for those waiting to transact business in the offices. Clock alley is physically discrete from the production areas of the plant.” 215 N. L. R. B. 271, 273 n. 7 (1974). 5 Young, a longtime employee of petitioner, was on leave to serve as president of Local 801. 6 Young testified that he had asked “permission for employees of the Company to be allowed to distribute this on non-working hours, on non-production areas, and specifically outside the clock alley; and if that area posed a problem, we would be willing to move to any area convenient to the Company, out on the end of the walk or guardhouse or parking lot, that we would only hand it out to employees leaving the plant, and where it wouldn’t cause a litter problem in the plant.” App. 8-9. The Administrative Law Judge credited Young’s testimony that the request was only for employees to distribute the newsletter. 215 N. L. R. B., at 273 n. 9. EASTEX, INC. v. NLRB 561 556 Opinion of the Court that employees would not be allowed to do so and that petitioner thought the union had other ways to communicate with employees. Local 801 then filed an unfair practice charge with the National Labor Relations Board (Board), alleging that petitioner’s refusal to allow employees to distribute the newsletter in nonworking areas of petitioner’s property during non working time interfered with, restrained, and coerced employees’ exercise of their § 7 rights in violation of § 8 (a) (I).7 At a hearing on the charge, Menius testified that he had no objection to the first and fourth sections of the newsletter. He had denied permission to distribute the newsletter because he “didn’t see any way in which [the second and third sections were] related to our association with the Union.” App. 19. The Administrative Law Judge held that although not all of the newsletter had immediate bearing on the relationship between petitioner and Local 801, distribution of all its contents was protected under § 7 as concerted activity for the “mutual aid or protection” of employees. Because petitioner had presented no evidence of “special circumstances” to justify a ban on the distribution of protected matter by employees in nonworking areas during nonworking time, the Administrative Law Judge held that petitioner had violated § 8 (a)(1) and ordered petitioner to cease and desist from the violation.8 The Board 7 Section 8 (a)(1) makes it an unfair labor practice "to interfere with, restrain, or coerce employees in the exercise of the rights guaranteed in” § 7 of the Act. 8 Because no evidence of "special circumstances” had been presented, the Administrative Law Judge did not consider whether alternative channels of communication were available to Local 801. 215 N. L. R. B.„ at 275 n. 13. In the alternative, the judge held that even if distribution of the second and third sections of the newsletter was not protected by § 7, distribution of the newsletter as a whole was protected. Id., at 274, relying on Samsonite Corp., 206 N. L. R. B. 343 (1973). The Administrative Law Judge also held that petitioner maintained an overbroad no-solicitation rule. 215 N. L. R. B., at 272. Petitioner did 562 OCTOBER TERM, 1977 Opinion of the Court 437U.S. affirmed the Administrative Law Judge’s rulings, findings, and conclusions, and adopted his recommended order. 215 N. L. R. B. 271 (1974). The Court of Appeals enforced the order. 550 F. 2d 198 (CA5 1977). It rejected petitioner’s argument that the “mutual aid or protection” clause of § 7 protects only concerted activity by employees that is directed at conditions that their employer has the authority or power to change or control. Without expressing an opinion as to the full range of § 7 rights “when exercised off the employer’s property,” 550 F. 2d, at 202, the court purported to balance those rights against the employer’s property rights and concluded that “whatever is reasonably related to the employees’ jobs or to their status or condition as employees in the plant may be the subject of such handouts as we treat of here, distributed on the plant premises in such a manner as not to interfere with the work . . . .” Id., at 203 (emphasis in original). The court further held that all of the material in the newsletter here met this test. Id., at 204—205.9 Because of apparent differences among the Courts of Appeals as to the scope of rights protected by the “mutual aid or protection” clause of § 7, see n. 17, infra, we granted certiorari. 434 U. S. 1045 (1978). We affirm. not rely on this rule in refusing to allow distribution of the newsletter, see id., at 272 n. 4, and its validity was not an issue in the Court of Appeals, see 550 F. 2d 198, 201 n. 3 (CA5 1977). That rule is not before us. See Brief for Petitioner 5 n. 2. 9 The court went on to disapprove the alternative ground for the Board’s decision, see n. 8, supra, stating that “the presence of some § 7 protected material will not rescue that which is significantly not protected.” 550 F. 2d, at 205. We do not find it necessary to express an opinion as to the correctness of this statement. In an opinion denying rehearing and rehearing en banc, the court reaffirmed that it had balanced the employer’s and employees’ rights, and it deleted two references in its first opinion to the First Amendment. 556 F. 2d 1280 (CA5 1977). EASTEX, INC. v. NLRB 563 556 Opinion of the Court II Two distinct questions are presented. The first is whether, apart from the location of the activity, distribution of the newsletter is the kind of concerted activity that is protected from employer interference by §§ 7 and 8(a)(1) of the National Labor Relations Act. If it is, then the second question is whether the fact that the activity takes place on petitioner’s property gives rise to a countervailing interest that outweighs the exercise of § 7 rights in that location. See Hudgens v. NLRB, 424 U. S. 507, 521-523 (1976); Central Hardware Co. v. NLRB, 407 U. S. 539, 542-545 (1972); NLRB v. Babcock & Wilcox Co., 351 U. S. 105, 112 (1956); Republic Aviation Corp. v. NLRB, 324 U. S. 793, 797-798 (1945). We address these questions in turn. A Section 7 provides that “[e]mployees shall have the right ... to engage in . . . concerted activities for the purpose of collective bargaining or other mutual aid or protection . . . .” 10 Petitioner contends that the activity here is not within the “mutual aid or protection” language because it does not relate to a “specific dispute” between employees and their own employer “over an issue which the employer has the right or power to affect.” Brief for Petitioner 13. In support of its position, petitioner asserts that the term “employees” in § 7 refers only to employees of a particular employer, so that only activity by employees on behalf of themselves or other em 10 Section 7, as amended, as set forth in 29 U. S. C. § 157, states in full: Employees shall have the right to self-organize, to form, join, or assist labor organizations, to bargain collectively through representatives of their own choosing, and to engage in other concerted activities for the purpose of collective bargaining or other mutual aid or protection, and shall also have the right to refrain from any or all of such activities except to the extent that such right may be affected by an agreement requiring member-ship in a labor organization as a condition of employment as authorized in section 158 (a) (3) of this title [29].” 564 OCTOBER TERM, 1977 Opinion of the Court 437U.S. ployees of the same employer is protected. Id., at 18, 24. Petitioner also argues that the term “collective bargaining” in § 7 “indicates a direct bargaining relationship whereas ‘other mutual aid or protection’ must refer to activities of a similar nature . ..Id., at 24. Thus, in petitioner’s view, under § 7 “the employee is only protected for activity within the scope of the employment relationship.” Id., at 13. Petitioner rejects the idea that § 7 might protect any activity that could be characterized as “political,” and suggests that the discharge of an employee who engages in any such activity would not violate the Act.11 We believe that petitioner misconceives the reach of the “mutual aid or protection” clause. The “employees” who may engage in concerted activities for “mutual aid or protection” are defined by § 2 (3) of the Act, 29 U. S. C. § 152 (3), to “include any employee, and shall not be limited to the employees of a particular employer, unless this subchapter explicitly states otherwise .. ..” This definition was intended to protect employees when they engage in otherwise proper concerted activities in support of employees of employers other than their own.12 In recognition of this intent, the Board and the courts long have held that the “mutual aid or protection” clause encompasses such activity.13 Petitioner’s 11 See Tr. of Oral Arg. 17: “QUESTION: [Suppose the] Union is banding together and they all want to oppose right-to-work laws, and they pass out literature out on the public street; and the employer says, T just don’t like you fellows getting into this kind of business, I’m going to fire you.’ “Now, is that an unfair labor practice? “MR. ABERCROMBIE: Your Honor, we would submit that it was not, that political activity is not protected under Section 7.” 12 See Phelps Dodge Corp. v. NLRB, 313 U. S. 177, 191-192 (1941); S. Rep. No. 573, 74th Cong., 1st Sess., 6 (1935); H. R. Rep. No. 1147, 74th Cong., 1st Sess., 9-10 (1935). 13 E. g., Fort Wayne Corrugated Paper Co. v. NLRB, 111 F. 2d 869, 874 (CA7 1940), enf’g Cayuga Linen & Cotten Mills, Inc., 11 N. L. R. B. 1, EASTEX, INC. v. NLRB 565 556 Opinion of the Court argument on this point ignores the language of the Act and its settled construction. We also find no warrant for petitioner’s view that employees lose their protection under the “mutual aid or protection” clause when they seek to improve terms and conditions of employment or otherwise improve their lot as employees through channels outside the immediate employee-employer relationship. The 74th Congress knew well enough that labor’s cause often is advanced on fronts other than collective bargaining and grievance settlement within the immediate employment context. It recognized this fact by choosing, as the language of § 7 makes clear, to protect concerted activities for the somewhat broader purpose of “mutual aid or protection” as well as for the narrower purposes of “self-organization” and “collective bargaining.”14 Thus, it has been held that the “mutual aid or 4-5 (1939) (right to assist in organizing another employer’s employees); NLRB v. J. G. Boswell Co., 136 F. 2d 585, 595 (CA9 1943), enf’g 35 N. L. R. B. 968 (1941) (right to express sympathy for striking employees of another employer); Redwing Carriers, Inc., 137 N. L. R. B. 1545, 1546-1547 (1962), enf’d sub nom. Teamsters n. NLRB, 117 U. S. App. D. C. 84, 325 F. 2d 1011 (1963), cert, denied, 377 U. S. 905 (1964) (right to honor picket line of another employer’s employees); NLRB v. Alamo Express Co.,430 F. 2d 1032, 1036 (CA5 1970), cert, denied, 400 U. S. 1021 (1971), enf’g 170 N. L. R. B. 315 (1968) (accord); Washington State Service Employees, 188 N. L. R. B. 957, 959 (1971) (right to demonstrate in support of another employer’s employees); Yellow Cab, Inc., 210 N. L. R. B. 568, 569 (1974) (right to distribute literature in support of another employer’s employees). We express no opinion, however, as to the correctness of the particular balance struck between employees’ exercise of § 7 rights and employers’ legitimate interests in any of the above-cited cases. 14 Congress modeled the language of § 7 after that found in § 2 of the Norris-LaGuardia Act, 47 Stat. 70, 29 U. S. C. § 102, which declares that it is the public policy of the United States that workers “shall be free from the interference, restraint, or coercion of employers of labor, or their agents, in the designation of . . . representatives or in self-organization or in other concerted activities for the purpose of collective bargaining or other mutual aid or protection . . . .” See S. Rep. No. 573, 74th Cong., 1st 566 OCTOBER TERM, 1977 Opinion of the Court 437U.S. protection” clause protects employees from retaliation by their employers when they seek to improve working conditions through resort to administrative and judicial forums,15 and that employees’ appeals to legislators to protect their interests as employees are within the scope of this clause.16 To hold that activity of this nature is entirely unprotected—irrespective of location or the means employed—would leave employees Sess., 9 (1935); H. R. Rep. No. 1147, 74th Cong., 1st Sess., 15 (1935). This section of the Norris-LaGuardia Act expresses Congress’ recognition of the “right of wage earners to organize and to act jointly in questions affecting wages, conditions of labor, and the welfare of labor generally ....” S. Rep. No. 163, 72d Cong., 1st Sess., 9 (1932) (emphasis supplied). Similar language is found in §7 (a)(1) of the National Industrial Recovery Act of 1933, 48 Stat. 198; § 1 of the National Labor Relations Act, 49 Stat. 449, 29 U. S. C. § 151 (declaration of policy); and § 2 (a) of the Labor-Management Reporting and Disclosure Act of 1959, 73 Stat. 519, 29 U. S. C. § 401 (a) (findings, purposes, and policy). 15 E. g., Walls Mfg. Co., 137 N. L. R. B. 1317 (1962), enf’d, 116 U. S. App. D. C. 140, 321 F. 2d 753, cert, denied, 375 U. S. 923 (1963); Socony Mobil Oil Co., 153 N. L. R. B. 1244 (1965), enf’d, 357 F. 2d 662 (CA2 1966); Altex Ready Mixed Concrete Corp. v. NLRB, 542 F. 2d 295, 297 (CA5 1976), enf’g 223 N. L. R. B. 696; Wray Electric Contracting, Inc., 210 N. L. R. B. 757 (1974); Alleluia Cushion Co., 221 N. L. R. B. 999 (1975); King Soopers, Inc., 222 N. L. R. B. 1011 (1976); Triangle Tool & Engineering, Inc., 226 N. L. R. B. 1354 (1976). We do not address here the question of what may constitute “concerted” activities in this context. Cf. NLRB v. Weingarten, Inc., 420 U. S. 251, 260-261 (1975). 16 Bethlehem Shipbuilding Corp. v. NLRB, 114 F. 2d 930, 937 (CAI 1940), dismissed on motion of petitioner, 312 U. S. 710 (1941), enf’g 11 N. L. R. B. 105 (1939); NLRB v. Peter Cailler Kohler Swiss Chocolates Co., 130 F. 2d 503, 506 (CA2 1942) (dicta), enf’g 33 N. L. R. B. 1170 (1941); Kaiser Engineers v. NLRB, 538 F. 2d 1379, 1384-1385 (CA9 1976), enf’g 213 N. L. R. B. 752 (1974); cf. Machinists n. Street, 367 U. S. 740, 800-801, 812-816 (1961) (Frankfurter, J., dissenting). Other laws, however, may place limits on concerted activity in the legislative and political spheres. See United States v. CIO, 335 U. S. 106 (1948); United States v. Auto Workers, 352 U. S. 567 (1957); Street, supra; Railway Clerks v. Allen, 373 U. S. 113 (1963); Pipefitters v. United States, 407 U. S. 385 (1972); Abood v. Detroit Bd. of Education., 431 U. S. 209 (1977). EASTEX, INC. v. NLRB 567 556 Opinion of the Court open to retaliation for much legitimate activity that could improve their lot as employees. As this could “frustrate the policy of the Act to protect the right of workers to act together to better their working conditions,” NLRB v. Washington Aluminum Co., 370 U. S. 9, 14 (1962), we do not think that Congress could have intended the protection of § 7 to be as narrow as petitioner insists.17 It is true, of course, that some concerted activity bears a less immediate relationship to employees’ interests as employees than other such activity. We may assume that at some point 17 Petitioner relies upon several cases said to construe § 7 more narrowly than do we. NLRB v. Leslie Metal Arts Co., 509 F. 2d 811 (CA6 1975), and Shelly & Anderson Furniture Mjg. Co. n. NLRB, 497 F. 2d 1200 (CA9 1974), both quote the same treatise for the proposition that to be protected under § 7, concerted activity must seek “a specific remedy” for a “work-related complaint or grievance.” 509 F. 2d, at 813, and 497 F. 2d, at 1202-1203, quoting 18B T. Kheel, Labor Law § 10.02 [3], pp. 10-21 (1973). It was unnecessary in those cases to decide whether the protection of § 7 went beyond the treatise’s formulation, for the activity in both cases was held to be protected. Moreover, in stating its “rule,” the treatise relied upon takes no note of the cases cited in nn. 13, 15, and 16, supra. Cf. R. Gorman, Labor Law 296-302 (1976). The Courts of Appeals for the Sixth and Ninth Circuits themselves have taken a broader view of the “mutual aid or protection” clause than the reference to the treatise in the above-cited cases would seem to suggest. See, e. g., Kellogg Co. v. NLRB, 457 F. 2d 519, 522-523 (CA6 1972), and cases there cited; Kaiser Engineers v. NLRB, supra, at 1384-1385. Similarly, although the Court of Appeals for the Fourth Circuit stated in NLRB v. Bretz Fuel Co., 210 F. 2d 392 (1954), that “concerted activity is protected only where such activity is intimately connected with the employees’ immediate employment,” id., at 396, the holding in that case turned more on the fact that the activity there consisted of a wildcat strike in violation of a collective-bargaining agreement than on a narrow view of the “mutual aid or protection” clause. See id., at 397-398. This leaves only G&W Electric Specialty Co. v. NLRB, 360 F. 2d 873 (CA7 1966), which refused to enforce a Board order because the concerted activity there—circulation of a petition concerning management of an employee-run credit union—“involved no request for any action upon the part of the Company and did not concern a matter over which the Com 568 OCTOBER TERM, 1977 Opinion of the Court 437 U. S. the relationship becomes so attenuated that an activity cannot fairly be deemed to come within the “mutual aid or protection” clause. It is neither necessary nor appropriate, however, for us to attempt to delineate precisely the boundaries of the “mutual aid or protection” clause. That task is for the Board to perform in the first instance as it considers the wide variety of cases that come before it.18 Republic Aviation Corp. v. NLRB, 324 U. S., at 798; Phelps Dodge Corp. v. NLRB, 313 U. S. 177, 194 (1941). To decide this case, it is enough to determine whether the Board erred in holding that distribution of the second and third sections of the newsletter is for the purpose of “mutual aid or protection.” pany had control.” Id., at 876. G&W Electric cites no authority for its narrowing of § 7, and it ignores a substantial weight of authority to the contrary, including the Seventh Circuit’s own prior holding in Fort Wayne Corrugated Peeper Co. v. NLRB, 111 F. 2d, at 874. See n. 13, supra. We therefore do not view any of these cases as persuasive authority for petitioner’s position. 18 See Ford Motor Co., 221 N. L. R. B. 663, 666 (1975), enf’d, 546 F. 2d 418 (CA3 1976) (holding distribution on employer’s premises of a “purely political tract” unprotected even though “the election of any political candidate may have an ultimate effect on employment conditions”); cf. Ford Motor Co. (Rouge Complex), 233 N. L. R. B. 698, 705 (1977) (decision of Administrative Law Judge) (concession of General Counsel that distributions on employer’s premises of literature urging participation in Revolutionary Communist Party celebration, and of Party’s newspaper, were unprotected). The Board has not yet made clear whether it considers distributions like those in the above-cited cases to be unprotected altogether, or only on the employer’s premises. In addition, even when concerted activity comes within the scope of the “mutual aid or protection” clause, the forms such activity permissibly may take may well depend on the object of the activity. “The argument that the employer’s lack of interest or control affords a legitimate basis for holding that a subject does not come within 'mutual aid or protection’ is unconvincing. The argument that economic pressure should be unprotected in such cases is more convincing.” Getman, The Protection of Economic Pressure by Section 7 of the National Labor Relations Act, 115 U. Pa. L. Rev. 1195,1221 (1967). EASTEX, INC. v. NLRB 569 556 Opinion of the Court The Board determined that distribution of the second section, urging employees to write their legislators to oppose incorporation of the state “right-to-work” statute into a revised state constitution, was protected because union security is central to the union concept of strength through solidarity” and a mandatory subject of bargaining in other than right-to-work states.” 215 N. L. R. B., at 274. The newsletter warned that incorporation could affect employees adversely “by weakening Unions and improving the edge business has at the bargaining table.” The fact that Texas already has a “right-to-work” statute does not render employees’ interest in this matter any less strong, for, as the Court of Appeals noted, it is “one thing to face a statutory scheme which is open to legislative modification or repeal” and “quite another thing to face the prospect that such a scheme will be frozen in a concrete constitutional mandate.” 550 F. 2d, at 205. We cannot say that the Board erred in holding that this section of the newsletter bears such a relation to employees’ interests as to come within the guarantee of the “mutual aid or protection” clause. See cases cited in n. 16, supra. The Board held that distribution of the third section, criticizing a Presidential veto of an increase in the federal minimum wage and urging employees to register to vote to “defeat our enemies and elect our friends,” was protected despite the fact that petitioner’s employees were paid more than the vetoed minimum wage. It reasoned that the “mini-mum wage inevitably influences wage levels derived from collective bargaining, even those far above the minimum,” and that “concern by [petitioner’s] employees for the plight of other employees might gain support for them at some future time when they might have a dispute with their employer.” 215 N. L. R. B., at 274 (internal quotation marks omitted). We think that the Board acted within the range of its discretion in so holding. Few topics are of such immediate concern to employees as the level of their wages. The Board was 570 OCTOBER TERM, 1977 Opinion of the Court 437 U. S. entitled to note the widely recognized impact that a rise in the minimum wage may have on the level of negotiated wages generally,19 a phenomenon that would not have been lost on petitioner’s employees. The union’s call, in the circumstances of this case, for these employees to back persons who support an increase in the minimum wage, and to oppose those who oppose it, fairly is characterized as concerted activity for the “mutual aid or protection” of petitioner’s employees and of employees generally. In sum, we hold that distribution of both the second and the third sections of the newsletter is protected under the “mutual aid or protection” clause of § 7.20 B The question that remains is whether the Board erred in holding that petitioner’s employees may distribute the newsletter in nonworking areas of petitioner’s property during nonworking time. Consideration of this issue must begin with the Court’s decisions in Republic Aviation Corp. v. NLRB, supra, and NLRB v. Babcock & Wilcox Co., 351 U. S. 105 (1956). In Republic Aviation the Court upheld the Board’s ruling that an employer may not prohibit its employees from 19 See N. Chamberlain, Labor 435-437 (1958); L. Reynolds, Labor Economics and Labor Relations 272 (5th ed. 1970). 20 Petitioner argues that the “right to work” and minimum wage issues are “political,” and that advancing a union’s political views is not protected by § 7. As almost every issue can be viewed by some as political, the clear purpose of the “mutual aid or protection” clause would be frustrated if the mere characterization of conduct or speech removed it from the protection of the Act. See cases cited in n. 16, supra. Moreover, what may be viewed as political in one context can be viewed quite differently in another. There may well be types of conduct or speech that are so purely political or so remotely connected to the concerns of employees as employees as to be beyond the protection of the clause. But this is a determination that should be left for case-by-case consideration. Cf. cases cited in n. 18, supra. EASTEX, INC. v. NLRB 571 556 Opinion of the Court distributing union organizational literature in nonworking areas of its industrial property during non working time, absent a showing by the employer that a ban is necessary to maintain plant discipline or production. This ruling obtained even though the employees had not shown that distribution off the employer’s property would be ineffective. 324 U. S., at 798-799, 801. In the Court’s view, the Board had reached an acceptable “adjustment between the undisputed right of self-organization assured to employees under the Wagner Act and the equally undisputed right of employers to maintain discipline in their establishments.” Id., at 797-798.21 In Babcock de Wilcox, on the other hand, nonemployees sought to enter an employer’s property to distribute union organizational literature. The Board applied the rule of Republic Aviation in this situation, but the Court held that there is a distinction “of substance” between “rules of law applicable to employees and those applicable to nonemployees.” 351 U. S., at 113. The difference was that the nonemployees in Babcock <& Wilcox sought to trespass on the employer’s property, whereas the employees in Republic Aviation did not. Striking a balance between § 7 organizational rights and an employer’s right to keep strangers from entering on its property, the Court held that the employer in Babcock de Wilcox was entitled to prevent “nonemployee distribution of union literature [on its property] if reasonable efforts by the union through other available channels of communication will enable it to reach the employees with its message . . . Id., at 112. The Court recently has emphasized the distinction between the two cases: “A wholly different balance was 21 In Republic Aviation the Court also upheld Board rulings that employees may solicit other employees to join a union on the employer’s property during nonworking time, and may wear union insignia on the employer’s property. The Board since has distinguished between distributions of literature and oral solicitation, holding that the latter but not the former may take place in working areas during nonworking time. Stoddard-Quirk Mfg. Co., 138 N. L. R. B. 615 (1962). 572 OCTOBER TERM, 1977 Opinion of the Court 437U.S. struck when the organizational activity was carried on by employees already rightfully on the employer’s property, since the employer’s management interests rather than his property interests were there involved.” Hudgens v. NLRB, 424 U. 8., at 521-522, n. 10; see also Central Hardware Co. v. NLRB, 407 U. 8., at 543-545. It is apparent that the instant case resembles Republic Aviation rather closely. Here, as there, employees sought to distribute literature in non working areas of their employer’s industrial property during nonworking time. Here, as there, the employer has not attempted to show that distribution would interfere with plant discipline or production. And here, as there, distribution of the newsletter clearly would be protected by § 7 against employer discipline if it took place off the employer’s property. The only possible ground of distinction is that part of the newsletter in this case does not address purely organizational matters, but rather concerns other activity protected by § 7. The question, then, is whether this difference required the Board to apply a different rule here than it applied in Republic Aviation. Petitioner contends that the Board must distinguish among distributions of protected matter by employees on an employer’s property on the basis of the content of each distribution. Echoing its earlier argument, petitioner urges that the Republic Aviation rule should not be applied if a distribution “does not involve a request for any action on the part of the employer, or does not concern a matter over which the employer has any degree of control . . . Brief for Petitioner 28. In petitioner’s view, distribution of any other matter protected by §7 would be an “unnecessary intrusio[n] on the employer’s property rights,” id., at 29, in the absence of a showing by employees that no alternative channels of communication with fellow employees are available. We hold that the Board was not required to adopt this view in the case at hand. In the first place, petitioner’s reliance on EASTEX, INC. v. NLRB 573 556 Opinion of the Court its property right is largely misplaced. Here, as in Republic Aviation, petitioner’s employees are “already rightfully on the employer’s property,” so that in the context of this case it is the “employer’s management interests rather than [its] property interests” that primarily are implicated. Hudgens, supra, at 521-522, n. 10. As already noted, petitioner made no attempt to show that its management interests would be prejudiced in any way by the exercise of § 7 rights proposed by its employees here. Even if the mere distribution by employees of material protected by § 7 can be said to intrude on petitioner’s property rights in any meaningful sense, the degree of intrusion does not vary with the content of the material. Petitioner’s only cognizable property right in this respect is in preventing employees from bringing literature onto its property and distributing it there—not in choosing which distributions protected by § 7 it wishes to suppress.22 On the other side of the balance, it may be argued that the employees’ interest in distributing literature that deals with matters affecting them as employees, but not with selforganization or collective bargaining, is so removed from the central concerns of the Act as to justify application of a different rule than in Republic Aviation. Although such an argument may have force in some circumstances, see Hudgens, supra, at 522, the Board to date generally has chosen not to engage in such refinement of its rules regarding the distribution 22 In addition, we doubt whether the test proposed by petitioner for the protection of its property rights can be squared with Republic Aviation itself, for the organizational literature in that case did not “involve a request for any action on the part of the employer, or . . . concern a matter over which the employer [had] any degree of control.” To be sure, if the material distributed on the premises of the employer were inflammatory to the point of threatening disorder or other interruption of the normal functioning of the business, the exception noted in Republic Aviation with respect to interference with discipline or production would be fully applicable. See Procter & Gamble Mjg. Co., 160 N. L. R. B. 334, 395 (1966). 574 OCTOBER TERM, 1977 Opinion of the Court 437U.S. of literature by employees during nonworking time in nonworking areas of their employers’ property. We are not prepared to say in this case that the Board erred in the view it took. It is apparent that the complexity of the Board’s rules and the difficulty of the Board’s task might be compounded greatly if it were required to distinguish not only between literature that is within and without the protection of § 7, but also among subcategories of literature within that protection. In addition, whatever the strength of the employees’ § 7 interest in distributing particular literature, the Board is entitled to view the intrusion by employees on the property rights of their employer as quite limited in this context as long as the employer’s management interests are adequately protected. The Board also properly may take into account the fact that the plant is a particularly appropriate place for the distribution of § 7 material, because it "is the one place where [employees] clearly share common interests and where they traditionally seek to persuade fellow workers in matters affecting their union organizational life and other matters related to their status as employees.” Gale Products, 142 N. L. R. B. 1246, 1249 (1963). We need not go so far in this case, however, as to hold that the Republic Aviation rule properly is applied to every in-plant distribution of literature that falls within the protective ambit of § 7. This is a new area for the Board and the courts which has not yet received mature consideration.23 It may be that the 23 In addition to the instant case, the Board has extended the rule of Republic Aviation to a limited extent to encompass nonorganizational literature complaining about an incumbent union’s leadership or bargaining position. Samsonite Corp., 206 N. L. R. B. 343 (1973); McDonnell Douglas Corp., 210 N. L. R. B. 280 (1974); General Motors Corp., 212 N. L. R. B. 133 (1974); The Singer Co., 220 N. L. R. B. 1179 (1975); Ford Motor Co., 221 N. L. R. B. 663 (1975), enf’d, 546 F. 2d 418 (CA3 1976). In one case it applied the rule to literature exhorting employees EASTEX, INC. v. NLRB 575 556 Opinion of the Court “nature of the problem, as revealed by unfolding variant situations,” requires “an evolutionary process for its rational response, not a quick, definitive formula as a comprehensive answer.” Electrical Workers v. NLRB, 366 U. S. 667, 674 (1961). For this reason, we confine our holding to the facts of this case. Petitioner concedes that its employees were entitled to distribute a substantial portion of this newsletter on its property. In addition, as we have held above, the sections to which petitioner objected concern activity which petitioner, in the absence of a countervailing interest of its own, is not entitled to suppress. Yet petitioner made no attempt to show that its management interests would be prejudiced in any manner by distribution of these sections, and in our view any incremental intrusion on petitioner’s property rights from their distribution together with the other sections would be minimal. Moreover, it is undisputed that the union undertook the distribution in order to boost its support and improve its bargaining position in upcoming contract negotiations with petitioner. Thus, viewed in context, the distribution was closely tied to vital concerns of the Act.24 In these circum- “to support employees of other employers who were on strike and to oppose an alleged antilabor combination.” Yellow Cab, Inc., 210 N. L. R. B., at 569. On the other hand, it has not allowed distribution of “purely political” material on employers’ premises, even when the material might arguably be within the scope of § 7. See n. 18, supra. This Court already has approved the Board’s limited extension of the Republic Aviation rule to cover the distribution of literature by dissident employees advocating the displacement of a union. See NLRB v. Magnavox Co., 415 U. S. 322 (1974); id., at 327 (Stewart, J., concurring in part and dissenting in part). 24 As we have had occasion to state: “Unions have a legitimate and substantial interest in continuing organizational efforts after recognition. Whether the goal is merely to strengthen or preserve the union’s majority, or is to achieve 100% employee membership—a particularly substantial union concern where union security agreements are not permitted, as they are not here . . .—these organizing efforts are equally entitled to the 576 437 U.S. OCTOBER TERM, 1977 Appendix to opinion of the Court stances, we hold that the Board did not err in applying the Republic Aviation rule to the facts of this case. The judgment of the Court of Appeals therefore is Affirmed. APPENDIX TO OPINION OF THE COURT NEWS BULLETIN TO LOCAL 801 MEMBERS FROM BOYD YOUNG—PRESIDENT WE NEED YOU As a member, we need you to help build the Union through your support and understanding. Too often members become disinterested and look upon their Union as being something separate from themselves. Nothing could be further from the truth. This Union or any Union will only be as good as the members make it. The policies and practices of this Union are made by the membership—the active membership. If this Union has ever missed its target it may be because not enough members made their views known where the final decisions are made— The Union Meeting. It would be impossible to satisfy everyone with the decisions that are made but the active member has the opportunity to bring the majority around to his way of thinking. This is how a democratic organization works and it’s the best system around. Through participation you can make your voice felt not only in this Local but throughout the International Union. A PHONY LABEL—“right to work” Wages are determined at the bargaining table and the stronger the Union, the better the opportunity for improvements. The “right to work” law is simply an attempt to weaken the strength of Unions. The misleading title of protection of § 7 ... .” Letter Carriers n. Austin, 418 U. S. 264, 279 (1974). EASTEX, INC. v. NLRB 577 556 Appendix to opinion of the Court “right to work” cannot guarantee anyone a job. It simply weakens the negotiating power of Unions by outlawing provisions in contracts for Union shops, agency shops, and modified Union shops. These laws do not improve wages or working conditions but just protect free riders. Free riders are people who take all the benefits of Unions without paying dues. They ride on the dues that members pay to build an organization to protect their rights and improve their way of life. At this time there is a very well organized and financed attempt to place the “right to work” law in our new state constitution. This drive is supported and financed by big business, namely, the National Right-To-Work Committee and the National Chamber of Commerce. If their attempt is successful, it will more than pay for itself by weakening Unions and improving the edge business has at the bargaining table. States that have no “right-to-work” law consistently have higher wages and better working conditions. Texas is well known for its weak laws concerning the working class and the “right-to-work” law would only add insult to injury. If you fail to take action against the “right-to-work” law it may well show up in wages negotiated in the future. I urge every member to write their state congressman and senator in protest of the “right-to-work” law being incorporated into the state constitution. Write your state representative and state senator and let the delegate know how you feel. POLITICS AND INFLATION The Minimum Wage Bill, HR 7935, was vetoed by President Nixon. The President termed the bill as inflationary. The bill would raise the present $1.60 to $2.00 per hour for most covered workers. It seems almost unbelievable that the President could term $2.00 per hour as inflationary and at the same time remain silent about oil companies profits ranging from 56% to 280%. It also seems disturbing, that after the price of gasoline has increased to over 50 cents a gallon, that the fuel crisis is 578 OCTOBER TERM, 1977 White, J., concurring 437U.S. beginning to disappear. If the price of gasoline ever reaches 70 cents a gallon you probably couldn’t find a closed filling station or empty pump in the Northern Hemisphere. Congress is now pr[o]ceeding with a second minimum wage bill that hopefully the President will sign into law. At $1.60 per hour you could work 40 hours a week, 52 weeks a year and never earn enough money to support a family. As working men and women we must defeat our enemies and elect our friends. If you haven’t registered to vote, please do so today. FOOD FOR THOUGHT In Union there is strength, justice, and moderation; In disunion, nothing but an alternating humility and insolence. COMING TOGETHER WAS A BEGINNING STAYING TOGETHER IS PROGRESS WORKING TOGETHER MEANS SUCCESS THE PERSON WHO STANDS NEUTRAL, STANDS FOR NOTHING! Mr. Justice White, concurring. As I understand the record in this case, the only issue before the Administrative Law Judge and before the Board was whether the activity engaged in here by the employees was the kind of activity protected by § 7 of the National Labor Relations Act. The Administrative Law Judge held that the circulars were related to matters encompassed by § 7 and noted that there had been no attempt or evidence to show that even though the distributions were § 7 activity, there were nevertheless circumstances that permitted the employer to forbid the distributions on his property. The Board adopted the report of the Administrative Law Judge. I agree that the employees here were engaged in activity protected by § 7, at least in the sense that the employer could not discharge employees for propagandizing their fellow workers with materials concerning minimum wages and right-to- EASTEX, INC. v. NLRB 579 556 Rehnquist, J., dissenting work laws, so long as the distribution takes place off the employer’s property. I agree further that under current law and the facts and claims in this record, the distributions could take place on the employer’s property. Accordingly, the Board was entitled to have its order enforced and I join the judgment and opinion of the Court. In doing so, I should say that it is not easy to explain, why an employer need permit his property to be used for distributions about subjects unrelated to his relationship with his employees simply because it is convenient for the latter to use his property in this manner and simply because there is no interference with “management interests.” Ownership of property normally confers the right to control the use of that property. Here there was no finding by the Board that the literature sought to be distributed was connected with the bargaining relationship; and I doubt that federal law requires the employer always to permit his property to be used for solicitations and distributions having § 7 protection, even by and among employees in nonworking areas and during nonworking times. Such distributions might concern goals and ends about which his work force, considered as a whole, as well as the public, may be deeply divided, with which he may have no sympathy whatsoever, or in connection with which he would not care to have it inferred that he supports one side or the other. All of these, if substantiated by the record, would appear to be substantial factors to be weighed in the balance when determining whether the employer has violated the Labor Act’s strictures concerning his relationship with his employees. However this may be, on the record before us, I am content to affirm the judgment of the Court of Appeals. Mr. Justice Rehnquist, with whom The Chief Justice joins, dissenting. It is not necessary to determine the scope of the “mutual aid or protection” language of § 7 of the National Labor Rela- 580 437 U. S. OCTOBER TERM, 1977 Rehnquist, J., dissenting tions Act to conclude that Congress never intended to require the opening of private property to the sort of political advocacy involved in this case. Petitioner’s right as a property owner to prescribe the conditions under which strangers may enter its property is fully recognized under Texas law. “ 'A licensee who goes beyond the rights and privileges granted by the license becomes a trespasser.’ ” Burton Construction & Shipbuilding Co. v. Broussard, 154 Tex. 50, 58, 273 S. W. 2d 598, 603 (1954) (citation omitted). See also Brown v. Dellinger, 355 S. W. 2d 742 (Tex. Civ. App. 1962); 56 Tex. Jur. 2d, Trespass §4 (1964). Thus, the employees’ effort to distribute their leaflet in defiance of petitioner’s wishes would clearly be a trespass infringing upon petitioner’s property right. There is no indication that Texas takes so narrow a view of petitioner’s rights that it may fairly be said that its “only cognizable property right in this respect is in preventing employees from bringing literature onto its property and distributing it there.” Ante, at 573. So far as appears, a Texas property owner may admit certain leaflets onto his property and exclude others, as it pleases him. The Court can only mean that the Board need not take cognizance of any greater property right because the Congress has clearly and constitutionally said so. From its earliest cases construing the National Labor Relations Act the Court has recognized the weight of an employer’s property rights, rights which are explicitly protected from federal interference by the Fifth Amendment to the Constitution. The Court has not been quick to conclude in a given instance that Congress has authorized the displacement of those rights by the federally created rights of the employees. In NLRB v. Bansteel Metallurgical Corp., 306 U. S. 240 (1939), construing another section of the Act, this Court dealt with the Board’s efforts to compel the reinstatement of employees who had been discharged after violating their EASTEX, INC. v. NLRB 581 556 Rehnquist, J., dissenting employer’s property rights by engaging in a sitdown strike. Mr. Chief Justice Hughes wrote for the Court: “We are unable to conclude that Congress intended to compel employers to retain persons in their employ regardless of their unlawful conduct,—to invest those who go on strike with an immunity from discharge for acts of trespass or violence against the employer’s property, which they would not have enjoyed had they remained at work. Apart from the question of the constitutional validity of an enactment of that sort, it is enough to say that such legislative intention should be found in some definite and unmistakable expression. We find no such expression in the cited provision.” Id., at 255. See also id., at 265 (Stone, J., concurring in part). An employer’s property rights must give way only where necessary to effectuate the central purposes of the Act: “to safeguard the rights of self-organization and collective bargaining, and thus by the promotion of industrial peace to remove obstructions to the free flow of commerce as defined in the Act.” Id., at 257. Those rights of self-organization were again recognized six years later in Republic Aviation Corp. v. NLRB, 324 U. S. 793 (1945). There, the Court held that Congress had authorized the Board to displace the property rights of employers where necessary to accommodate the rights of employees to distribute union organizational literature and to wear union insignia. In NLRB n. Babcock & Wilcox Co., 351 U. S. 105 (1956), the Court recognized that nonemployees could also invoke this right to solicit union membership, but it held that the Board’s authority to displace the employer’s property rights in such circumstances was extremely limited.1 Later, 1The Court’s assertion to the contrary notwithstanding, both Babcock and Republic Aviation, Eke this case, involved a “trespass on the employer’s property,” ante, at 571, in that union members sought to over- 582 OCTOBER TERM, 1977 Rehnquist, J., dissenting 437U.S. the Court in Central Hardware Co. v. NLRB, 407 U. S. 539 (1972), explained the limited nature of the intrusion upon property rights permitted by Babcock: “The principle of Babcock is limited to this accommodation between organization rights and property rights. This principle requires a ‘yielding’ of property rights only in the context of an organization campaign. Moreover, the allowed intrusion on property rights is limited to that necessary to facilitate the exercise of employees’ § 7 rights. After the requisite need for access to the employer’s property has been shown, the access is limited to (i) union organizers; (ii) prescribed non working areas of the employer’s premises; and (iii) the duration of organization activity. In short, the principle of accommodation announced in Babcock is limited to labor organization campaigns and the ‘yielding’ of property rights it may require is both temporary and minimal.” 407 U. S., at 544-545? ride the employer’s right to prescribe the conditions of entry to its property. It cannot accept the implications of the dictum in Hudgens v. NLRB, 424 U. S. 507, 521-522, n. 10 (1976), which may in turn be traced back to that portion of the Board’s opinion quoted in Republic Aviation, 324 U. S., at 803-804, n. 10, that this constitutionally protected right may be disregarded where employees are involved simply by characterizing it as a “management inheres [t].” The employer has a property right under Texas law to decide not only who shall come on his property but also the conditions which must be complied with to remain there. The fact that this right may be subordinated by various governmental enactments makes it no less a property right. 21 do not read the reference in Central Hardware to “§ 7 rights” as a suggestion that all rights protected under that section may be allowed to intrude upon an employer’s property rights. The rest of the paragraph clearly limits its application to organization rights, and the Court in a later case suggested that distinctions might be drawn between “lawful economic strike activity” and “organizational activity,” both of which are protected rights under § 7. Hudgens v. NLRB, supra, at 522. Earlier this Term, in Sears, Roebuck & Co. v. Carpenters, 436 U. S. 180 (1978), the Court EASTEX, INC. v. NLRB 583 556 Rehnquist, J., dissenting The Court today cites no case in which it has ever held that anyone, whether an employee or a nonemployee, has a protected right to engage in anything other than organizational activity on an employer’s property. The simple question before us is whether Congress has authorized the Board to displace an employer’s right to prevent the distribution on his property of political material concerning matters over which he has no control.3 In eschewing any analysis of this question, in deference to the supposed expertise of the Board, the Court permits a “ ‘yielding’ of property rights” which is certainly not “temporary”; and I cannot conclude that the deprivation of such a right of property can be dismissed as “minimal.” It may be that Congress has power under the Commerce Clause to require an employer to open his property to such political advocacy, but, if Congress intended to do so, “such a legislative intention should be found in some definite and unmistakable expression.” Fansteel, 306 U. S., at 255. Finding no such expression in the Act, I would not permit the Board to balance away petitioner’s right to exclude political literature from its property. I would reverse the judgment of the Court of Appeals. conceded that trespassory picketing might be protected in some circumstances, but went on to state: “Even on the assumption that picketing to enforce area standards is entitled to the same deference in the Babcock accommodation analysis as organizational solicitation, it would be unprotected in most instances.” Id., at 206 (footnote omitted). No holding of this Court has ever found such a trespass protected. 3 The Court’s complaint that “almost every issue can be viewed by some as political,” ante, at 570 n. 20, contrasts markedly with its earlier assurance, in another context, that “common-sense” distinctions may be drawn between political speech and commercial speech. Ohralik v. Ohio State Bar Assn., 436 U. S. 447, 455-456 (1978). In any case, there is little difficulty in determining whether the employer has the power to affect those matters of which his employees complain. Where he does not, there is no reason to require him to permit such advocacy on his property, even though such activity might arguably be protected under § 7 if committed elsewhere. 584 OCTOBER TERM, 1977 437 U. S. Syllabus PARKER, ACTING COMMISSIONER OF PATENTS AND TRADEMARKS v. FLOOK CERTIORARI TO THE COURT OF CUSTOMS AND PATENT APPEALS No. 77-642. Argued April 25, 1978—Decided June 22, 1978 Respondent’s method for updating alarm limits during catalytic conversion processes, in which the only novel feature is a mathematical formula, held not patentable under § 101 of the Patent Act. The identification of a limited category of useful, though conventional, postsolution applications of such a formula does not make the method eligible for patent protection, since assuming the formula to be within prior art, as it must be, O’Reilly v. Morse, 15 How. 62, respondent’s application contains no patentable invention. The chemical processes involved in catalytic conversion are well known, as are the monitoring of process variables, the use of alarm limits to trigger alarms, the notion that alarm limit values must be recomputed and readjusted, and the use of computers for “automatic process monitoring.” Pp. 588-596. 559 F. 2d 21, reversed. Stevens, J., delivered the opinion of the Court, in which Brennan, White, Marshall, Blackmun, and Powell, JJ., joined. Stewart, J., filed a dissenting opinion, in which Burger, C. J., and Rehnquist, J., joined, post, p. 598. Deputy Solicitor General Wallace argued the cause for petitioner. On the briefs were Solicitor General McCree, Assistant Attorney General Shene field, Richard H. Stern, Joseph F. Nakamura, and Jere W. Sears. D. Dennis Allegretti argued the cause for respondent. With him on the brief were Charles G. Call, Edward W. Remus, and Frank J. Uxa, Jr* *John S. Voorhees and Kenneth E. Krosin filed a brief for the Computer Business Equipment Manufacturers Assn, as amicus curiae urging reversal. Briefs of amici curiae urging affirmance were filed by Carol A. Cohen for Applied Data Research, Inc.; and by Morton C. Jacobs and David Cohen for the Association of Data Processing Service Organizations. Briefs of amici curiae were filed by James W. Geriak for the American PARKER v. FLOOK 585 584 Opinion of the Court Mr. Justice Stevens delivered the opinion of the Court. Respondent applied for a patent on a “Method for Updating Alarm Limits.” The only novel feature of the method is a mathematical formula. In Gottschalk v. Benson, 409 U. S. 63, we held that the discovery of a novel and useful mathematical formula may not be patented. The question in this case is whether the identification of a limited category of useful, though conventional, post-solution applications of such a formula makes respondent’s method eligible for patent protection. I An “alarm limit” is a number. During catalytic conversion processes, operating conditions such as temperature, pressure, and flow rates are constantly monitored. When any of these “process variables” exceeds a predetermined “alarm limit,” an alarm may signal the presence of an abnormal condition indicating either inefficiency or perhaps danger. Fixed alarm limits may be appropriate for a steady operation, but during transient operating situations, such as start-up, it may be necessary to “update” the alarm limits periodically. Respondent’s patent application describes a method of updating alarm limits. In essence, the method consists of three steps: an initial step which merely measures the present value of the process variable (e. g., the temperature); an intermediate step which uses an algorithm1 to calculate an updated alarm-limit value; and a final step in which the actual alarm limit is adjusted to the updated value.2 The only difference Patent Law Assn, et al.; by Richard E. Kurtz, Michael G. Gilman, and Charles A. Huggett for Mobil Oil Corp.; and by Reed C. Lawlor and Theodore H. Lassagne for Software Associates, Inc. 1 We use the word “algorithm” in this case, a« we did in Gottschalk v. Benson, 409 U. S. 63, 65, to mean “[a] procedure for solving a given type of mathematical problem . . . .” 2 Claim 1 of the patent is set forth in the appendix to this opinion, which also contains a more complete description of these three steps. 586 OCTOBER TERM, 1977 Opinion of the Court 437U.S. between the conventional methods of changing alarm limits and that described in respondent’s application rests in the second step—the mathematical algorithm or formula. Using the formula, an operator can calculate an updated alarm limit once he knows the original alarm base, the appropriate margin of safety, the time interval that should elapse between each updating, the current temperature (or other process variable), and the appropriate weighting factor to be used to average the original alarm base and the current temperature. The patent application does not purport to explain how to select the appropriate margin of safety, the weighting factor, or any of the other variables. Nor does it purport to contain any disclosure relating to the chemical processes at work, the monitoring of process variables, or the means of setting off an alarm or adjusting an alarm system. All that it provides is a formula for computing an updated alarm limit. Although the computations can be made by pencil and paper calculations, the abstract of disclosure makes it clear that the formula is primarily useful for computerized calculations producing automatic adjustments in alarm settings.3 The patent claims cover any use of respondent’s formula for updating the value of an alarm limit on any process variable involved in a process comprising the catalytic chemical conversion of hydrocarbons. Since there are numerous processes of that kind in the petrochemical and oil-refining industries,4 the claims cover a broad range of potential uses of the method. They do not, however, cover every conceivable application of the formula. 3App. 13A. 4 Examples mentioned in the abstract of disclosure include naphtha reforming, petroleum distillate and petroleum residuum cracking, hydro- cracking and desulfurization, aromatic hydrocarbon and paraffin isomerization and disproportionation, paraffin-olefin alkylation, and the like. Id., at 8A. PARKER v. FLOOK 587 584 Opinion of the Court II The patent examiner rejected the application. He found that the mathematical formula constituted the only difference between respondent’s claims and the prior art and therefore a patent on this method “would in practical effect be a patent on the formula or mathematics itself.” 5 The examiner concluded that the claims did not describe a discovery that was eligible for patent protection. The Board of Appeals of the Patent and Trademark Office sustained the examiner’s rejection. The Board also concluded that the “point of novelty in [respondent’s] claimed method”6 lay in the formula or algorithm described in the claims, a subject matter that was unpatentable under Benson, supra. The Court of Customs and Patent Appeals reversed. In re Flook, 559 F. 2d 21. It read Benson as applying only to claims that entirely pre-empt a mathematical formula or algorithm, and noted that respondent was only claiming on the use of his method to update alarm limits in a process comprising the catalytic chemical conversion of hydrocarbons. The court reasoned that since the mere solution of the algorithm would not constitute infringement of the claims, a patent on the method would not pre-empt the formula. The Acting Commissioner of Patents and Trademarks filed a petition for a writ of certiorari, urging that the decision of the Court of Customs and Patent Appeals will have a debilitating effect on the rapidly expanding computer “software” industry,7 and will require him to process thousands of addi 6 Id., at 47A. 6 Id., at 60A. 7 The term “software” is used in the industry to describe computer programs. The value of computer programs in use in the United States in 1976 was placed at $43.1 billion, and projected at $70.7 billion by 1980 according to one industry estimate. See Brief for the Computer & Business Equipment Manufacturers Assn, as Amicus Curiae 17-18, n. 16. 588 OCTOBER TERM, 1977 Opinion of the Court 437 U. S. tional patent applications. Because of the importance of the question, we granted certiorari, 434 U. S. 1033. Ill This case turns entirely on the proper construction of § 101 of the Patent Act, which describes the subject matter that is eligible for patent protection.8 It does not involve the familiar issues of novelty and obviousness that routinely arise under §§ 102 and 103 when the validity of a patent is challenged. For the purpose of our analysis, we assume that respondent’s formula is novel and useful and that he discovered it. We also assume, since respondent does not challenge the examiner’s finding, that the formula is the only novel feature of respondent’s method. The question is whether the discovery of this feature makes an otherwise conventional method eligible for patent protection. The plain language of § 101 does not answer the question. It is true, as respondent argues, that his method is a “process” in the ordinary sense of the word? But that was also true of the algorithm, which described a method for converting binary-coded decimal numerals into pure binary numerals, 8 Title 35 U. S. C. § 101 provides: “Whoever invents or discovers any new and useful process, machine, manufacture, or composition of matter, or any new and useful improvement thereof, may obtain a patent therefor, subject to the conditions and requirements of this title.” Section 100 (b) provides: “The term ‘process? means process, art or method, and includes a new use of a known process, machine, manufacture, composition of matter, or material.” 9 The statutory definition of “process” is broad. See n. 8, supra. An argument can be made, however, that this Court has only recognized a process as within the statutory definition when it either was tied to a particular apparatus or operated to change materials to a “different state or thing.” See Cochrane v. Deener, 94 U. S. 780, 787-788. As in Benson, we assume that a valid process patent may issue even if it does not meet one of these qualifications of our earlier precedents. 409 U. S., at 71. PARKER v. FLOOK 589 584 Opinion of the Court that was involved in Gottschalk v. Benson. The holding that the discovery of that method could not be patented as a “process” forecloses a purely literal reading of § 101.10 Reasoning that an algorithm, or mathematical formula, is like a law of nature, Benson applied the established rule that a law of nature cannot be the subject of a patent. Quoting from earlier cases, we said: “ ‘A principle, in the abstract, is a fundamental truth; an original cause; a motive; these cannot be patented, as no one can claim in either of them an exclusive right.’ Le Roy v. Tatham, 14 How. 156, 175. Phenomena of nature, though just discovered, mental processes, and abstract intellectual concepts are not patentable, as they are the basic tools of scientific and technological work.” 409 U. S., at 67. The line between a patentable “process” and an unpatentable “principle” is not always clear. Both are “conception [s] of the mind, seen only by [their] effects when being executed or performed.” Tilghman v. Proctor, 102 U. S. 707, 728. In Benson we concluded that the process application in fact sought to patent an idea, noting that “[t]he mathematical formula involved here has no substantial practical application except in connection with a digital computer, which means that if the judgment below is affirmed, the patent would wholly pre-empt the mathematical formula and in practical effect would be a patent on the algorithm itself.” 409 U. S., at 71-72. Respondent correctly points out that this language does not apply to his claims. He does not seek to “wholly preempt the mathematical formula,” since there are uses of his 10 In Benson we phrased the issue in this way: “The question is whether the method described and claimed is a ‘process’ within the meaning of the Patent Act.” Id., at 64. 590 OCTOBER TERM, 1977 Opinion of the Court 437U.S. formula outside the petrochemical and oil-refining industries that remain in the public domain. And he argues that the presence of specific “post-solution” activity—the adjustment of the alarm limit to the figure computed according to the formula—distinguishes this case from Benson and makes his process patentable. We cannot agree. The notion that post-solution activity, no matter how conventional or obvious in itself, can transform an unpatentable principle into a patentable process exalts form over substance. A competent draftsman could attach some form of post-solution activity to almost any mathematical formula; the Pythagorean theorem would not have been patentable, or partially patentable, because a patent application contained a final step indicating that the formula, when solved, could be usefully applied to existing surveying techniques.11 The concept of patentable subject matter under § 101 is not “like a nose of wax which may be turned and twisted in any direction ... White v. Dunbar, 119 U. S. 47, 51. Yet it is equally clear that a process is not unpatentable simply because it contains a law of nature or a mathematical algorithm. See Eibel Process Co. v. Minnesota & Ontario Paper Co., 261 U. S. 45; Tilghman v. Proctor, supra?2 For 11 It should be noted that in Benson there was a specific end use contemplated for the algorithm—utilization of the algorithm in computer programming. See In re Chatfield, 545 F. 2d 152, 161 (CCPA 1976) (Rich, J., dissenting). Of course, as the Court pointed out, the formula had no other practical application; but it is not entirely clear why a process claim is any more or less patentable because the specific end use contemplated is the only one for which the algorithm has any practical application. 12 In Eibel Process Co. the Court upheld a patent on an improvement on a papermaking machine that made use of the law of gravity to enhance the flow of the product. The patentee, of course, did not claim to have discovered the force of gravity, but that force was an element in his novel conception. Tilghman v. Proctor involved a process claim for “ ‘the manufacturing PARKER v. FLOOK 591 584 Opinion of the Court instance, in Mackay Radio & Telegraph Co. v. Radio Corp, of America, 306 U. S. 86, the applicant sought a patent on a directional antenna system in which the wire arrangement was determined by the logical application of a mathematical formula. Putting the question of patentability to one side as a preface to his analysis of the infringement issue, Mr. Justice Stone, writing for the Court, explained: “While a scientific truth, or the mathematical expression of it, is not patentable invention, a novel and useful structure created with the aid of knowledge of scientific truth may be.” Id., at 94. Funk Bros. Seed Co. v. Kalo Co., 333 U. S. 127, 130, expresses a similar approach: “He who discovers a hitherto unknown phenomenon of nature has no claim to a monopoly of it which the law recognizes. If there is to be invention from such a discovery, it must come from the application of the law of nature to a new and useful end.” Mackay Radio and Funk Bros, point to the proper analysis for this case: The process itself, not merely the mathematical algorithm, must be new and useful. Indeed, the novelty of the mathematical algorithm is not a determining factor at all. Whether the algorithm was in fact known or unknown at the time of the claimed invention, as one of the “basic tools of scientific and technological work,” see Gottschalk v. Benson, of fat acids and glycerine from fatty bodies.’ ” The Court distinguished the process from the principle involved as follows: “[T]he claim of the patent is not for a mere principle. The chemical principle or scientific fact upon which it is founded is, that the elements of neutral fat require to be severally united with an atomic equivalent of water in order to separate from each other and become free. This chemical fact was not discovered by Tilghman. He only claims to have invented a particular mode of bringing about the desired chemical union between the fatty elements and water.” 102 U. S., at 729. 592 OCTOBER TEAM, 1977 Opinion of the Court 437U.S. 409 U. S., at 67, it is treated as though it were a familiar part of the prior art. This is also the teaching of our landmark decision in O’Reilly v. Morse, 15 How. 62. In that case the Court rejected Samuel Morse’s broad claim covering any use of electromagnetism for printing intelligible signs, characters, or letters at a distance. Id., at 112-121. In reviewing earlier cases applying the rule that a scientific principle cannot be patented, the Court placed particular emphasis on the English case of Neilson v. Harford, Web. Pat. Cases 295, 371 (1844), which involved the circulation of heated air in a furnace system to increase its efficiency. The English court rejected the argument that the patent merely covered the principle that furnace temperature could be increased by injecting hot air, instead of cold into the furnace. That court’s explanation of its decision was relied on by this Court in Morse: 11 ‘It is very difficult to distinguish it [the Neilson patent] from the specification of a patent for a principle, and this at first created in the minds of the court much difficulty; but after full consideration, we think that the plaintiff does not merely claim a principle, but a machine, embodying a principle, and a very valuable one. We think the case must be considered as if the principle being well known, the plaintiff had first invented a mode of applying it . . . .’ ” 15 How., at 115 (emphasis added).13 We think this case must also be considered as if the principle or mathematical formula were well known. Respondent argues that this approach improperly imports into § 101 the considerations of “inventiveness” which are the proper concerns of §§ 102 and 103.14 This argument is based on two fundamental misconceptions. 13 See also Risdon Locomotive Works n. Medart, 158 U. S. 68; Tilghman v. Proctor, supra. 14 Sections 102 and 103 establish certain conditions, such as novelty and nonobviousness, to patentability. PARKER v. FLOOK 593 584 Opinion of the Court First, respondent incorrectly assumes that if a process application implements a principle in some specific fashion, it automatically falls within the patentable subject matter of § 101 and the substantive patentability of the particular process can then be determined by the conditions of §§ 102 and 103. This assumption is based on respondent’s narrow reading of Benson, and is as untenable in the context of § 101 as it is in the context of that case. It would make the determination of patentable subject matter depend simply on the draftsman’s art and would ill serve the principles underlying the prohibition against patents for “ideas” or phenomena of nature. The rule that the discovery of a law of nature cannot be patented rests, not on the notion that natural phenomena are not processes, but rather on the more fundamental understanding that they are not the kind of “discoveries” that the statute was enacted to protect.15 The obligation to determine what type of discovery is sought to be patented must precede the determination of whether that discovery is, in fact, new or obvious. Second, respondent assumes that the fatal objection to his application is the fact that one of its components—the mathe 15 The underlying notion is that a scientific principle, such as that expressed in respondent’s algorithm, reveals a relationship that has always existed. “An example of such a discovery [of a scientific principle] was Newton’s formulation of the law of universal gravitation, relating the force of attraction between two bodies, F, to their masses, m and m', and the square of the distance, d, between their centers, according to the equation F=mm'/d2. But this relationship always existed—even before Newton announced his celebrated law. Such 'mere’ recognition of a theretofore existing phenomenon or relationship carries with it no rights to exclude others from its enjoyment. . . . Patentable subject matter must be new (novel); not merely heretofore unknown. There is a very compelling reason for this rule. The reason is founded upon the proposition that in granting patent rights, the public must not be deprived of any rights that it theretofore freely enjoyed.” P. Rosenberg, Patent Law Fundamentals, §4, p. 13 (1975). 594 OCTOBER TERM, 1977 Opinion of the Court 437 U. S. matical formula—consists of unpatentable subject matter. In countering this supposed objection, respondent relies on opinions by the Court of Customs and Patent Appeals which reject the notion “that a claim may be dissected, the claim components searched in the prior art, and, if the only component found novel is outside the statutory classes of invention, the claim may be rejected under 35 U. S. C. § 101.” In re Chatfield, 545 F. 2d 152, 158 (CCPA 1976).16 Our approach to respondent’s application is, however, not at all inconsistent with the view that a patent claim must be considered as a whole. Respondent’s process is unpatentable under § 101, not because it contains a mathematical algorithm as one component, but because once that algorithm is assumed to be within the prior art, the application, considered as a whole, contains no patentable invention. Even though a phenomenon of nature or mathematical formula may be well known, an inventive application of the principle may be patented. Conversely, the discovery of such a phenomenon cannot support a patent unless there is some other inventive concept in its application. Here it is absolutely clear that respondent’s application contains no claim of patentable invention. The chemical processes involved in catalytic conversion of hydrocarbons are well known, as are the practice of monitoring the chemical process variables, the use of alarm limits to trigger alarms, the notion that alarm limit values must be recomputed and readjusted, and the use of computers for “automatic monitoring-alarming.” 17 Respondent’s application simply provides a new and presumably better method for calculating alarm limit 16 Section 103, by its own terms, requires that a determination of obviousness be made by considering “the subject matter as a whole.” 35 U. S. C. § 103. Although this does not necessarily require that analysis of what is patentable subject matter under § 101 proceed on the same basis, we agree that it should. 17 App. 22. PARKER v. FLOOK 595 584 Opinion of the Court values. If we assume that that method was also known, as we must under the reasoning in Morse, then respondent’s claim is, in effect, comparable to a claim that the formula 2?rr can be usefully applied in determining the circumference of a wheel.18 As the Court of Customs and Patent Appeals has explained, “if a claim is directed essentially to a method of calculating, using a mathematical formula, even if the solution is for a specific purpose, the claimed method is nonstatutory.” In re Richman, 563 F. 2d 1026,1030 (1977). To a large extent our conclusion is based on reasoning derived from opinions written before the modern business of developing programs for computers was conceived. The youth of the industry may explain the complete absence of precedent supporting patentability. Neither the dearth of precedent, nor this decision, should therefore be interpreted as reflecting a judgment that patent protection of certain novel and useful computer programs will not promote the progress of science and the useful arts, or that such protection is undesirable as a matter of policy. Difficult questions of policy concerning the kinds of programs that may be appropriate for patent protection and the form and duration of such protection can be answered by Congress on the basis of current empirical data not equally available to this tribunal.19 18 Respondent argues that the inventiveness of his process must be determined as of “the time the invention is made” under § 103, and that, therefore, it is improper to judge the obviousness of his process by assessing the application of the formula as though the formula were part of the prior art. This argument confuses the issue of patentable subject matter under § 101 with that of obviousness under § 103. Whether or not respondent’s formula can be characterized as “obvious,” his process patent rests solely on the claim that his mathematical algorithm, when related to a computer program, will improve the existing process for updating alarm units. Very simply, our holding today is that a claim for an improved method of calculation, even when tied to a specific end use, is unpatentable subject matter under § 101. 19 Articles assessing the merits and demerits of patent protection for computer programming are numerous. See, e. g., Davis, Computer Pro- 596 OCTOBER TERM, 1977 Appendix to opinion of the Court 437 U. S. It is our duty to construe the patent statutes as they now read, in light of our prior precedents, and we must proceed cautiously when we are asked to extend patent rights into areas wholly unforeseen by Congress. As Mr. Justice White explained in writing for the Court in Deepsouth Packing Co. v. Laitram Corp., 406 U. S. 518, 531: “[W]e should not expand patent rights by overruling or modifying our prior cases construing the patent statutes, unless the argument for expansion of privilege is based on more than mere inference from ambiguous statutory language. We would require a clear and certain signal from Congress before approving the position of a litigant who, as respondent here, argues that the beachhead of privilege is wider, and the area of public use narrower, than courts had previously thought. No such signal legitimizes respondent’s position in this litigation.” The judgment of the Court of Customs and Patent Appeals is Reversed. APPENDIX TO OPINION OF THE COURT Claim 1 of the patent describes the method as follows: “1. A method for updating the value of at least one alarm limit on at least one process variable involved in a process comprising the catalytic chemical conversion of hydrocarbons wherein said alarm limit has a current value of Bo+K “wherein Bo is the current alarm base and K is a predetermined alarm offset which comprises: grams and Subject Matter Patentability, 6 Rutgers J. of Computers and Law 1 (1977), and articles cited therein, at 2 n. 5. Even among those who favor patentability of computer programs, there is questioning of whether the 17-year protection afforded by the current Patent Act is either needed or appropriate. See id., at 20 n. 133. PARKER V. FLOOK 597 584 Appendix to opinion of the Court “(1) Determining the present value of said process variable, said present value being defined as PVL; “(2) Determining a new alarm base Bi, using the following equation: Bi=Bo(1.0-F)+PVL(F) “where F is a predetermined number greater than zero and less than 1.0; “(3) Determining an updated alarm limit which is defined as Bi-j-K; and thereafter “(4) Adjusting said alarm limit to said updated alarm limit value.” App. 63A. In order to use respondent’s method for computing a new limit, the operator must make four decisions. Based on his knowledge of normal operating conditions, he first selects the original “alarm base” (Bo); if a temperature of 400 degrees is normal, that may be the alarm base. He next decides on an appropriate margin of safety, perhaps 50 degrees; that is his “alarm offset” (K). The sum of the alarm base and the alarm offset equals the alarm limit. Then he decides on the time interval that will elapse between each updating; that interval has no effect on the computation although it may, of course, be of great practical importance. Finally, he selects a weighting factor (F), which may be any number between 99% and 1%,* and which is used in the updating calculation. If the operator has decided in advance to use an original alarm base (Bo) of 400 degrees, a constant alarm offset (K) of 50 degrees, and a weighting factor (F) of 80%, the only additional information he needs in order to compute an updated alarm limit (UAV), is the present value of the process variable (PVL). The computation of the updated alarm limit according to respondent’s method involves these three steps: First, at the predetermined interval, the process variable *More precisely, it is defined as a number greater than 0, but less than 1. 598 OCTOBER TERM, 1977 Stewart, J., dissenting 437U.S. is measured; if we assume the temperature is then 425 degrees, PVL will then equal 425. Second, the solution of respondent’s novel formula will produce a new alarm base (Bi) that will be a weighted average of the preceding alarm base (Bo) of 400 degrees and the current temperature (PVL) of 425. It will be closer to one or the other depending on the value of the weighting factor (F) selected by the operator. If F is 80%, that percentage of 425 (340) plus 20% (1—F) of 400 (80) will produce a new alarm base of 420 degrees. Third, the alarm offset (K) of 50 degrees is then added to the new alarm base (Bi) of 420 to produce the updated alarm limit (UAV) of 470. The process is repeated at the selected time intervals. In each updating computation, the most recently calculated alarm base and the current measurement of the process variable will be substituted for the corresponding numbers in the original calculation, but the alarm offset and the weighting factor will remain constant. Mr. Justice Stewart, with whom The Chief Justice and Mr. Justice Rehnquist join, dissenting. It is a commonplace that laws of nature, physical phenomena, and abstract ideas are not patentable subject matter.1 A patent could not issue, in other words, on the law of gravity, or the multiplication tables, or the phenomena of magnetism, or the fact that water at sea level boils at 100 degrees centigrade and freezes at zero—even though newly discovered. Le Roy v. Tatham, 14 How. 156, 175; O'Reilly v. Morse, 15 How. 62, 112-121; Rubber-Tip Pencil Co. v. Howard, 20 Wall. 1 Title 35 U. S. C. § 101 provides: “Whoever invents or discovers any new and useful process, machine, manufacture, or composition of matter, or any new and useful improvement thereof, may obtain a patent therefor, subject to the conditions and requirements of this title.” PARKER v. FLOOK 599 584 Stewart, J., dissenting 498, 507; Tilghman v. Proctor, 102 U. S. 707; Mackay Radio & Telegraph Co. v. Radio Corp, of America, 306 U. S. 86, 94; Funk Bros. Seed Co. v. Kalo Co., 333 U. S. 127, 130. The recent case of Gottschalk v. Benson, 409 U. S. 63, stands for no more than this long-established principle, which the Court there stated in the following words: “Phenomena of nature, though just discovered, mental processes, and abstract intellectual concepts are not patentable, as they are the basic tools of scientific and technological work.” Id., at 67. In Benson the Court held unpatentable claims for an algorithm that “were not limited to any particular art or technology, to any particular apparatus or machinery, or to any particular end use.” Id., at 64. A patent on such claims, the Court said, “would wholly pre-empt the mathematical formula and in practical effect would be a patent on the algorithm itself.” Id., at 72. The present case is a far different one. The issue here is whether a claimed process2 loses its status of subject-matter patentability simply because one step in the process would not be patentable subject matter if considered in isolation. The Court of Customs and Patent Appeals held that the process is patentable subject matter, Benson being inapplicable since “ [t]he present claims do not preempt the formula or algorithm contained therein, because solution of the algorithm, per se, would not infringe the claims.” In re Flook, 559 F. 2d 21, 23. That decision seems to me wholly in conformity with basic principles of patent law. Indeed, I suppose that thousands of processes and combinations have been patented that contained one or more steps or elements that themselves would have been . 2 Title 35 IT. S. C. § 100 (b) provides: “The term 'process’ means process, art or method, and includes a new use of a known process, machine, manufacture, composition of matter, or material.” 600 OCTOBER TERM, 1977 Stewart, J., dissenting 437U.S. unpatentable subject matter.3 Eibel Process Co. v. Minnesota & Ontario Paper Co., 261 U. S. 45, is a case in point. There the Court upheld the validity of an improvement patent that made use of the law of gravity, which by itself was clearly unpatentable. See also, e. g., Tilghman v. Proctor, supra. The Court today says it does not turn its back on these well-settled precedents, ante, at 594, but it strikes what seems to me an equally damaging blow at basic principles of patent law by importing into its inquiry under 35 U. S. C. § 101 the criteria of novelty and inventiveness. Section 101 is concerned only with subject-matter patentability. Whether a patent will actually issue depends upon the criteria of §§ 102 and 103, which include novelty and inventiveness, among many others. It may well be that under the criteria of §§ 102 and 103 no patent should issue on the process claimed in this case, because of anticipation, abandonment, obviousness, or for some other reason. But in my view the claimed process clearly meets the standards of subject-matter patentability of § 101. In short, I agree with the Court of Customs and Patent Appeals in this case, and with the carefully considered opinions of that court in other cases presenting the same basic issue. See In re Freeman, 573 F. 2d 1237; In re Richman, 563 F. 2d 1026; In re De Castelet, 562 F. 2d 1236; In re Deutsch, 553 F. 2d 689; In re Chatfield, 545 F. 2d 152. Accordingly, I would affirm the judgment before us. 3 In Gottschalk v. Benson, the Court equated process and product patents for the purpose of its inquiry: “We dealt there with a 'product’ claim, while the present case deals with a 'process’ claim. But we think the same principle applies.” 409 U. S., at 67-68. CALIFORNIA v. TEXAS 601 Per Curiam CALIFORNIA v. TEXAS ON MOTION FOR LEAVE TO FILE A BILL OF COMPLAINT No. 76, Orig. Argued March 29, 1978—Decided June 22, 1978 Jerome B. Falk, Jr., argued the cause for plaintiff. With him on the briefs were Myron Siedorj, James R. Bimberg, and Steven L. Mayer. John L. Hill, Attorney General of Texas, argued the cause for defendant. With him on the brief were David M. Kendall, First Assistant Attorney General, Lee C. Clyburn, Administrative Assistant Attorney General, Rick Harrison, Special Assistant Attorney General, and David Deaderick and Rick Arnett, Assistant Attorneys General. Per Curiam. The motion for leave to file a bill of complaint is denied. Mr. Justice Brennan, concurring. I agree with Mr. Justice Stewart and Mr. Justice Powell that “in light of Edelman v. Jordan, 415 U. S. 651 (1974), this Court’s decision in Worcester County Trust Co. v. Riley, 302 U. S. 292 (1937), no longer can be regarded as a bar against the use of federal interpleader by estates threatened with double death taxation because of possible inconsistent adjudications of domicile.” Post, at 615. I am not so sure as they that Texas v. Florida, 306 U. S. 398 (1939), was wrongly decided. But, whatever the case, I would still deny California’s motion to file a bill of complaint at this time. If we have jurisdiction at all, that jurisdiction certainly does not attach until it can be shown that two States may possibly be able to obtain conflicting adjudications of domicile. That showing has not been made at this time in this case, since it may well be possible for the Hughes estate to 602 437 U. S. OCTOBER TERM, 1977 Stewart, J., concurring obtain a judgment under the Federal Interpleader Statute, 28 U. S. C. § 1335, from a United States district court, which would be binding on both California and Texas. In this event, the precondition for our original jurisdiction would be lacking. Accordingly, I would deny California’s motion, at least until such time as it is shown that such a statutory interpleader action cannot or will not be brought. Mr. Justice Stewart, with whom Mr. Justice Powell and Mr. Justice Stevens join, concurring. California seeks to invoke the original and exclusive jurisdiction of this Court to settle a dispute with the State of Texas over the question of which State has the power to collect death taxes from the estate of the late Howard Robard Hughes. The Court today, without explanation of any kind, evidently concludes that California’s complaint does not state a claim within our original and exclusive jurisdiction. This conclusion seems to me squarely contrary to a longstanding precedent of this Court, the case of Texas v. Florida, 306 U. S. 398. I have joined in the order denying California’s motion for leave to file this complaint only because I think Texas v. Florida was wrongly decided and should be overruled. I According to the complaint, California imposes an inheritance tax on the real and tangible personal property located within its borders, and upon the intangible personalty wherever situated, of a person domiciled in the State at the time of his death, and Texas follows precisely the same policy.1 1 Tangible personal property and realty are constitutionally subject to taxation only at the place of situs. See Union Refrigerator Transit Co. N. Kentucky, 199 U. S. 194; City Bank Farmers Trust Co. n. Schnader, 293 U. S. 112. As will be developed more fully, infra, at 607-610, intangible personal property may, at least theoretically, be taxed only at the place of the owner’s domicile. First Nat. Bank v. Maine, 284 U. S. 312. CALIFORNIA v. TEXAS 603 601 Stewart, J., concurring The complaint alleges that the taxing authorities in each State are claiming in good faith that the decedent Hughes was domiciled in their State at the time of his death, and have instituted proceedings to tax all the assets of the estate within the jurisdiction, as well as the intangibles (consisting of shares of stock in a single holding company) that constitute the great bulk of the estate’s assets.2 The common law in both States recognizes, as a theoretical matter, that a person has only one domicile for purposes of death taxes. Nevertheless, the complaint alleges, since neither Texas nor California is or will become a party to the proceedings in the other’s courts, neither will be bound by an adverse determination of domicile in the other’s forum. Finally, and at the crux of the dispute, the complaint alleges that if both California and Texas obtain judgments for estate taxes in their respective courts and impose their taxes on the basis of the valuation of assets set forth in the federal estate tax return, the estate’s total liability for federal and state taxes will exceed its net value. Thus, the complaint alleges that if the United States and Texas were to collect the taxes claimed by them, and if the California courts should ultimately determine that Hughes was a domiciliary of California at the time of his death, then California would be left with an entirely valid tax judgment that would be uncollectible to the extent of about $21 million. In sum, the complaint alleges that “because there is no other means by which the conflicting tax claims of Texas and California can be resolved, this Court is the only forum which can determine the question of decedent’s domicile in a man-ner that will bind the interested parties and assure that the state of domicile, if California or Texas, will be able to collect the tax.” California invokes the original and exclusive juris 2 In each State the personal representative of the Hughes estate is contesting the tax claim, asserting that Hughes died domiciled in Nevada— the only State in the Union without death taxes. 604 OCTOBER TERM, 1977 Stewart, J., concurring 437U.S. diction of this Court on the authority of Texas v. Florida, supra. II In Texas n. Florida this Court accepted original jurisdiction of Texas’ complaint “in the nature of a bill of interpleader, brought to determine the true domicile of [a] decedent as the basis of rival claims of four states for death taxes upon his estate . . . .” 306 U. S7 at 401. Texas and each of the three defendant States claimed that the decedent, Colonel Edward Green, son of the legendary Hetty Green,3 was its domiciliary and that it was entitled to collect death taxes upon his intangible property wherever located, as well as upon his tangible property within the State. None of the States had reduced its tax claim to judgment, but all conceded that the decedent’s estate was insufficient to satisfy the total amount of taxes claimed: that is, if all four States were successful in their own courts and obtained judgments for taxes in the full amount claimed, the estate would be insufficient to cover all of the claims.4 Although none of the parties raised any question of this Court’s jurisdiction, the Court considered the question sua sponte. It held that since the suit was between States, Art. Ill, § 2, of the Constitution conferred original jurisdiction to decide the case so long as “the issue framed by the pleadings 3 See 7 Dictionary of American Biography 545 (1931). 4 The case had been assigned to a Special Master and fully litigated on the merits before the Court raised the question of its jurisdiction sua sponte. The Special Master found that the net estate would amount to $36,137,335, and that the total tax claims of the United States and the four claiming States was $37,727,213—roughly $17.5 million by the United States, $4.6 naillion each by Texas and Florida, $5 million by Massachusetts, and $6 million by New York. 306 U. S., at 409 n. 2. Since the assets of the estate fell short of the total tax claims by only about $1.6 million, it was clear that there would be no shortfall unless all four state claims were sustained, and indeed that no State would go completely unsatisfied in its tax judgment even if the claims of all four States were sustained. CALIFORNIA v. TEXAS 605 601 Stewart, J., concurring constitutes a justiciable ‘case’ or ‘controversy’ within the meaning of the Constitutional provision, and . . . the facts alleged and found afford an adequate basis for relief according to accepted doctrines of the common law or equity systems of jurisprudence . . . .” 306 U. S., at 405. The Court found such a basis for relief by analogizing the suit to a bill in the nature of interpleader. This procedure had developed in equity to avert the “risk of loss ensuing from the demands in separate suits of rival claimants to the same debt or legal duty” by requiring the claimants to “litigate in a single suit their ownership of the asserted claim.” Id., at 405-406.5 Since the law of each of the claiming States provided that a decedent could be domiciled in only one State for purposes of death taxes, the Court held that the competing tax claims were in fact conflicting claims to the same single legal duty. Thus viewing the suit as one in the nature of interpleader, the Court also found that the controversy was ripe for decision. Since each State’s claim was sufficiently substantial to support a finding of domicile, there was a “fair probability” that each would be successful in its own courts and that the estate’s assets would be insufficient to meet all of the claims. The Court therefore found a justiciable present controversy in the substantial “risk of loss [to] the state lawfully entitled to collect the tax.” Id., at 410-411. The Court perceived no jurisdictional frailty in the fact that none of the claiming States had completed proceedings to collect its inheritance tax, since a plaintiff in an interpleader action was ordinarily not required to await actual institution of independent suits: “ [I] t is enough if he shows that conflicting claims are asserted 5 In true interpleader the stakeholder bringing suit asserts no interest in the fund. The bill in the nature of interpleader, by contrast, allows an interested claimant to seek adjudication of all claims to the fund including his own. See id., at 406. 606 OCTOBER TERM, 1977 Stewart, J., concurring 437U.S. and that the consequent risk of loss is substantial.” Id., at 406.6 The facts alleged in the complaint now before us are indistinguishable in all material respects from those on which jurisdiction was based in Texas n. Florida.1 This Court has original and exclusive jurisdiction of disputes between two or more States, 28 U. S. C. § 1251 (a)(1), and it has a responsibility to exercise that jurisdiction when it is properly invoked. See Cohens n. Virginia, 6 Wheat. 264, 404; Massachusetts v. Missouri, 308 U. S. 1, 19-20. If Texas n. Florida was correctly decided, the Court, therefore, is under a duty in this case to grant California’s motion to file its complaint. I believe, however, that Texas v. Florida was wrongly decided. Its conclusion that there was a case or controversy among the claiming States depended entirely on the analogy to a suit in the nature of interpleader to settle the question of the decedent’s domicile. Yet it seems to me that in resting upon that analogy the Court focused erroneously on the plight of the estate, which was indeed confronted with a “substantial likelihood” of multiple and inconsistent tax claims, and overlooked the fact that the dispute among the claiming States—stemming solely from the possibility that the estate might be insufficient to satisfy all of their claims—was not a case or controversy in the constitutional sense. 6 On the merits the Court confirmed the Master’s finding that Colonel Green was domiciled in Massachusetts at the time of his death, and that Massachusetts was therefore the only State lawfully entitled to tax the intangible personal property in his estate. 7 Texas does not concede that all tax claims will necessarily exceed the value of the Hughes estate, and argues that this fact distinguishes the present case from Texas v. Florida. But in that case it was not the concessions of the parties that did or could confer jurisdiction upon the Court. Rather, the Court held that a mere “fair probability” of inconsistent adjudications and consequent “substantial” risk of loss was sufficient to create a constitutional case or controversy in the nature of interpleader. The claims here are, in fact, no more speculative than the claims in that case. See n. 4, supra. CALIFORNIA v. TEXAS 607 601 Stewart, J., concurring III The Court’s readiness in Texas v. Florida to accept the interpleader analogy is understandable in the context of the then state of the law governing multiple taxation of intangibles. Before 1931 it had been taken as settled that, because the question of domicile was purely one of state law, it “must in many cases be impossible to have a single controlling decision upon the question,” unless all interested parties could by chance or voluntary appearance be brought before a single forum. Baker v. Baker, Eccles de Co., 242 U. S. 394, 405. But when this Court held in 1931 that shares of stock and other intangible property could constitutionally “be subjected to a death transfer tax by one state only,” that being the State of the decedent’s domicile, First National Bank n. Maine, 284 U. S. 312, 328-330, it seemed implicit that there must be some means of protecting that right in a federal forum. The obvious next question was under what federal-court procedures conflicting state claims of domicile were to be resolved.8 The somewhat unexpected answer came in Worcester County Trust Co. v. Riley, 302 U. S. 292, which held that, at least for the ordinary estate, there was no means of forcing unwilling States to litigate the question of domicile, and the consequent right to tax the estate’s intangibles, in a federal district court. In that case the estate of a decedent attempted to sue the taxing officials of two different States under the recently enacted Federal Interpleader Statute, 28 U. S. C. § 1335, to obtain a single, binding determination of the decedent’s domicile at the time of his death. Despite the broad language of the First National Bank case, the Court held that “[n]either the Fourteenth Amendment nor the full faith and credit clause requires uniformity in the decisions of the courts 8 See Chafee, Federal Interpleader Since the Act of 1936, 49 Yale L. J. 377, 383-393 (1940), and authorities collected, id., at 383 n. 17; Nash, And Again Multiple Taxation?, 26 Geo. L. J. 288, 297 (1938). 608 OCTOBER TERM, 1977 Stewart, J., concurring 437U.S. of different states as to the place of domicil, where the exertion of state power is dependent upon domicil within its boundaries.” 302 U. S., at 299. After thus making clear that the imposition of multiple estate taxes on the basis of inconsistent adjudications of domicile presented no federal constitutional question, the opinion of the Court went on to foreclose recourse to the federal interpleader jurisdiction. Federal interpleader is based on diversity-of-citizenship jurisdiction, see State Farm Fire & Casualty Co. v. Tashire, 386 U. S. 523, 530-531, and a federal question is ordinarily not required.9 But because the state tax officials were not acting unconstitutionally in attempting to impose taxes on the basis of valid state-court judgments, the Court held that the interpleader action was in substance a suit against the States themselves, and therefore barred by the Eleventh Amendment. See Ex parte Young, 209 U. S. 123.10 9 There was no doubt that the dispute was in fact ideally suited to resolution by means of federal interpleader. Professor Chafee, upon whose work the Federal Interpleader Statute was largely based, believed that conflicting state claims of domicile presented a situation in which interpleader was “badly needed.” Chafee, supra, n. 8, at 379. It is, he observed, “highly unfair for both state governments to tell the taxpayer, ‘You have to pay only one tax,’ and then make him pay twice.” Id., at 384. He pointed out that the paradox of inconsistent adjudications of a theoretically single domicile is one created by our federal system of government: “In a nation with a unified government, the situation in which estates of decedents are here left remediless would be impossible. Either only one agency would impose death taxes; or else a single court of review would determine domicile as between two local taxing agencies. . . . Somewhere within that federal system we should be able to find remedies for the frictions which that system creates.” Id., at 388. I believe such a remedy is now available. See n. 10, infra. 101 think this holding has been substantially undercut by subsequent developments. In Edelman v. Jordan, 415 U. S. 651, the Court expressed an understanding of the Eleventh Amendment quite different from that manifested in Worcester County Trust Co. v. Riley, 302 U. S. 292. Thus it would appear that an estate confronted with multiple tax claims by two or more States could now bring an interpleader action in a federal CALIFORNIA v. TEXAS 609 601 Stewart, J., concurring When the identical type of dispute was placed before this Court two years later in Texas v. Florida, the Court was thus understandably persuaded to view the complaint as presenting a question of domicile resolvable by a suit in the nature of interpleader to determine which State could alone impose the death tax.11 But the issue of the decedent’s domicile in that case was merely a coincidental premise to the real basis of the dispute among the States—the risk that the claims of the competing States would exceed the net value of the estate, and that “the state lawfully entitled to collect the tax” would find itself unable to do so.12 As the opinion in Texas v. Florida made clear, insofar as the rights of the estate were concerned, each of the four States was “lawfully entitled” to collect the tax: “[T]wo or more states may each constitutionally assess death taxes on a decedent’s intangibles upon a judicial determination that the district court seeking declaratory and injunctive relief against the tax officials of each State. I do not believe that the Tax Injunction Act, 28 U. S. C. § 1341, would preclude such a suit, if it were clear that the taxing States would not afford the estate a “plain, speedy and efficient remedy” for its claim that it should not be subjected to multiple taxes, e. g., by recognizing an earlier determination of domicile by a sister State. 11 At least one commentator so viewed the case when it was pending before the Court: “Texas v. Florida may become the wedge to open the door slammed in Worcester County Trust Co. v. Riley. It is so hard to believe that the Court will persist in its refusal to aid the states in the difficulty, one seizes on the slightest possibility to hope that there may yet come a solution.” Nash, supra n. 8, at 314. 12 At oral argument on Texas’ original motion for leave to file a bill of complaint in that case “the Court indicated that there was no justiciable controversy unless the assets of the estate were insufficient to pay the tax claims of all four of the states.” Tweed & Sargent, Death and Taxes are Certain—But What of Domicile, 53 Harv. L. Rev. 68, 75 (1939). This first complaint was dismissed without prejudice. Texas v. New York, 300 U. S. 642. It was upon Texas’ amended complaint, plainly alleging “on information and belief” that the assets were insufficient to meet all claims, that the Court took jurisdiction in Texas v. Florida. See also Massachusetts v. Missouri, 308 U. S. 1, 15. 610 OCTOBER TERM, 1977 Stewart, J., concurring 437 U. S. decedent was domiciled within it . . . .” 306 U. S., at 410. And a few months later the Court elaborated on this doctrine when it denied a motion to file a complaint in Massachusetts v. Missouri, 308 U. S. 1. There the Court made clear that both States were equally entitled to impose a tax so long as there was no risk that the estate would be depleted: “Missouri, in claiming a right to recover taxes from the respondent trustees, or in taking proceedings for collection, is not injuring Massachusetts. By the allegations, the property held in Missouri is amply sufficient to answer the claims of both States and recovery by either does not impair the exercise of any right the other may have.” Id., at 15. Thus, even after Texas v. Florida, there was still no forum in which an estate confronted with conflicting tax claims could obtain a single, binding adjudication of domicile. So long as it was able to pay each State’s claim, it was required to pay taxes to any State that obtained a judgment of domicile in its own courts. And, so long as the assets of the estate were sufficient to answer all claims, a State could not obtain an adjudication in this Court as to which State had “the jurisdiction and lawful right” to impose inheritance taxes. Only in the very rare situation when a decedent’s estate was threatened with death tax claims of two or more States that together exceeded its assets, and only if one of the competing States then invoked this Court’s original jurisdiction, would the Court undertake to decide the decedent’s true domicile and grant one State the exclusive right to tax the decedent’s estate. IV In reality the facts in Texas v. Florida, as well as the allegations in the complaint now before us, contain the seeds of two distinct lawsuits. One is a dispute between two States as to the proper division of a finite sum of money. The other is a suit in the nature of interpleader to settle the question of a decedent’s domicile for purposes of the taxes to be imposed CALIFORNIA v. TEXAS 611 601 Stewart, J., concurring upon his estate. But the suit in the nature of interpleader is not within the original and exclusive jurisdiction of this Court because it is not a dispute between States. And the dispute between the States, if indeed it is justiciable at all, is certainly not yet a case or controversy within the constitutional meaning of that term. A What California seeks in the present complaint is a determination of where Howard Hughes was domiciled at the time of his death. It is clear to me that, if presented by a proper party in a proper forum, this determination could and should be made in response to a bill of interpleader. See nn. 9 and 10, supra. But if interpleader generally affords no remedy to a decedent’s estate that is faced with the threat of multiple taxation, there is no logical reason why the remedy should be available in the rare situation where the multiple taxation would wipe the estate out entirely. If it is unfair to subject an estate to two domicile-based taxes when all agree that it is possible to have only one domicile, that unfairness is just as great, if not greater, when a decedent’s estate is able to pay the taxes to both States. It must be recognized, however, that what is involved is unfairness to the estate, not to the taxing States. The remedy of interpleader exists, if at all, to require litigation of the inconsistent tax claims in a single forum in order to avert the risk of loss to the estate that would result from separate adjudications. But the only live controversy in such a suit is between each State and the decedent’s estate as to the legal obligation to pay death taxes. There is, in fact, no present dispute between the claiming States. In the present case, it would be of no possible concern to either California or Texas that the other might adjudge Hughes a domiciliary and succeed in taxing his estate, except for the possibility that the other’s tax might exhaust the 612 OCTOBER TERM, 1977 Stewart, J., concurring 437U.S. estate entirely before it is able to satisfy its own tax judgment. Thus to the extent that the concern of this action is to prevent the possibility that the estate will be subjected to double taxation, it does not present a dispute between two States within the original and exclusive jurisdiction of this Court. For a State may seek the aid of this Court only to protect its own interests, not the interests of others. See Massachusetts v. Missouri, supra, at 15. B The dispute between California and Texas, therefore, is not really over which of them has the right to impose a domiciliary tax upon the Hughes estate. Indeed the dilemma of multiple taxation arises only because the Constitution permits both States to impose the tax. Worcester County Trust Co. n. Riley, 302 U. S. 292.13 The real dispute arises solely from the risk that one of the States will be left with an entirely valid but uncollectible tax judgment. Massachusetts v. Missouri, supra, at 15, The conflict would be equally real if the two States were staking their tax claims to the finite assets of the estate on entirely different grounds, or if both States claimed as judgment creditors on the basis of completely different debts incurred while Hughes was still alive. 13 In Western Union Telegraph Co. n. Pennsylvania, 368 U. S. 71, the Court held, by contrast, that a holder of tangible property is denied due process by a state-court judgment of escheat that does not and cannot protect the holder from the escheat claim of another State, and that the proper procedure was for the competing States to invoke the original jurisdiction of this Court. Because the Court held that the States could not constitutionally enforce their escheat laws in their own courts, this Court was the only remaining forum in which a State could escheat property that other States claimed. The situation in which the present case arises is quite different, since there is no constitutional impediment to both California and Texas imposing death taxes upon the Hughes estate by proceedings in their own courts. CALIFORNIA v. TEXAS 613 601 Stewart, J., concurring In the latter situation the question of domicile would be irrelevant, and there is no compelling reason why it should have been the dispositive question in Texas v. Florida. For when this Court exercises its original jurisdiction to settle a dispute between two States it does not look to the law of each State, but rather creates its own rules of decision. “The determination of the relative rights of contending States in respect of the use of streams flowing through them does not depend upon the same considerations and is not governed by the same rules of law that are applied in such States for the solution of similar questions of private right.” Connecticut v. Massachusetts, 282 U. S. 660, 670. The determination of the relative rights of two States that both claim the power to tax a decedent’s estate similarly should not necessarily depend on the same considerations that would govern the question under state law. In deciding the controversy between Texas and California the Court could, of course, determine, according to its own rules of decision where Hughes was domiciled when he died, and permit only the State of domicile to tax the estate. Cf. Texas v. New Jersey, 379 U. S. 674. But assuming there are sufficient contacts with each State to support a finding of domicile under each State’s law—a premise of jurisdiction in Texas v. Florida—the Court could with equal validity decide that the proper disposition was a division of the assets of the estate based on a judgment as to the relative strength of the domicile claims, or on almost any other basis that seemed just. Indeed, for purposes of this Court’s resolution of a dispute between two sovereign States, each of which has an equally valid claim under its own law, it would seem more appropriate to decide the case on some neutral principle rather than attempt to determine a single “correct” answer Under state common law. In any event the question for decision would be one to be resolved under federal law, not under the state law of domicile. 614 OCTOBER TERM, 1977 Stewart, J., concurring 437U.S. A prior adjudication of domicile in the courts of either of the claiming States would not bind this Court in any respect, or prevent it from affording whatever relief it deemed appropriate. Thus California, unlike the ordinary claimant in an interpleader action, will not be met with the bar of res judicata if its potential conflict with Texas is not pre-empted at this incipient stage. Cf. Tremies v. Sunshine Mining Co., 308 U. S. 66, 74-78. The original jurisdiction of this Court exists to remedy real and substantial injuries inflicted by sovereign States upon their sister States. New York v. New Jersey, 256 U. S. 296, 309; Massachusetts v. Missouri, 308 U. S. 1. As yet, California has suffered no injury at the hand of Texas, and there is indeed a “fair probability” that the injury will never come to pass. California has not obtained a judgment in its own courts that Hughes died domiciled there, but merely a conditional agreement from the estate’s representative not to contest California’s assertion of domicile in this Court if the present complaint is accepted for filing. Moreover, whether or not the estate will in fact be insufficient to meet the various tax claims may depend on how the assets are finally evaluated and what deductions the various taxing authorities allow. While the risk of conflict poses a sufficiently real threat to the estate to present a ripe controversy if an interpleader suit were filed by the appropriate parties in a federal district court,14 that risk certainly does not amount to “clear and convincing evidence” of an actual injury of “serious magnitude” inflicted by one State upon another. New York n. New Jersey, supra, at 309; Missouri v. Illinois, 200 U. S. 496, 521. Indeed it is not at all clear to me that the injury threatened here—essentially that one State will be left with an uncollectible judgment because another State has exhausted a debtor’s funds—would be sufficient to justify the exercise of this Court’s original jurisdiction even if the injury actually 14 See nn. 9 and 10, supra. CALIFORNIA v. TEXAS 615 601 Powell, J., concurring occurred.15 But even assuming that it would be, such jurisdiction surely does not exist until each State has finally established an enforceable claim under state law, and it is clear that the estate’s assets are insufficient to meet both claims. It is for these reasons that I join in the order of the Court denying California’s motion for leave to file its complaint. Mr. Justice Powell, concurring. I join the excellent opinion of Mr. Justice Stewart and write simply to emphasize his conclusion that, in light of Edelman v. Jordan, 415 U. S. 651 (1974), this Court’s decision in Worcester County Trust Co. v. Riley, 302 U. S. 292 (1937), no longer can be regarded as a bar against the use of federal interpleader by estates threatened with double death taxation because of possible inconsistent adjudications of domicile. As Professor Zechariah Chafee, the father of federal statu 15 The injury would be the same whatever the source of each State’s claim upon the debtor. The closest analogue of the State’s complaint would seem to be the petition for a declaration of involuntary bankruptcy—a remedy created entirely by statute, not by “accepted doctrines of the common law or equity systems of jurisprudence, which are guides to decision of cases within the original jurisdiction of this Court.” Texas v. Florida, 306 U. S., at 405. See generally 1 W. Collier on Bankruptcy, fl 0.01-0.03 (1974). I am not certain that our duty to “exercise [the] jurisdiction which is given,” Cohens n. Virginia, 6 Wheat. 264, 404, compels or even empowers us to create such a remedy for the sovereign States. The status of unsatisfied creditor does not necessarily create the kind of controversy between States that can or should be resolved by means of adjudication under this Court’s original jurisdiction. This may, rather, be the kind of dispute that is best resolved by the contending States through negotiation or arbitration. See New York v. New Jersey, 256 U. S. 296, 313; Texas v. Florida, supra, at 428 (Frankfurter, J., dissenting). Tweed & Sargent, supra n. 12, at 77. Indeed many States have adopted procedures for arbitration or compromise of precisely the kind of dispute presented here. See Uniform Interstate Arbitration of Death Taxes Act, 8 U. L. A. 255 (1972); 4 CCH Inh. Est. & Gift Tax Rep. J 12,035 (1975). 616 OCTOBER TERM, 1977 Powell, J., concurring 437U.S. tory interpleader, pointed out: “It is our federal system which creates the possibility of double taxation. Somewhere within that federal system we should be able to find remedies for the frictions which that system creates.” Federal Interpleader Since the Act of 1936, 49 Yale L. J. 377, 388 (1940). The Worcester County Court, much to Professor Chafee’s regret, 49 Yale L. J., at 388, held that the Eleventh Amendment precluded resort to federal interpleader as a remedy for the particularly unfair “friction” that can result from conflicting adjudications of domicile in death taxation cases. But as noted by Mr. Justice Stewart, ante, at 608-609, n. 10, Worcester County has been effectively undercut by subsequent developments. Edelman made it clear that the Eleventh Amendment bars only suits “by private parties seeking to impose a liability which must be paid from public funds in the state treasury,” 415 U. S., at 663, and not actions which may have “fiscal consequences to state treasuries . . . [that are] the necessary result of compliance with decrees which by their terms [are] prospective in nature,” id., at 667-668, at least in a case such as this, where the very controversy is a result of our federal system. An interpleader action to prevent competing States’ taxing officials from levying death taxes on the basis of possible inconsistent adjudications of domicile unquestionably would fall into the latter category. Accordingly, it would appear that resort to federal interpleader no longer is proscribed by the Eleventh Amendment in this situation. PHILADELPHIA v. NEW JERSEY 617 Syllabus CITY OF PHILADELPHIA et al. v. NEW JERSEY et al. APPEAL FROM THE SUPREME COURT OF NEW JERSEY No. 77-404. Argued March 27, 1978—Decided June 23, 1978 New Jersey statute (ch. 363) that prohibits the importation of most “solid or liquid waste which originated or was collected outside the territorial limits of the State . . held to violate the Commerce Clause of the United States Constitution. Pp. 621-629. (a) All objects of interstate trade merit Commerce Clause protection and none is excluded from the definition of “commerce” at the outset; hence, contrary to the suggestion of the court below, there can be no doubt that the banning of “valueless” out-of-state wastes by ch. 363 implicates constitutional protection. Bowman v. Chicago & Northwestern R. Co., 125 U. S. 465, distinguished. Pp. 621-623. (b) The crucial inquiry here must be directed to determining whether ch. 363 is basically an economic protectionist measure, and thus virtually per se invalid, or a law directed at legitimate local concerns that has only incidental effects on interstate commerce. Pike v. Bruce Church, Inc., 397 U. S. 137, 142. Pp. 623-624. (c) Since the evil of protectionism can reside in legislative means as well as legislative ends, it is immaterial whether the legislative purpose of ch. 363 is to protect New Jersey’s environment or its economy, for whatever the purpose, it may not be accomplished by discriminating against articles of commerce coming from outside the State unless there is some reason, apart from their origin, to treat them differently. Both on its face and in its plain effect ch. 363 violates this principle of nondiscrimination. A State may not attempt to isolate itself from a problem common to many by erecting a barrier against the movement of interstate trade, as ch. 363 seeks to do by imposing on out-of-state commercial interests the full burden of conserving New Jersey’s remaining landfill space. Pp. 625-628. (d) The New Jersey statute cannot be likened to a quarantine law which bans importation of articles of commerce because of their innate harmfulness and not because of their origin. Though New Jersey concedes that out-of-state waste is no different from domestic waste, it has banned the former while leaving its landfill sites open to the latter, thus trying to saddle those outside the State with the entire burden of slowing the flow of wastes into New Jersey’s remaining landfill sites. Pp. 628-629. 73 N. J. 562, 376 A. 2d 888, reversed. 618 OCTOBER TERM, 1977 Opinion of the Court 437 U. S. Stewart, J., delivered the opinion of the Court, in which Brennan, White, Marshall, Blackmun, Powell, and Stevens, JJ., joined. Rehnquist, J., filed a dissenting opinion, in which Burger, C. J., joined, post, p. 629. Herbert F. Moore argued the cause for appellants. With him on the briefs was Arthur Meisel. Stephen Skillman, Assistant Attorney General of New Jersey, argued the cause for appellees. With him on the brief were John J. Degnan, Attorney General, and Deborah Poritz and Nathan Edelstein, Deputy Attorneys General.* Mr. Justice Stewart delivered the opinion of the Court. A New Jersey law prohibits the importation of most “solid or liquid waste which originated or was collected outside the territorial limits of the State . . . .” In this case we are required to decide whether this statutory prohibition violates the Commerce Clause of the United States Constitution. I The statutory provision in question is ch. 363 of 1973 N. J. Laws, which took effect in early 1974. In pertinent part it provides: “No person shall bring into this State any solid or liquid waste which originated or was collected outside the territorial limits of the State, except garbage to be fed to swine in the State of New Jersey, until the commissioner [of the State Department of Environmental Protection] shall determine that such action can be permitted without endangering the public health, safety and *M. Jefferson Davis and Michael J. Hogan filed a brief for the Board of Chosen Freeholders of the County of Burlington, N. J., as amicus curiae urging affirmance. Briefs of amici curiae were filed by Jeffrey B. Schwartz for the American Public Health Assn.; and by William C. Brashares for the National Solid Wastes Management Assn. PHILADELPHIA v. NEW JERSEY 619 617 Opinion of the Court welfare and has promulgated regulations permitting and regulating the treatment and disposal of such waste in this State.” N. J. Stat. Ann. § 13:17-10 (West Supp. 1978).1 As authorized by ch. 363, the Commissioner promulgated regulations permitting four categories of waste to enter the State.2 Apart from these narrow exceptions, however, New Jersey closed its borders to all waste from other States. Immediately affected by these developments were the operators of private landfills in New Jersey, and several cities in other States that had agreements with these operators for waste disposal. They brought suit against New Jersey and its Department of Environmental Protection in state court, attacking the statute and regulations on a number of state and federal grounds. In an oral opinion granting the plaintiffs’ motion for summary judgment, the trial court declared the law unconstitutional because it discriminated against interstate commerce. The New Jersey Supreme Court consolidated this case with another reaching the same conclusion, 1 New Jersey enacted a Waste Control Act, N. J. Stat. Ann. § 13: 1Z—1 et seq. (West Supp. 1978), in early 1973. This Act empowered the State Commissioner of Environmental Protection to promulgate rules banning the movement of solid waste into the State. Within a year, the state legislature enacted ch. 363, which reversed the presumption and blocked the importation of all categories of waste unless excepted by rules of the Commissioner. 2 Effective as of February 1974, these regulations provided as follows: “ (a) No person shall bring into this State, or accept for disposal in this State, any solid or liquid waste which originated or was collected outside the territorial limits of this State. This Section shall not apply to: “1. Garbage to be fed to swine in the State of New Jersey; “2. Any separated waste material, including newsprint, paper, glass and metals, that is free from putrescible materials and not mixed with other solid or liquid waste that is intended for a recycling or reclamation facility; “3. Municipal solid waste to be separated or processed into usable secondary materials, including fuel and heat, at a resource recovery facility 620 OCTOBER TERM, 1977 Opinion of the Court 437 U. S. Hackensack Meadowlands Development Comm’n v. Municipal Sanitary Landfill Auth., 127 N. J. Super. 160, 316 A. 2d 711, and reversed, 68 N. J. 451, 348 A. 2d 505. It found that ch. 363 advanced vital health and environmental objectives with no economic discrimination against, and with little burden upon, interstate commerce, and that the law was therefore permissible under the Commerce Clause of the Constitution. The court also found no congressional intent to pre-empt ch. 363 by enacting in 1965 the Solid Waste Disposal Act, 79 Stat. 997, 42 U. S. C. § 3251 et seq., as amended by the Resource Recovery Act of 1970, 84 Stat. 1227. The plaintiffs then appealed to this Court.3 After noting probable jurisdiction, 425 U. S. 910, and hearing oral argument, we remanded for reconsideration of the appellants’ preemption claim in light of the newly enacted Resource Conservation and Recovery Act of 1976, 90 Stat. 2795. 430 U. S. 141. Again the New Jersey Supreme Court found no federal pre-emption of the state law, 73 N. J. 562, 376 A. 2d 888, and again we noted probable jurisdiction, 434 U. S. 964. We agree with the New Jersey court that the state law has not been pre-empted by federal legislation.4 The dispositive provided that not less than 70 per cent of the thru-put of any such facility is to be separated or processed into usable secondary materials; and “4. Pesticides, hazardous waste, chemical waste, bulk liquid, bulk semiliquid, which is to be treated, processed or recovered in a solid waste disposal facility which is registered with the Department for such treatment, processing or recovery, other than by disposal on or in the lands of this State.” N. J. Admin. Code 7:1-4.2 (Supp. 1977). 3 The decision of the New Jersey Supreme Court disposed of the appellants’ pre-emption and Commerce Clause claims, but remanded the case to the trial court for further proceedings on the other claims. The appellants then dismissed with prejudice the other counts in their complaint so that there would be a final judgment from which they could appeal to this Court. 4 The surviving provisions of the 1965 Solid Waste Disposal Act, 79 Stat. 997, the Resource Discovery Act of 1970, 84 Stat. 1227, and the Resource PHILADELPHIA v. NEW JERSEY 621 617 Opinion of the Court question, therefore, is whether the law is constitutionally permissible in light of the Commerce Clause of the Constitution.5 II Before it addressed the merits of the appellants’ claim, the New Jersey Supreme Court questioned whether the interstate movement of those wastes banned by ch. 363 is “commerce” at all within the meaning of the Commerce Clause. Any doubts on that score should be laid to rest at the outset. The state court expressed the view that there may be two definitions of “commerce” for constitutional purposes. When relied on “to support some exertion of federal control or regulation,” the Commerce Clause permits “a very sweeping concept” of commerce. 68 N. J., at 469, 348 A. 2d, at 514. But when relied on “to strike down or restrict state legislation,” that Clause and the term “commerce” have a “much more confined . . . reach.” Ibid. The state court reached this conclusion in an attempt to Conservation and Recovery Act of 1976, 90 Stat. 2795, are now codified as the Solid Waste Disposal Act, found at 42 U. S. C. § 6901 et seq. (1976 ed.). From our review of this federal legislation, we find no “clear and manifest, purpose of Congress,” Rice v. Santa Fe Elevator Corp., 331 U. S. 218, 230, to pre-empt the entire field of interstate waste management or transportation, either by express statutory command, see Jones v. Rath Packing Co., 430 U. S. 519, 530-531, or by implicit legislative design, see City of Burbank v. Lockheed Air Terminal, 411 U. S. 624, 633. To the contrary, Congress expressly has provided that “the collection and disposal of solid wastes should continue to be primarily the function of State, regional, and local agencies . . . .” 42 U. S. C. §6901 (a)(4) (1976 ed.). Similarly, ch. 363 is not pre-empted because of a square conflict with particular provisions of federal law or because of general incompatibility with basic federal objectives. See Ray v. Atlantic Richfield Co., 435 U. S. 151, 158; Jones v. Rath Packing Co., supra, at 540-541. In short, we agree with the New Jersey Supreme Court that ch. 363 can be enforced consistently with the program goals and the respective federal-state roles intended by Congress when it enacted the federal legislation. 6 U. S. Const., Art. I, § 8, cl. 3. 622 OCTOBER TERM, 1977 Opinion of the Court 437U.S. reconcile modern Commerce Clause concepts with several old cases of this Court holding that States can prohibit the importation of some objects because they “are not legitimate subjects of trade and commerce.” Bowman v. Chicago & Northwestern R. Co., 125 U. S. 465, 489. These articles include items “which, on account of their existing condition, would bring in and spread disease, pestilence, and death, such as rags or other substances infected with the germs of yellow fever or the virus of small-pox, or cattle or meat or other provisions that are diseased or decayed, or otherwise, from their condition and quality, unfit for human use or consumption.” Ibid. See also Baldwin v. G. A. F. Seelig, Inc., 294 U. S. 511, 525, and cases cited therein. The state court found that ch. 363 as narrowed by the state regulations, see n. 2, supra, banned only “those wastes which can [not] be put to effective use,” and therefore those wastes were not commerce at all, unless “the mere transportation and disposal of valueless waste between states constitutes interstate commerce within the meaning of the constitutional provision.” 68 N. J., at 468, 348 A. 2d, at 514. We think the state court misread our cases, and thus erred in assuming that they require a two-tiered definition of commerce. In saying that innately harmful articles “are not legitimate subjects of trade and commerce,” the Bowman Court was stating its conclusion, not the starting point of its reasoning. All objects of interstate trade merit Commerce Clause protection; none is excluded by definition at the outset. In Bowman and similar cases, the Court held simply that because the articles’ worth in interstate commerce was far outweighed by the dangers inhering in their very movement, States could prohibit their transportation across state lines. Hence, we reject the state court’s suggestion that the banning of “valueless” out-of-state wastes by ch. 363 implicates no constitutional protection. Just as Congress has power to regulate the interstate movement of these wastes, States are PHILADELPHIA v. NEW JERSEY 623 617 Opinion of the Court not free from constitutional scrutiny when they restrict that movement. Cf. Hughes N. Alexandria Scrap Corp., 426 U. S. 794, 802-814; Meat Drivers n. United States, 371 U. S. 94. Ill A Although the Constitution gives Congress the power to regulate commerce among the States, many subjects of potential federal regulation under that power inevitably escape congressional attention “because of their local character and their number and diversity.” South Carolina State Highway Dept. v. Barnwell Bros., Inc., 303 U. S. 177, 185. In the absence of federal legislation, these subjects are open to control by the States so long as they act within the restraints imposed by the Commerce Clause itself. See Raymond Motor Transportation, Inc. v. Rice, 434 U. S. 429, 440. The bounds of these restraints appear nowhere in the words of the Commerce Clause, but have emerged gradually in the decisions of this Court giving effect to its basic purpose. That broad purpose was well expressed by Mr. Justice Jackson in his opinion for the Court in H. P. Hood Ac Sons, Inc. v. Du Mond, 336 U. S. 525, 537-538: “This principle that our economic unit is the Nation, which alone has the gamut of powers necessary to control of the economy, including the vital power of erecting customs barriers against foreign competition, has as its corollary that the states are not separable economic units. As the Court said in Baldwin v. Seelig, 294 U. S. [511], 527, ‘what is ultimate is the principle that one state in its dealings with another may not place itself in a position of economic isolation.’ ” The opinions of the Court through the years have reflected an alertness to the evils of “economic isolation” and protectionism, while at the same time recognizing that incidental 624 OCTOBER TERM, 1977 Opinion of the Court 437U.S. burdens on interstate commerce may be unavoidable when a State legislates to safeguard the health and safety of its people. Thus, where simple economic protectionism is effected by state legislation, a virtually per se rule of invalidity has been erected. See, e. g., H. P. Hood & Sons, Inc., v. Du Mond, supra; Toomer v. Witsell, 334 U. S. 385, 403-406; Baldwin v. G. A. F. Seelig, Inc., supra; Buck v. Kuykendall, 267 U. S. 307, 315-316. The clearest example of such legislation is a law that overtly blocks the flow of interstate commerce at a State’s borders. Cf. Welton v. Missouri, 91 U. S. 275. But where other legislative objectives are credibly advanced and there is no patent discrimination against interstate trade, the Court has adopted a much more flexible approach, the general contours of which were outlined in Pike v. Bruce Church, Inc., 397 U. S. 137, 142: “Where the statute regulates evenhandedly to effectuate a legitimate local public interest, and its effects on interstate commerce are only incidental, it will be upheld unless the burden imposed on such commerce is clearly excessive in relation to the putative local benefits. . . . If a legitimate local purpose is found, then the question becomes one of degree. And the extent of the burden that will be tolerated will of course depend on the nature of the local interest involved, and on whether it could be promoted as well with a lesser impact on interstate activities.” See also Raymond Motor Transportation, Inc. v. Rice, supra, at 441-442; Hunt v. Washington Apple Advertising Comm’n, 432 U. S. 333, 352-354; Great A&P Tea Co. v. Cottrell, 424 U. S. 366, 371-372. The crucial inquiry, therefore, must be directed to determining whether ch. 363 is basically a protectionist measure, or whether it can fairly be viewed as a law directed to legitimate local concerns, with effects upon interstate commerce that are only incidental. PHILADELPHIA v. NEW JERSEY 625 617 Opinion of the Court B The purpose of ch. 363 is set out in the statute itself as follows: “The Legislature finds and determines that . . . the volume of solid and liquid waste continues to rapidly increase, that the treatment and disposal of these wastes continues to pose an even greater threat to the quality of the environment of New Jersey, that the available and appropriate land fill sites within the State are being diminished, that the environment continues to be threatened by the treatment and disposal of waste which originated or was collected outside the State, and that the public health, safety and welfare require that the treatment and disposal within this State of all wastes generated outside of the State be prohibited.” The New Jersey Supreme Court accepted this statement of the state legislature’s purpose. The state court additionally found that New Jersey’s existing landfill sites will be exhausted within a few years; that to go on using these sites or to develop new ones will take a heavy environmental toll, both from pollution and from loss of scarce open lands; that new techniques to divert waste from landfills to other methods of disposal and resource recovery processes are under development, but that these changes will require time; and finally, that “the extension of the lifespan of existing landfills, resulting from the exclusion of out-of-state waste, may be of crucial importance in preventing further virgin wetlands or other undeveloped lands from being devoted to landfill purposes.” 68 N. J., at 460-465, 348 A. 2d, at 509-512. Based on these findings, the court concluded that ch. 363 was designed to protect, not the State’s economy, but its environment, and that its substantial benefits outweigh its “slight” burden on interstate commerce. Id., at 471-478, 348 A. 2d, at 515—519. The appellants strenuously contend that ch. 363, “while outwardly cloaked ‘in the currently fashionable garb of environ 626 OCTOBER TERM, 1977 Opinion of the Court 437U.S. mental protection/ ... is actually no more than a legislative effort to suppress competition and stabilize the cost of solid waste disposal for New Jersey residents . . . .” They cite passages of legislative history suggesting that the problem addressed by ch. 363 is primarily financial: Stemming the flow of out-of-state waste into certain landfill sites will extend their lives, thus delaying the day when New Jersey cities must transport their waste to more distant and expensive sites. The appellees, on the other hand, deny that ch. 363 was motivated by financial concerns or economic protectionism. In the words of their brief, “[n]o New Jersey commercial interests stand to gain advantage over competitors from outside the state as a result of the ban on dumping out-of-state waste.” Noting that New Jersey landfill operators are among the plaintiffs, the appellee’s brief argues that “[t]he complaint is not that New Jersey has forged an economic preference for its own commercial interests, but rather that it has denied a small group of its entrepreneurs an economic opportunity to traffic in waste in order to protect the health, safety and welfare of the citizenry at large.” This dispute about ultimate legislative purpose need not be resolved, because its resolution would not be relevant to the constitutional issue to be decided in this case. Contrary to the evident assumption of the state court and the parties, the evil of protectionism can reside in legislative means as well as legislative ends. Thus, it does not matter whether the ultimate aim of ch. 363 is to reduce the waste disposal costs of New Jersey residents or to save remaining open lands from pollution, for we assume New Jersey has every right to protect its residents’ pocketbooks as well as their environment. And it may be assumed as well that New Jersey may pursue those ends by slowing the flow of all waste into the State’s remaining landfills, even though interstate commerce may incidentally be affected. But whatever New Jersey’s ultimate purpose, it may not be accomplished by discriminating against PHILADELPHIA v. NEW JERSEY 627 617 Opinion of the Court articles of commerce coming from outside the State unless there is some reason, apart from their origin, to treat them differently. Both on its face and in its plain effect, ch. 363 violates this principle of nondiscrimination. The Court has consistently found parochial legislation of this kind to be constitutionally invalid, whether the ultimate aim of the legislation was to assure a steady supply of milk by erecting barriers to allegedly ruinous outside competition, Baldwin v. G. A. F. Seelig, Inc., 294 U. S., at 522-524; or to create jobs by keeping industry within the State, FosterFountain Packing Co. v. Hay del, 278 U. S. 1, 10; Johnson v. Hay del, 278 U. S. 16; Toomer v. Witsell, 334 U. S., at 403-404; or to preserve the State’s financial resources from depletion by fencing out indigent immigrants, Edwards v. California, 314 U. S. 160, 173—174. In each of these cases, a presumably legitimate goal was sought to be achieved by the illegitimate means of isolating the State from the national economy. Also relevant here are the Court’s decisions holding that a State may not accord its own inhabitants' a preferred right of access over consumers in other States to natural resources located within its borders. West v. Kansas Natural Gas Co., 221 U. S. 229; Pennsylvania v. West Virginia, 262 U. S. 553. These cases stand for the basic principle that a “State is without power to prevent privately owned articles of trade from being shipped and sold in interstate commerce on the ground that they are required to satisfy local demands or because they are needed by the people of the State.”6 FosterFountain Packing Co. v. Haydel, supra, at 10. 6 We express no opinion about New Jersey’s power, consistent with the Commerce Clause, to restrict to state residents access to state-owned resources, compare Douglas v. Seacoast Products, Inc., 431 U. S. 265, 283-287, with id., at 287-290 (Rehnquist, J., concurring and dissenting); Toomer v. Witsell, 334 U. S. 385, 404; or New Jersey’s power* to spend state funds solely on behalf of state residents and businesses, compere Hughes v. Alexandria Scrap Corp., 426 U. S. 794, 805-810; id., at 815 628 OCTOBER TERM, 1977 Opinion of the Court 437U.S. The New Jersey law at issue in this case falls squarely within the area that the Commerce Clause puts off limits to state regulation. On its face, it imposes on out-of-state commercial interests the full burden of conserving the State’s remaining landfill space. It is true that in our previous cases the scarce natural resource was itself the article of commerce, whereas here the scarce resource and the article of commerce are distinct. But that difference is without consequence. In both instances, the State has overtly moved to slow or freeze the flow of commerce for protectionist reasons. It does not matter that the State has shut the article of commerce inside the State in one case and outside the State in the other. What is crucial is the attempt by one State to isolate itself from a problem common to many by erecting a barrier against the movement of interstate trade. The appellees argue that not all laws which facially discriminate against out-of-state commerce are forbidden protectionist regulations. In particular, they point to quarantine laws, which this Court has repeatedly upheld even though they appear to single out interstate commerce for special treatment. See Baldwin v. G. A. F. Seelig, Inc., supra, at 525; Bowman v. Chicago & Northwestern R. Co., 125 U. S., at 489. In the appellees’ view, ch. 363 is analogous to such health-protective measures, since it reduces the exposure of New Jersey residents to the allegedly harmful effects of landfill sites. It is true that certain quarantine laws have not been considered forbidden protectionist measures, even though they were directed against out-of-state commerce. See Asbell v. Kansas, 209 U. S. 251; Reid v. Colorado, 187 U. S. 137; Bowman v. Chicago & Northwestern R. Co., supra, at 489. But those quarantine laws banned the importation of articles such as diseased livestock that required destruction as soon (Stevens, J., concurring), with id., at 817 (Brennan, J., dissenting). Also compare South Carolina State Highway Dept. v. Barnwell Bros., Inc., 303 U. S. 177, 187, with Southern Pacific Co. v. Arizona ex rel. Sullivan, 325 U. S. 761,783. PHILADELPHIA v. NEW JERSEY 629 617 Rehnquist, J., dissenting as possible because their very movement risked contagion and other evils. Those laws thus did not discriminate against interstate commerce as such, but simply prevented traffic in noxious articles, whatever their origin. The New Jersey statute is not such a quarantine law. There has been no claim here that the very movement of waste into or through New Jersey endangers health, or that waste must be disposed of as soon and as close to its point of generation as possible. The harms caused by waste are said to arise after its disposal in landfill sites, and at that point, as New Jersey concedes, there is no basis to distinguish out-of-state waste from domestic waste. If one is inherently harmful, so is the other. Yet New Jersey has banned the former while leaving its landfill sites open to the latter. The New Jersey law blocks the importation of waste in an obvious effort to saddle those outside the State with the entire burden of slowing the flow of refuse into New Jersey’s remaining landfill sites. That legislative effort is clearly impermissible under the Commerce Clause of the Constitution. Today, cities in Pennsylvania and New York find it expedient or necessary to send their waste into New Jersey for disposal, and New Jersey claims the right to close its borders to such traffic. Tomorrow, cities in New Jersey may find it expedient or necessary to send their waste into Pennsylvania or New York for disposal, and those States might then claim the right to close their borders. The Commerce Clause will protect New Jersey in the future, just as it protects her neighbors now, from efforts by one State to isolate itself in the stream of interstate commerce from a problem shared by all. The judgment is Reversed. Mr. Justice Rehnquist, with whom The Chief Justice joins, dissenting. A growing problem in our Nation is the sanitary treatment and disposal of solid waste.1 For many years, solid waste was 1 Congress specifically recognized the substantial dangers to the environ 630 OCTOBER TERM, 1977 Rehnquist, J., dissenting 437U.S. incinerated. Because of the significant environmental problems attendant on incineration, however, this method of solid waste disposal has declined in use in many localities, including New Jersey. “Sanitary” landfills have replaced incineration as the principal method of disposing of solid waste. In ch. 363 of the 1973 N. J. Laws, the State of New Jersey legislatively recognized the unfortunate fact that landfills also present extremely serious health and safety problems. First, in New Jersey, “virtually all sanitary landfills can be expected to produce leachate, a noxious and highly polluted liquid which is seldom visible and frequently pollutes . . . ground and surface waters.” App. 149. The natural decomposition process which occurs in landfills also produces large quantities of methane and thereby presents a significant explosion hazard. Id., at 149, 156-157. Landfills can also generate “health hazards caused by rodents, fires and scavenger birds”* and, “needless to say, do not help New Jersey’s aesthetic appearance nor New Jersey’s noise or water or air pollution problems.” Supp. App. 5. The health and safety hazards associated with landfills present appellees with a currently unsolvable dilemma. Other, hopefully safer, methods of disposing of solid wastes are still in the development stage and cannot presently be used. But appellees obviously cannot completely stop the tide of solid waste that its citizens will produce in the interim. For the moment, therefore, appellees must continue to use sanitary landfills to dispose of New Jersey’s own solid waste despite the critical environmental problems thereby created. ment and public health that are posed by current methods of disposing of solid waste in the Resource Conservation and Recovery Act of 1976, 90 Stat. 2795. As the Court recognizes, ante, at 621 n. 4, the laws under challenge here “can be enforced consistently with the program goals and the respective federal-state roles intended by Congress when it enacted” this and other legislation and are thus not pre-empted by any federal statutes. PHILADELPHIA v. NEW JERSEY 631 617 Rehnquist, J., dissenting The question presented in this case is whether New Jersey must also continue to receive and dispose of solid waste from neighboring States, even though these will inexorably increase the health problems discussed above.2 The Court answers this question in the affirmative. New Jersey must either prohibit all landfill operations, leaving itself to cast about for a presently nonexistent solution to the serious problem of disposing of the waste generated within its own borders, or it must accept waste from every portion of the United States, thereby multiplying the health and safety problems which would result if it dealt only with such wastes generated within the State. Because past precedents establish that the Commerce Clause does not present appellees with such a Hobson’s choice, I dissent. The Court recognizes, ante, at 621-622, that States can prohibit the importation of items “ ‘which, on account of their existing condition, would bring in and spread disease, pestilence, and death, such as rags or other substances infected with the germs of yellow fever or the virus of small-pox, or cattle or meat or other provisions that are diseased or decayed, or otherwise, from their condition and quality, unfit for human use or consumption.’ ” Bowman v. Chicago & Northwestern R. Co., 125 U. S. 465, 489 (1888). See Baldwin v. G. A. F. Seelig, Inc., 294 U. S. 511, 525 (1935); Sligh v. Kirkwood, 237 U. S. 52, 59-60 (Í915); Asbell v. Kansas, 209 U. S. 251 (1908); Railroad Co. v. Husen, 95 U. S. 465, 472 (1878). As the Court points out, such “quarantine laws have not been considered forbidden protectionist measures, even though they were directed against out-oj-state commerce.” Ante, at 628 (emphasis added). 2 Regulations of the New Jersey Department of Environmental Protection “except from the ban on out-of-state refuse those types of solid waste which may have a value for recycling or for use as fuel.” App. 47. Thus, the ban under challenge would appear to be strictly limited to that waste which will be disposed of in sanitary landfills and thereby pose health and safety dangers to the citizens of New Jersey. 632 OCTOBER TERM, 1977 Rehnquist, J., dissenting 437U.S. In my opinion, these cases are dispositive of the present one. Under them, New Jersey may require germ-infected rags or diseased meat to be disposed of as best as possible within the State, but at the same time prohibit the importation of such items for disposal at the facilities that are set up within New Jersey for disposal of such material generated within the State. The physical fact of life that New Jersey must somehow dispose of its own noxious items does not mean that it must serve as a depository for those of every other State. Similarly, New Jersey should be free under our past precedents to prohibit the importation of solid waste because of the health and safety problems that such waste poses to its citizens. The fact that New Jersey continues to, and indeed must continue to, dispose of its own solid waste does not mean that New Jersey may not prohibit the importation of even more solid waste into the State. I simply see no way to distinguish solid waste, on the record of this case, from germ-infected rags, diseased meat, and other noxious items. The Court’s effort to distinguish these prior cases is unconvincing. It first asserts that the quarantine laws which have previously been upheld “banned the importation of articles such as diseased livestock that required destruction as soon as possible because their very movement risked contagion and other evils.” Ante, at 628-629. According to the Court, the New Jersey law is distinguishable from these other laws, and invalid, because the concern of New Jersey is not with the movement of solid waste but with the present inability to safely dispose of it once it reaches its destination. But I think it far from clear that the State’s law has as limited a focus as the Court imputes to it: Solid waste which is a health hazard when it reaches its destination may in all likelihood be an equally great health hazard in transit. Even if the Court is correct in its characterization of New Jersey’s concerns, I do not see why a State may ban the importation of items whose movement risks contagion, but PHILADELPHIA v. NEW JERSEY 633 617 Rehnquist, J., dissenting cannot ban the importation of items which, although they may be transported into the State without undue hazard, will then simply pile up in an ever increasing danger to the public’s health and safety. The Commerce Clause was not drawn with a view to having the validity of state laws turn on such pointless distinctions. Second, the Court implies that the challenged laws must be invalidated because New Jersey has left its landfills open to domestic waste. But, as the Court notes, ante, at 628, this Court has repeatedly upheld quarantine laws “even though they appear to single out interstate commerce for special treatment.” The fact that New Jersey has left its landfill sites open for domestic waste does not, of course, mean that solid waste is not innately harmful. Nor does it mean that New Jersey prohibits importation of solid waste for reasons other than the health and safety of its population. New Jersey must out of sheer necessity treat and dispose of its solid waste in some fashion, just as it must treat New Jersey cattle suffering from hoof-and-mouth disease. It does not follow that New Jersey must, under the Commerce Clause, accept solid waste or diseased cattle from outside its borders and thereby exacerbate its problems. The Supreme Court of New Jersey expressly found that ch. 363 was passed “to preserve the health of New Jersey residents by keeping their exposure to solid waste and landfill areas to a minimum.” 68 N. J. 451, 473, 348 A. 2d 505, 516. The Court points to absolutely no evidence that would contradict this finding by the New Jersey Supreme Court. Because I find no basis for distinguishing the laws under challenge here from our past cases upholding state laws that prohibit the importation of items that could endanger the population of the State, I dissent. 634 OCTOBER TERM, 1977 Syllabus 437 U. S. UNITED STATES v. JOHN et al. CERTIORARI TO THE UNITED STATES COURT OF APPEALS FOR THE FIFTH CIRCUIT No. 77-836. Argued April 19, 1978—Decided June 23, 1978* Lands designated as a reservation for Choctaw Indians residing in central Mississippi held, on the basis of the history of the relations between the Mississippi Choctaws and the United States, to be “Indian country,” as defined in 18 U. S. C. § 1151 (1976 ed.) to include “all land within the limits of any Indian reservation under the jurisdiction of the United States Government,” and as used in the Major Crimes Act, 18 U. S. C. § 1153, which makes any Indian who commits certain specified offenses “within the Indian country . . . subject to the same laws and penalties as all other persons committing [such] offenses, within the exclusive jurisdiction of the United States.” Neither the fact that the Choctaws in Mississippi are merely a remnant of a larger group of Indians, nor the fact that federal supervision over them has not been continuous, affects the federal power to deal with them under these statutes. Hence, the Major Crimes Act provided a proper basis for federal prosecution of a Choctaw Indian for assault with intent to kill (one of the specified offenses) occurring on such lands, and Mississippi had no power similarly to prosecute him for the same offense. Pp. 638-654. No. 77-836, 560 F. 2d 1202, reversed and remanded; No. 77-575, 347 So. 2d 959, reversed. Blackmun, J., delivered the opinion for a unanimous Court. H. Bartow Farr III argued the cause for the United States in No. 77-836. With him on the brief were Solicitor General McCree, Assistant Attorney General Moorman, Raymond N. Zagone, Carl Strass, and Larry G. Gutterridge. Richard B. Collins argued the cause for appellants in No. 77-575 and respondents in No. 77-836. With him on the briefs was Edwin R. Smith. *Together with No. 77-575, John et al. v. Mississippi, on appeal from the Supreme Court of Mississippi. UNITED STATES v. JOHN 635 634 Opinion of the Court Carl F. Andre argued the cause for appellee in No. 77-575. With him on the brief were A. F. Summer, Attorney General of Mississippi, and Catherine Walker Underwood, Special Assistant Attorney General. + Mr. Justice Blackmun delivered the opinion of the Court. These cases present issues concerning state and federal jurisdiction over certain crimes committed on lands within the area designated as a reservation for the Choctaw Indians residing in central Mississippi. More precisely, the questions presented are whether the lands are “Indian country,” as that phrase is defined in 18 U. S. C. § 1151 (1976 ed.) and as it was used in the Major Crimes Act of 1885, being § 9 of the Act of Mar. 3, 1885, 23 Stat. 385, later codified as 18 U. S. C. § 1153, and, if so, whether these federal statutes operate to preclude the exercise of state criminal jurisdiction over the offenses. I In October 1975, in the Southern District of Mississippi, Smith John1 was indicted by a federal grand jury for assault with intent to kill Artis Jenkins, in violation of 18 U. S. C. §§ 1153 and 113 (a).2 He was tried before a jury and, on \Harry R. Sachse filed a brief for the Mississippi Band of Choctaw Indians as amicus curiae urging reversal in both cases. Arthur Lazarus, Jr., filed a brief for the Association of Indian Affairs, Inc., as amicus curiae urging reversal in No. 77-836. 1 Smith John’s son, Harry Smith John, also was charged jointly with his father in the federal indictment. The United States and counsel for the Johns have advised the Court of Harry Smith John’s death on February 18, 1978, and concede that as to him the case is moot. Brief for United States 3: Brief for John et al. 1. The brief for the State of Mississippi is silent as to this. We agree that both cases are moot as to Harry Smith John. 2 At the time of the alleged offense, 18 U. S. C. § 1153 read: “Any Indian who commits against the person or property of another Indian or other person any of the following offenses, namely, murder, 636 OCTOBER TERM, 1977 Opinion of the Court 437 U. S. December 15, was convicted of the lesser included offense of simple assault.3 A sentence of 90 days in a local jail-type institution and a fine of $300 were imposed. On appeal, the United States Court of Appeals for the Fifth Circuit, considering the issue on its own motion, see App. to Pet. for Cert, in manslaughter, rape, carnal knowledge of any female, not his wife, who has not attained the age of sixteen years, assault with intent to commit rape, incest, assault with intent to kill, assault with a dangerous weapon, assault resulting in serious bodily injury, arson, burglary, robbery, and larceny within the Indian country, shall be subject to the same laws and penalties as all other persons committing any of the above offenses, within the exclusive jurisdiction of the United States. “As used, in this section, the offenses of rape and assault with intent to commit rape shall be defined in accordance with the laws of the State in which the offense was committed, and any Indian who commits the offenses of rape or assault with intent to commit rape upon any female Indian within the Indian country shall be imprisoned at the discretion of the court. “As used in this section, the offenses of burglary, assault with a dangerous weapon, assault resulting in serious bodily injury, and incest shall be defined and punished in accordance with the laws of the State in which such offense was committed.” This section has since been amended by the Indian Crimes Act of 1976, 90 Stat. 585, which added kidnaping to the list of offenses covered and made changes, not pertinent to these cases, in the ways in which state law is incorporated. Section 113, the statute specifying punishment for assaults committed within the special territorial jurisdiction of the United States, including those for which federal prosecutions are authorized by § 1153, was also amended by the same Act. See H. R. Rep. No. 94^1038 (1976); S. Rep. No. 94-620 (1976). 3 Under Keeble v. United States, 412 U. S. 205 (1973), Smith John was entitled to instructions regarding this lesser included offense. It appears, however, see Brief for John et al. 5; Brief for United States 4, and n. 6, that Smith John argued before the Court of Appeals that although he was entitled to such instructions, the District Court had no jurisdiction to enter a judgment of conviction for the lesser offense, a misdemeanor not listed in § 1153. The Court of Appeals, in deciding that the statute did not apply even to the extent urged by the United States, did not reach the issue. It has not been argued before this Court. See, however, Felicia v. United States, 495 F. 2d 353 (CA8), cert, denied, 419 U. S. 849 (1974). UNITED STATES v. JOHN 637 634 Opinion of the Court No. 77-836, p. 39A, ruled that the District Court was without jurisdiction over the case because the lands designated as a reservation for the Choctaw Indians residing in Mississippi, and on which the offense took place, were not “Indian country,” and that, therefore, § 1153 did not provide a basis for federal prosecution. 560 F. 2d 1202, 1205-1206 (1977). The United States sought review, and we granted its petition for certiorari in No. 77-836. 434 U. S. 1032 (1978). In April 1976, Smith John4 was indicted by a grand jury of Leake County, Miss., for aggravated assault upon the same Artis Jenkins, in violation of Miss. Code Ann. § 97-3-7 (2) (Supp. 1977). The incident that was the subject of the state indictment was the same as that to which the federal indictment related. A motion to dismiss the charge on the ground the federal jurisdiction was exclusive was denied. John was tried before a jury in the Circuit Court of Leake County and, in May 1976, was convicted of the offense charged. He was sentenced to two years in the state penitentiary. On appeal, the Supreme Court of Mississippi, relying on its earlier decision in Tubby v. State, 327 So. 2d 272 (1976), and on the decision of the United States Court of Appeals for the Fifth Circuit in United States v. State Tax Comm’n, 505 F. 2d 633 (1974), rehearing denied, 535 F. 2d 300, rehearing en banc denied, 541 F. 2d 469 (1976), held that the United States District Court had had no jurisdiction to prosecute Smith John, and that, therefore, his arguments against state-court jurisdiction were without merit. 347 So. 2d 959 (1977). Characterizing the case as one falling within this Court’s jurisdiction under 28 U. S. C. § 1257 (2) (1976 ed.), Smith John filed notice of an appeal in No. 77-575. We 4 Harry Smith John was also jointly charged with his father under the Mississippi indictment, and was convicted. As stated above, counsel for Harry Smith John concedes that the death of Harry Smith John on February 18, 1978, renders the state case moot as to him. Brief for John et al. 1. 638 437 U. S. OCTOBER TERM, 1977 Opinion of the Court postponed jurisdiction, 434 U. S. 1032 (1978). We now note jurisdiction. Antoine v. Washington, 420 U. S. 194 (1975) ; McClanahan v. Arizona State Tax Comm’n, 411 U. S. 164 (1973). II There is no dispute that Smith John is a Choctaw Indian, and it is presumed by all that he is a descendant of the Choctaws who for hundreds of years made their homes in what is now central Mississippi. The story of these Indians, and of their brethren who left Mississippi to settle in what is now the State of Oklahoma, has been told in the pages of the reports of this Court and of other federal courts. See, e. g., Choctaw Nation v. Oklahoma, 397 U. S. 620 (1970); Winton v. Amos, 255 U. S. 373 (1921); Fleming v. McCurtain, 215 U. S. 56 (1909); United States v. Choctaw Nation, 179 U. S. 494 (1900); Choctaw Nation v. United States, 119 U. S. 1 (1886) ; Chitto v. United States, 133 Ct. Cl. 643, 138 F. Supp. 253, cert, denied, 352 U. S. 841 (1956); Choctaw Nation v. United States, 81 Ct. Cl. 1, cert, denied, 296 U. S. 643 (1935). At the time of the Revolutionary War, these Indians occupied large areas of what is now the State of Mississippi. In the years just after the formation of our country, they entered into a treaty of friendship with the United States. Treaty at Hopewell, 7 Stat. 21 (1786). But the United States became anxious to secure the lands the Indians occupied in order to allow for westward expansion. The Choctaws, in an attempt to avoid what proved to be their fate, entered into a series of treaties gradually relinquishing their claims to these lands.5 5 Treaty at Fort Adams, 7 Stat. 66 (1801) (2^ million acres ceded); Treaty at Fort Confederation, 7 Stat. 73 (1802) (establishment of bound-aries generally); Treaty at Hoe-Buckin-too-pa, 7 Stat. 80 (1803) (900,000 acres in conformity with the Fort Confederation agreement) ; Treaty at Mount Dexter, 7 Stat. 98 (1805) (4 million acres) ; Treaty at Fort St. Stephens, 7 Stat. 152 (1816) (ceding a relatively sma.11 tract where UNITED STATES v. JOHN 639 634 Opinion of the Court Despite these concessions, when Mississippi became a State on December 10, 1817, the Choctaws still retained claims, recognized by the Federal Government, to more than three-quarters of the land within the State’s boundaries. The popular pressure to make these lands available to non-Indian settlement, and the responsibility for these Indians felt by some in the Government, combined to shape a federal policy aimed at persuading the Choctaws to give up their lands in Mississippi completely and to remove to new lands in what for many years was known as the Indian Territory, now a part of Oklahoma and Arkansas. The first attempt to effectuate this' policy, the Treaty at Doak’s Stand, 7 Stat. 210 (1820), resulted in an exchange of more than 5 million acres. Because, however, of complications arising when it was discovered that much of the land promised the Indians already had been settled, most Choctaws remained in Mississippi. A delegation of Choctaws went to Washington, D. C., to untangle the situation and to negotiate yet another treaty. See 7 Stat. 234 (1825). Still, few Choctaws moved. Only after the election of Andrew Jackson to the Presidency in 1828 did the federal efforts to persuade the Choctaws to leave Mississippi meet with some success.6 Even before Columbus, Miss., now stands). See A. DeRosier, Jr., The Removal of the Choctaw Indians 29 (1970). 6 Andrew Jackson had been one of the two commissioners sent to negotiate the Treaty at Doak’s Stand. From the land ceded by the Choctaws under that treaty, a new state capital, to be named Jackson, was planned. P. Fortune, The Formative Period, in 1 A History of Mississippi 255 (R. McLemore ed., 1973). Jackson’s position with regard to the removal of the Indians played a significant role in his Presidential election and in his popularity in Mississippi. Id., at 277. See generally DeRosier, supra n. 5, at 100-115; M. Young, Redskins, Rufileshirts, and Rednecks: Indian Allotments in Alabama and Mississippi, 1830-1860, pp. 14-21 (1961); G. Foreman, Indian Removal: The Emigration of the Five Civilized Tribes of Indians 21 (1953 ed.); F. Cohen, Handbook of Federal Indian Law 56-59 (1941); Prucha, Andrew Jackson’s Indian Policy: A Reassessment, 56 J. of Am. Hist. 527 (1969). 640 437 U. S. OCTOBER TERM, 1977 Opinion of the Court Jackson himself had acted on behalf of the Federal Government, however, the State of Mississippi, grown impatient with federal policies, had taken steps to assert jurisdiction over the lands occupied by the Choctaws. In early 1829, legislation was enacted purporting to extend legal process into the Choctaw territory. 1824r-1838 Miss. Gen. Laws 195 (Act of Feb. 4, 1829). In his first annual address to Congress on December 8, 1829, President Jackson made known his position on the Indian question and his support of immediate removal. S. Doc. No. 1, 21st Cong., 1st Sess., 15-16 (1829). Further encouraged, the Mississippi Legislature passed an Act purporting to abolish the Choctaw government and to impose a fine upon anyone assuming the role of chief. The Act also declared that the rights of white persons living within the State were to be enjoyed by the Indians, and that the laws of the State were to be in effect throughout the territory they occupied. 1824-1838 Miss. Gen. Laws 207 (Act of Jan. 19, 1830). In Washington, Congress debated whether the States had power to assert such jurisdiction and whether such assertions were wise.7 But the only message heard by the Choctaws in Mississippi was that the Federal Government no longer would stand between the States and the Indians. Appreciating these realities, the Choctaws again agreed to deal with the Federal Government. On September 27, 1830, the Treaty at Dancing 7 See, e. g., 6 Cong. Deb. 585 (1830). These debates culminated on May 28, 1830, in the passage of the Indian Removal Bill. 4 Stat. 411. See generally A. Abel, The History of Events Resulting in Indian Consolidation West of the Mississippi River, in 1906 Annual Report of the American Historical Assn. 377-382 (1908). They also set the stage for the constitutional crisis surrounding this Court’s decision in Worcester v. Georgia, 6 Pet. 515 (1832), that the States had no power over the Indians and the Indian lands within their boundaries. See generally Burke, The Cherokee Cases: A Study in Law, Politics, and Morality, 21 Stan. L. Rev. 500 (1969); Miles, After John Marshall’s Decision: Worcester v. Georgia and the Nullification Crisis, 39 J. of So. Hist. 519 (1973). UNITED STATES v. JOHN 641 634 Opinion of the Court Rabbit Creek, 7 Stat. 333, was signed.8 It provided that the Choctaws would cede to the United States all lands still occupied by them east of the Mississippi, more than 10 million acres. They were to remove to lands west of the river, where they would remain perpetually free of federal or state control, by the fall of 1833. The Government would help plan and pay for this move. Each Choctaw “head of a family being desirous to remain and become a citizen of the States,” id., at 335, however, was to be permitted to do so by signifying his intention within six months to the federal agent assigned to the area. Lands were to be reserved, at least 640 acres per household, to be held by the Indians in fee simple if they would remain upon the lands for five years. Ibid. Other lands were reserved to the various chiefs and to others already residing on improved lands. Id., at 335-336. Those who remained, however, were not to “lose the priviledge of a Choctaw citizen,” id., at 335, although they were to receive no share of the annuity provided for those who chose to remove. The relations between the Federal Government and the Choctaws remaining in Mississippi did not end with the formal ratification of the Treaty at Dancing Rabbit Creek by the United States Senate in February 1831. 7 Cong. Deb. 347 (1831). The account of the federal attempts to satisfy 8 Perhaps the best evidence of the circumstances surrounding this treaty lies in its very words. As signed by the Choctaws, it contained the following preamble: “Whereas the General Assembly of the State of Mississippi has extended the laws of said State to persons and property within the chartered limits of the [Choctaw lands], and the President of the United States has said that he cannot protect the Choctaw people from the operation of these laws; Now therefore that the Choctaw may Eve under their own laws in peace with the United States and the State of Mississippi they have determined to seU their lands east of the Mississippi and have accordingly agreed to the following articles of treaty.” The preamble was stricken from the treaty as ratified by the Senate. 7 Cong. Deb. 346-347 (1831). 642 437 U. S. OCTOBER TERM, 1977 Opinion of the Court the obligations of the United States both to those who remained,9 and to those who removed,10 is one best left to historians. It is enough to say here that the failure of these 9 See generally Chitto v. United States, 133 Ct. Cl. 643, 138 F. Supp. 253, cert, denied, 352 U. S. 841 (1956); Young, supra n. 6, at 47-72; Riley, Choctaw Land Claims, 8 Publications of the Mississippi Historical Society 345 (1904). It is generally acknowledged that, whether anxious to conceal the fact that far more Choctaws had remained in Mississippi than he had anticipated originally, or simply because he was disinterested in his job and generally dissolute, the agent in charge of the task refused to record the claims of those who elected to remain. See, e. g., Coleman v. Doe, 12 Miss. 40 (1844); Chitto v. United States, 133 Ct. Cl., at 648-649, 138 F. Supp., at 257. Speculators soon began pressing the cause of those who had been refused. Perhaps in large part due to their efforts, and the cloud created on the ceded lands as they were put up for sale without the proper recordation of Indian claims, Congress soon authorized investigation of the situation. See 7 American State Papers, Public Lands 448-525 (1860); H. R. Rep. No. 663, 24th Cong., 1st Sess. (1836). Although one might wonder whether it was concern for the preservation of the claims for the Indians, or simply concern for the preservation of the claims, that motivated subsequent events, measures were taken to remedy the situation and to provide substitute lands for the Choctaws to replace those lands sold despite their attempt to file claims One measure provided that the claimants would be issued scrip enabling them to claim substitute lands, but half the scrip was not to be delivered unless the claimants removed to territory west of the Mississippi. Act of Aug. 23, 1842, 5 Stat. 513. The administration of this statute was as unsuccessful as had been the administration of the original treaty. It appears that in practice, none of the scrip was delivered before removal, Chitto v. United States, 133 Ct. Cl., at 649, 138 F. Supp., at 257, and that Congress later established a fund to be paid in lieu of part of the scrip. 5 Stat. 777 (1845). After an attempt at settlement in 1852 proved unsuccessful, the United States and the Choctaws in Oklahoma in 1855 entered into still another treaty that provided that the Senate would make a determination of the amounts owing to the Choctaws generally for the failure of the United States to abide by its various treaty promises. Treaty of June 22, 1855, 11 Stat. 611. In March 1859, the Senate approved the general formula under which [Footnote 10 is on p. 0^3] UNITED STATES v. JOHN 643 634 Opinion of the Court attempts, characterized by incompetence, if not corruption, proved an embarrassment and an intractable problem for the Federal Government for at least a century. See, e. g., Chitto v. United States, 133 Ct. Cl. 643, 138 F. Supp. 253 (1956). It remained federal policy, however, to try to induce these Indians to leave Mississippi. During the 1890’s, the Federal Government became acutely aware of the fact that not all the Choctaws had left Mississippi. At that time federal policy toward the Indians favored the allotment of tribal holdings, including the Choctaw holdings in the Indian Territory, in order to make way for Oklahoma’s statehood. The inclusion of the Choctaws then residing in Mississippi in the distribution of these holdings proved among the largest obstacles encountered during the allotment effort.11 But even during this era, when federal policy again those amounts were to be calculated, Cong. Globe, 35th Cong., 2d Sess., 1691; S. Rep. No. 374, 35th Cong., 2d Sess. (1859), and the Secretary of the Interior, pursuant to this direction, computed the total to be almost $3 million. See H. R. Exec. Doc. No. 82, 36th Cong., 1st Sess. (1860), reprinted in H. R. Rep. No. 251, 45th Cong., 2d Sess., 12 (1878). The War Between the States interrupted the payment of this Senate award, and, after the war, the Choctaws found themselves forced to prove their claims once again, this time in the federal courts. See Choctaw Nation v. United States, 119 U. S. 1 (1886), rev’g 21 Ct. Cl. 59. 10 See generally DeRosier, supra n. 5, at 129-167; Wright, The Removal of the Choctaws to the Indian Territory 1830-1833, 6 Chronicles of Oklahoma 103 (1928); A. Debo, The Rise and Fall of the Choctaw Republic 56 (2d ed. 1961); n. 9, supra. 11 The potential right of the Choctaws who had not removed to participate in any general allotment of the Oklahoma lands was acknowledged in the treaty entered into by the United States and the Choctaws and Chickasaws at the close of the war. 14 Stat. 774 (1866). But a new series of frauds and speculation made implementation of this policy difficult when the allotment eventually took place. See the essentially contemporaneous account of these events provided in Wade, The Removal of the 397 (1904). In response to a flood of claims of those purporting to be Mississippi Choctaws to whom a portion of its holdings in Oklahoma should 644 OCTOBER TERM, 1977 Opinion of the Court 437 U. S. supported the removal of the Mississippi Choctaws to join their brethren in the West, there was no doubt that there remained persons in Mississippi who were properly regarded both by the Congress and by the Executive Branch as Indians. It was not until 1916 that this federal recognition of the presence of Indians in Mississippi was manifested by other than attempts to secure their removal. The appropriations for the Bureau of Indian Affairs in that year included an item (for $1,000) to enable the Secretary of the Interior “to investigate the condition of the Indians living in Mississippi” and to report to Congress “as to their need for additional land and school facilities.” 39 Stat. 138. See H. R. Doc. No. 1464, 64th Cong., 2d Sess. (1916). In March 1917, hearings were held in Union, Miss., by the House Committee on Investigation of the Indian Service, again exploring the desirability of providing federal services for these Indians. The efforts resulted in an inclusion in the general appropriation for the Bureau of Indian Affairs in 1918. This appropriation, passed only after debate in the House, 56 Cong. Rec. 1136-1140 (1918), included funds for the establishment of an agency with a physician, for the maintenance of schools, and for the purchase of land and farm equipment.12 Lands purchased be distributed, the Choctaw Nation resisted attempts to include Mississippi Choctaws on its rolls. Between 1897 and 1907, when the Choctaw rolls were finally closed, repeated efforts were made by the Dawes Commission, and by Congress, to determine the appropriate criteria for enrollment of the Mississippi Choctaws, and their participation in the allotment. Again, any participation was conditioned on removal from Mississippi. See the complete account of these efforts in Estate of Winton v. Amos, 51 Ct. Cl. 284 (1916), rev’d in part and aff’d in part, 255 U. S. 373 (1921). 12 40 Stat. 573 (1918). See Hearings on Indian Appropriation Bill before a Subcommittee of the House Committee on Indian Affairs, 65th Cong., 2d Sess., 153, 175-176 (1918). Shortly after this appropriation was made, Cato Sells, Commissioner of Indian Affairs, traveled to Mississippi to gain firsthand information about the Indians there. In his annual report, he observed: “Practically all of the Mississippi Choctaws are full-bloods. Very few UNITED STATES v. JOHN 645 634 Opinion of the Court through these appropriations were to be sold on contract to individuals in keeping with the general pattern of providing lands eventually to be held in fee by individual Indians, rather than held collectively. Further provisions for the Choctaws in Mississippi were made in similar appropriations in later years.13 In the 1930’s, the federal Indian policy had shifted back toward the preservation of Indian communities generally. This shift led to the enactment of the Indian Reorganization Act of 1934, 48 Stat. 984, and the discontinuance of the allotment program. The Choctaws in Mississippi were among the many groups who, before the legislation was enacted, voted to support its passage. This vote was reported to Congress by the Bureau of Indian Affairs. See Hearings on S. 2755 and S. 3645 before the Senate Committee on Indian Affairs, 73d Cong., 2d Sess., pt. 2, p. 82 (1934); Hearings on H. R. 7902 before the House Committee on Indian Affairs, 73d Cong., 2d Sess., 423 (1934). On March 30, 1935, the Mississippi Choctaws voted, as anticipated by § 18 of the Act, 48 Stat. 988, 25 U. S. C. § 478 (1976 ed.), to accept the provisions of the own their homes. They are almost entirely farm laborers or share croppers. They are industrious, honest, and necessarily frugal. Most of them barely exist, and some suffer from want of the necessaries of life and medical aid. In many of the homes visited by me there was conspicuous evidence of pitiable poverty. I discovered families with from three to five children, of proper age, not one of whom had spent a day of their fife in school. With very few exceptions they indicated willingness to go to school, as did their parents to send them. Several young Choctaw boys and girls expressed an ardent desire for an education.” Report of the Commissioner of Indian Affairs, in 2 Reports of the Department of the Interior, 1918, pp. 79-80 (1919). 13 41 Stat. 15 (1919); 41 Stat. 420 (1920); 41 Stat. 1236 (1921); 42 Stat. 570 (1922); 42 Stat. 1191 (1923); 43 Stat. 409 (1924); 43 Stat. 1149, 1155, 1159 (1925); 44 Stat. 461, 468, 472 (1926); 44 Stat. 941, 947, 951 (1927); 45 Stat. 206, 216, 220 (1928); 45 Stat. 1568, 1578, 1581 (1929); 46 Stat. 286, 299 (1930); 46 Stat. 1121, 1135 (1931); 47 Stat. 109 (1932). 646 OCTOBER TERM, 1977 Opinion of the Court 437U.S. Act. T. Haas, Ten Years of Tribal Government Under I. R. A. 17 (U. S. Indian Service, Tribal Relations Pamphlet No. 1 (1947)). By this time, it had become obvious that the original method of land purchase authorized by the 1918 appropriations—by contract to a particular Indian purchaser—not only was inconsistent with the new federal policy of encouraging the preservation of Indian communities with commonly held lands, but also was not providing the Mississippi Choctaws with the benefits intended. See H. R. Rep. No. 194, 76th Cong., 1st Sess. (1939). In 1939, Congress passed an Act providing essentially that title to all the lands previously purchased for the Mississippi Choctaws would be “in the United States in trust for such Choctaw Indians of one-half or more Indian blood, resident in Mississippi, as shall be designated by the Secretary of the Interior.” Ch. 235, 53 Stat. 851. In December 1944, the Assistant Secretary of the Department of the Interior officially proclaimed all the lands then purchased in aid of the Choctaws in Mississippi, totaling at that time more than 15,000 acres, to be a reservation. 9 Fed. Reg. 14907.14 In April 1945, again as anticipated by the Indian Reorganization Act, § 16, 48 Stat. 987, 25 U. S. C. § 476 (1976 ed.), the Mississippi Band of Choctaw Indians adopted a constitution and bylaws; these were duly approved by the appropriate federal authorities in May 1945.15 14 By its language, the 1939 Act affected only those lands that were “not under contract for resale to Choctaw Indians, or on which existing contracts of resale may hereafter be canceled.” The 1944 Proclamation of Reservation recited specifically that it was issued “by virtue of the authority contained in the act of June 21,1939, and in section 7 of the act of June 18, 1934,” and that no such acquired lands were covered by any outstanding contract “for the resale of any part thereof to any Choctaw or other Indian.” 15 This constitution has since been amended in response to the Indian Civil Rights Act of 1968, 25 U. S. C. § 1301 et seq. (1976 ed.). UNITED STATES v. JOHN 647 634 Opinion of the Court With this historical sketch as background, we turn to the jurisdictional issues presented by Smith John’s case. Ill In order to determine whether there is federal jurisdiction over the offense with which Smith John was charged (alleged in the federal indictment to have been committed “on and within the Choctaw Indian Reservation and on land within the Indian country under the jurisdiction of the United States of America”), we first look to the terms of the statute upon which the United States relies, that is, the Major Crimes Act, 18 U. S. C. § 1153. This Act, as codified at the time of the alleged offense, provided: “Any Indian who commits . . . assault with intent to kill . . . within the Indian country, shall be subject to the same laws and penalties as all other persons committing any [such offense], within the exclusive jurisdiction of the United States.” The definition of “Indian country” as used here and elsewhere in chapter 53 of Title 18 is provided in § 1151.16 Both the Mississippi Supreme Court 16 As originally enacted, the Major Crimes Act made no reference to “Indian country” but, instead, referred to any “reservation” within the States and the Territories. See n. 22, infra. The legislation retained this general form when it was re-enacted as § 328 of the Criminal Code of 1909, 35 Stat. 1151 (codified from 1926 to 1948 as 18 U. S. C. §548), and amended, 47 Stat. 336 (1932) (adding incest to the list of crimes covered, deleting the reference to the Territories, and providing expressly that rights of way running through a reservation were to be included as part of the reservation). In the 1948 revision of Title 18, however, the express reference to “reservation” was deleted in favor of the use of the term “Indian country,” which was used in most of the other special statutes referring to Indians, and as defined in § 1151. See Reviser’s Note, and n. 18, infra. The Act has since been amended four times, 63 Stat. 94 (1949) (relating to the punishment for the crime of rape); 80 Stat. 1100 (1966) (adding carnal knowledge and assault with intent to rape); 82 Stat. 80 (1968) (adding assault resulting in serious bodily injury); 90 Stat. 585 (1976) (see n. 2, supra), but its form has not been changed substantially. 648 437 U.S. OCTOBER TERM, 1977 Opinion of the Court and the Court of Appeals concluded that the situs of the alleged offense did not constitute “Indian country,” and that therefore § 1153 did not afford a basis for the prosecution of Smith John in federal court. We do not agree. With certain exceptions not pertinent here, § 1151 includes within the term “Indian country” three categories of land. The first, with which we are here concerned,17 is “all land within the limits of any Indian reservation under the jurisdiction of the United States Government, notwithstanding the issuance of any patent.” This language first appeared in the Code in 1948 as a part of the general revision of Title 18. The Reviser’s Notes indicate that this definition was based on several decisions of this Court interpreting the term as it was used in various criminal statutes relating to Indians. In one of these cases, United States v. McGowan, 302 U. S. 535 (1938), the Court held that the Reno Indian Colony, consisting of 28.38 acres within the State of Nevada, purchased out of federal funds appropriated in 1917 and 1926 and occupied by several hundred Indians theretofore scattered throughout Nevada, was “Indian country” for the purposes of what was then 25 U. S. C. § 247. (the predecessor of 18 U. S. C. §3618 (1976 ed.)), providing for the forfeiture of a vehicle used to transport intoxicants into the Indian country. The Court noted that the “fundamental consideration of both Congress and the Department of the Interior in establishing this colony has been the protection of a dependent people.” 302 U. S., at 538. The principal test applied was drawn from 17 The second category for inclusion within the definition of “Indian country” is “all dependent Indian communities within the borders of the United States whether within the original or subsequently acquired territory thereof, and whether within or without the limits of a State.” The third category is “all Indian allotments, the Indian titles to which have not been extinguished, including rights-of-way running through the same.” Inasmuch as we find in the first category a sufficient basis for the exercise of federal jurisdiction in this case, we need not consider the second and third categories. UNITED STATES v. JOHN 649 634 Opinion of the Court an earlier case, United States v. Pelican, 232 U. S. 442 (1914), and was whether the land in question “had been validly set apart for the use of the Indians as such, under the superintendence of the Government.” Id., at 449; 302 U. S., at 539.18 The Mississippi lands in question here were declared by Congress to be held in trust by the Federal Government for the benefit of the Mississippi Choctaw Indians who were at that time under federal supervision. There is no apparent reason why these lands, which had been purchased in previous years for the aid of those Indians, did not become a “reservation,” at least for the purposes of federal criminal jurisdiction at that particular time. See United States n. Celestine, 215 U. S. 278, 285 (1909). But if there were any doubt about the matter in 1939 when, as hereinabove described, Congress declared that title to lands previously purchased for the Mississippi Choctaws would be held in trust, the situation was completely clarified by the proclamation in 1944 of a reservation and the subsequent approval of the constitution and bylaws adopted by the Mississippi Band. The Court of Appeals and the Mississippi Supreme Court held, and the State now argues, that the 1944 proclamation had no effect because the Indian Reorganization Act of 1934 was not intended to apply to the Mississippi Choctaws. Assuming for the moment that authority for the proclamation 18 Some earlier cases had suggested a more technical and limited definition of “Indian country.” See, e. g., Bates v. Clark, 95 U. S. 204 (1877). Throughout most of the 19th century, apparently the only statutory definition was that in § 1 of the Act of June 30, 1834, 4 Stat. 729. But this definition was dropped in the compilation of the Revised Statutes. See Ex parte Crow Dog, 109 U. S. 556 (1883). This Court was left with little choice but to continue to apply the principles established under the earlier statutory language and to develop them according to changing conditions. Seex e. g., Donnelly v. United States, 228 U. S. 243 (1913). It is the more expansive scope of the term that was incorporated in the 1948 revision of Title 18. 650 437 U. S. OCTOBER TERM, 1977 Opinion of the Court can be found only in the 1934 Act, we find this argument unpersuasive. The 1934 Act defined “Indians” not only as “all persons of Indian descent who are members of any recognized [in 1934] tribe now under Federal jurisdiction,” and their descendants who then were residing on any Indian reservation, but also as “all other persons of one-half or more Indian blood.” 48 Stat. 988, 25 U. S. C. § 479 (1976 ed.). There is no doubt that persons of this description lived in Mississippi, and were recognized as such by Congress and by the Department of the Interior, at the time the Act was passed.19 The references to the Mississippi Choctaws in the legislative history of the Act, see supra, at 645-646, confirm our view that the Mississippi Choctaws were not to be excepted from the general operation of the 1934 Act.20 19 A report completed just after the passage of the Act recounts: “After all the years of living in and among both white and colored race, it is indeed surprising to find that approximately 85 percent of this group are full bloods. Their racial integrity is intact in spite of the absence of permanent holdings or any sort of community life. Many of the older Choctaws do not speak English.” E. Groves, Notes on the Choctaw Indians, Feb. 20-Mar. 20, 1936, p. 1 (Bureau of Indian Affairs). 20 The State of Mississippi makes much of a sentence contained in an unpublished memorandum dated August 31, 1936, of the Solicitor for the Department of the Interior. It reads: “They [the Indians remaining in Mississippi] cannot now be regarded as a tribe.” See F. Cohen, Handbook of Federal Indian Law 273 (1941). A reading of the entire memorandum, however, convinces us that it supports the position of the United States in this case. The memorandum was concerned only with the proper description of the Indians in the deeds relating to lands purchased according to the provisions of the Indian Reorganization Act. At least one deed had been prepared designating the grantee as “the United States in trust for the Choctaw tribe of Mississippi.” The memorandum recommended that, because the Indians could not be regarded as a tribe at that time, the deeds be written designating the grantee as “[t]he United States in trust for such Choctaw Indians of one-half or more Indian blood, resident in Mississippi, as shall be designated by the Secretary of the Interior, until such time as the Choctaw Indians of Mississippi shall be organized as an Indian tribe pursuant to the act of June 18, 1934 (48 Stat. 984) [the UNITED STATES v. JOHN 651 634 Opinion of the Court IV Mississippi appears to concede, Brief for Appellee in No. 77-575, p. 44, that if § 1153 provides a basis for the prosecution of Smith John for the offense charged, the State has no similar jurisdiction. This concession, based on the assumption that §1153 ordinarily is pre-emptive of state jurisdiction when it applies, seems to us to be correct.21 It was a necessary premise of at least one of our earlier decisions. Seymour n. Superintendent, 368 U. S. 351 (1962). See also Williams v. Lee, 358 U. S. 217,220, and n. 5 (1959); Rice v. Olson, 324 U. S. 786 (1945); In re Carmen’s Petition, 165 F. Supp. 942 (ND Cal. 1958), aff’d sub nom. Dickson v. Carmen, 270 F. 2d 809 (CA9 1959), cert, denied, 361 U. S. 934 (I960).22 Indian Reorganization Act], and then in trust for such organized tribe.” Surely this is evidence that although there was no legal entity known as “the Choctaw tribe of Mississippi,” the Department of the Interior anticipated that a more formal legal entity, a tribe for the purposes of federal Indian law, soon would exist. 21 We do not consider here the more disputed question whether § 1153 also was intended to pre-empt tribal jurisdiction. See Oliphant v. Suquamish Indian Tribe, 435 U. S. 191, 203-204, n. 14 (1978); United States v. Wheeler, 435 U. S. 313, 325 n. 22 (1978). 22 There is much in the legislative history to support this view. The Major Crimes Act was approved on March 3, 1885, 23 Stat. 385, in part in response to the decision of this Court in Ex parte Crow Dog, 109 U. S. 556 (1883). See United States v. Kagama, 118 U. S. 375, 382-383 (1886). As originally proposed in the House, the bill provided that Indians committing the specified crimes “within any Territory of the United States, and either within or without an Indian reservation, shall be subject therefor to the laws of such Territory relating to said crimes,” and, similarly, that Indians committing the same crimes “within the boundaries of any State of the United States, and either within or without an Indian reservation, shall be subject to the same laws ... as are all other persons committing any of the above crimes within the exclusive jurisdiction of the United States.” 16 Cong. Rec. 934 (1885). It became apparent in conference on the bill that this language would have a far broader effect than originally intended, for the language proposed would “take away from State courts, whether there be a reservation in the 652 OCTOBER TERM, 1977 Opinion of the Court 437 U. S. The State argues, however, that the Federal Government has no power to produce this result. It suggests that since 1830 the Choctaws residing in Mississippi have become fully assimilated into the political and social life of the State, and that the Federal Government long ago abandoned its supervisory authority over these Indians. Because of this abandonment, and the long lapse in the federal recognition of a tribal organization in Mississippi, the power given Congress “[t]o regulate Commerce . . . with the Indian Tribes,” Const. Art. I, § 8, cl. 3, cannot provide a basis for federal jurisdiction. To recognize the Choctaws in Mississippi as Indians over whom special federal power may be exercised would be anomalous and arbitrary.23 We assume for purposes of argument, as does the United States, that there have been times when Mississippi’s jurisdiction over the Choctaws and their lands went unchallenged. But, particularly in view of the elaborate history, recounted above, of relations between the Mississippi Choctaws and the United States, we do not agree that Congress and the Execu State or not” jurisdiction over the listed crimes when committed by an Indian Id., at 2385. The provision was then amended to read “all such Indians committing any of the above crimes . . . within the boundaries of any State of the United States, and within the limits of any Indian reservation,” and was agreed to with this change. 23 Mississippi has made no effort, either in this Court or in the courts below, to support this argument with evidence of the assimilation of the Choctaw Indians in Mississippi, or with a demonstration of the services provided for them. There is evidence that some educational services have been provided by the State. See J. Peterson, The Mississippi Band of Choctaw Indians: Their Recent History and Current Social Relations 84, and passim (Ph. D. dissertation, University of Georgia 1970); J. Jennings, V. Beggs, & A. Caldwell, A Study of the Social and Economic Condition of the Choctaw Indians in Mississippi in Relation to the Educational Program 4 (Bureau of Indian Affairs 1945); T. Taylor, The States and Their Indian Citizens 177 (1972). But the provision of state services to Indians would not prove that the Federal Government had relinquished its ability to provide for these Indians under its Article I power. UNITED STATES v. JOHN 653 634 Opinion of the Court tive Branch have less power to deal with the affairs of the Mississippi Choctaws than with the affairs of other Indian groups. Neither the fact that the Choctaws in Mississippi are merely a remnant of a larger group of Indians, long ago removed from Mississippi, nor the fact that federal supervision over them has not been continuous, destroys the federal power to deal with them. United States v. Wright, 53 F. 2d 300 (CA4 1931), cert, denied, 285 U. S. 539 (1932).24 The State also argues that the Federal Government may not deal specially with the Indians within the State’s boundaries because to do so would be inconsistent with the Treaty at Dancing Rabbit Creek. This argument may seem to be a cruel joke to those familiar with the history of the execution of that treaty, and of the treaties that renegotiated claims arising from it. See supra, at 640—643. And even if that treaty were the only source regarding the status of these Indians in federal law, we see nothing in it inconsistent with the continued federal supervision of them under the Commerce Clause. It is true that this treaty anticipated that each of those electing to remain in Mississippi would become “a citizen of the States,” but the extension of citizenship status to Indans does not, in itself, end the powers given Congress to 24 We need not be concerned, as Mississippi hints, that the assumption of federal criminal jurisdiction over the Choctaw Indians in Mississippi, if not historically anomalous, is inconsistent with the intent of Congress. In the early 195O’s, when federal Indian policy again emphasized assimilation, a thorough survey was made of all the then recognized tribes and their economic and social conditions. These efforts led to a congressional resolution calling for the freedom of certain tribes from federal supervision “at the earliest possible time,” 67 Stat. B 132 (1953), conferring on certain designated States jurisdiction with respect to criminal offenses and civil causes committed or arising on Indian reservations, and granting federal consent to the assertion of state jurisdiction by other States. Id., at 588-590. The Mississippi Band of Choctaw Indians was among those for whom the Bureau of Indian Affairs recommended continued supervision. See H. R. Rep. No. 2680, 83d Cong., 2d Sess., 31-32, and passim (1954). See also H. R. Rep. No. 2503, 82d .Cong., 2d Sess., 313 (1953). 654 OCTOBER TERM, 1977 Opinion of the Court 437U.S. deal with them. See United States v. Celestine, 215 U. S. 278 (1909). V We therefore hold that § 1153 provides a proper basis for federal prosecution of the offense involved here, and that Mississippi has no power similarly to prosecute Smith John for that same offense. Accordingly, the judgment of the Supreme Court of Mississippi in No. 77-575 is reversed; further, the judgment of the United States Court of Appeals for the Fifth Circuit in No. 77-836 is reversed, and that case is remanded for further proceedings consistent with this opinion. It is so ordered. WILL v. COVERT FIRE INS. CO. 655 Syllabus WILL, U. S. DISTRICT JUDGE v. CALVERT FIRE INSURANCE CO. et al. CERTIORARI TO THE UNITED STATES COURT OF APPEALS FOR THE SEVENTH CIRCUIT No. 77-693. Argued April 19, 1978—Decided June 23, 1978 After Calvert Fire Insurance Co. (hereafter respondent) had advised American Mutual Reinsurance Co. (American) that respondent was rescinding its membership in a reinsurance pool that American operated, American sued respondent in an Illinois state court for a declaration that the pool agreement with respondent remained in effect. Six months later, respondent in its answer asserted the unenforceability of the pool agreement on the grounds that American had violated, inter alia, the Securities Act of 1933; Rule 10b-5, promulgated under the Securities Exchange Act of 1934 (hereafter 1934 Act); and the Illinois Securities Act, and counterclaimed for damages on all its defense claims except the one involving Rule 10b-5, which under the 1934 Act’s terms . was exclusively enforceable in the federal courts. Respondent on the same day filed a complaint against American in the Federal District Court for damages for American’s alleged Rule 10b-5 violation, and joined therewith claims based on each of the other defensive counts made in the state-court action. American moved to dismiss or abate the federal-court action, the motion to dismiss being based on the contention that the reinsurance agreement was not a “security” within the meaning of the 1933 or 1934 Act, and the motion to abate being on the ground that the earlier state proceeding included all issues except the one involving Rule 10b-5. Petitioner, the District Court Judge, granted American’s motion to defer the federal proceeding until completion of the state proceeding, except the Rule 10b-5 damages claim. He rejected respondent’s contention that the District Court should proceed with the entire case because of its exclusive jurisdiction over that claim, and noted that the state court was bound to provide the equitable relief sought by respondent by recognizing a valid Rule 10b-5 claim as a defense to the state action. Petitioner heard argument on, but has not yet decided, the question of whether respondent’s interest in the reinsurance pool constituted a “security” as defined in the 1934 Act. After petitioner had rejected motions to reconsider his stay order and refused to certify an interlocutory appeal, respondent petitioned the Court of Appeals for a writ of mandamus directing petitioner to adjudicate 656 OCTOBER TERM, 1977 Syllabus 437 U. S. the Rule 10b-5 claim. Thereafter that court, relying on Colorado River Water Conservation Dist. v. United States, 424 U. S. 800, granted the petition and directed petitioner to “proceed immediately with Calvert’s claim for damages and equitable relief” under the 1934 Act. Held: The judgment is reversed. Pp. 661-667; 667-668. 560 F. 2d 792, reversed. Mr. Justice Rehnquist, joined by Mr. Justice Stewart, Mr. Justice White, and Mr. Justice Stevens, concluded: Issuance of the writ of mandamus by the Court of Appeals imper-missibly interfered with petitioner’s discretion to control his docket Pp. 661-667. (a) Though a court of appeals has the power to issue a writ of mandamus directing a district court to proceed to judgment in a pending case when it is the district court’s duty to do so, the burden is on the moving party to show that its right to issuance of the writ is “clear and indisputable.” P. 662. (b) Where there is duplicative litigation in the state and federal courts, the decision whether or not to defer to the state courts is largely committed to the discretion of the district court, Brillhart v. Excess Ins. Co., 316 U. S. 491, 494, even when matters of federal law are involved, Colorado River, supra, at 820. Pp. 662-664. (c) This case, unlike Colorado River, did not involve outright dismissal of the action, and respondent remained free to urge petitioner to reconsider his decision to defer based on new information as to the progress of the state case; to that extent deferral (contrary to respondent’s argument) was not equivalent to dismissal. Pp. 664-665. (d) Though a district court’s exercise of discretion may be subject to review in a proper interlocutory appeal, it ought not be overridden by a, writ of mandamus. Where a matter is committed to a district court s discretion, it cannot be said that a litigant’s right to a particular result is “clear and indisputable.” Here petitioner has not heedlessly refused to adjudicate the Rule 10b-5 damages claim (the only issue that may not concurrently be resolved by both the state and federal courts), and as far as the record shows his delay in adjudicating that claim is simply the product of a district court’s normal excessive workload, compounded by “the unfortunate consequence of making the judge a litigant” in this mandamus proceeding. Ex parte Fahey, 332 U g 258 260. Pp. 665-667. Mr. Justice Blackmun, who is of the view that Brillhart v. Excess Ins. Co., 316 U. S. 491, a diversity case, has no application to this federal-issue case, concluded that the issuance of mandamus in this case WILL v. CALVERT FIRE INS. CO. 657 655 Opinion of Rehnquist, J. was premature. The judgment of the Court of Appeals must be reversed because the court should have done no more than require reconsideration by petitioner in light of Colorado River Water Conservation Dist. v. United States, 424 U. S. 800, which was decided after petitioner’s stay order. Pp. 667-668. Rehnquist, J., announced the Court’s judgment and delivered an opinion, in which Stewart, White, and Stevens, JJ., joined. Blackmun, J., filed an opinion concurring in the judgment, post, p. 667. Burger, C. J., filed a dissenting opinion, post, p. 668. Brennan, J., filed a dissenting opinion, in which Burger, C. J., and Marshall and Powell, JJ., joined, post, p. 668. Milton V. Freeman argued the cause for petitioner. With him on the briefs were Dennis G. Lyons, Werner J. Kronstein, and Stanley A. Kaplan. Louis Loss argued the cause for respondent Calvert Fire Insurance Co. With him on the brief was Michael L. Weissman. Thomas J. Wcithers and D. Kendall Griffith filed a brief for American Mutual Reinsurance Co., respondent under this Court’s Rule 21 (4), in support- of petitioner. Mr. Justice Rehnquist announced the judgment of the Court, and delivered an opinion in which Mr. Justice Stewart, Mr. Justice White, and Mr. Justice Stevens joined. On August 15, 1977, the Court of Appeals for the Seventh Circuit granted a petition for writ of mandamus ordering petitioner, a judge of the United States District Court for the Northern District of Illinois, “to proceed immediately” to adjudicate a claim based upon the Securities Exchange Act of 1934 and brought by respondent, Calvert Fire Insurance Co., against American Mutual Reinsurance Co., despite the pendency of a substantially identical proceeding between the same parties in the Illinois state courts. 560 F. 2d 792, 797. The Court of Appeals felt that our recent decision in Colorado River Water Conservation Dist. v. United States, 424 U. S. 800 (1976), compelled the issuance of the writ. We granted 658 OCTOBER TERM, 1977 Opinion of Rehnquist, J. 437 u. gt certiorari to consider the propriety of the use of mandamus to review a District Court’s decision to defer to concurrent state proceedings, 434 U. S. 1008, and we now reverse. I Respondent Calvert writes property and casualty insurance. American Mutual operates a reinsurance pool whereby a number of primary insurers protect themselves against unanticipated losses. Membership in the pool requires both the payment of premiums by pool members and indemnification of the pool in the event that losses exceed those upon which the premiums are calculated. Calvert joined the pool in early 1974, but in April of that year notified American Mutual of its election to rescind the agreement by which it became a member. In July 19/4, American Mutual sued in the Circuit Court of Cook County, Ill., to obtain a declaration that the pool agreement between it and Calvert was in full force and effect. Six months later, Calvert in its answer to that suit asserted that the pool agreement was not enforceable against it because of violations by American Mutual of the Securities Act of 1933, the Securities Exchange Act of 1934, the Illinois Securities Act, the Maryland Securities Law, and the state common law of fraud. With its answer Calvert filed a counterclaim seeking $2 million in damages from American Mutual on all of the grounds that it set up in defense except for the defense based on the Securities Exchange Act of 1934. Since § 27 of that Act, 48 Stat. 902, as amended, 15 U. S. C. § 78aa (1976 ed.), granted the district courts of the United States exclusive jurisdiction to enforce the Act, Calvert on the same day filed a complaint in the United States District Court for the Northern District of Illinois seeking damages from American Mutual for an alleged violation of Rule 10b-5, 17 CFR § 240.10b-5 (1977), issued under § 10 (b) of the Act, 15 U. S. C. § 78j (b) (1976 ed.). Joined with this Rule 10b-5 WILL v. CALVERT FIRE INS. CO. 659 655 Opinion of Rehnquist, J. count were claims based on each of the other grounds asserted by it in defense to American Mutual’s state-court action. In February 1975, more than seven months after it had begun its state-court action, but less than one month after Calvert had filed its answer and counterclaim in that action and its complaint in the federal court, American Mutual moved to dismiss or abate the latter. The claim for dismissal was based on the substantive assertion that the reinsurance agreement was not a “security” within the meaning of the 1933 or 1934 Act. The motion to abate was based on the fact that the state proceedings commenced six months before the federal proceedings included every claim and defense except the claim for damages based on Rule 10b-5 under the 1934 Act. In May 1975, Judge Will substantially granted American Mutual’s motion to defer the federal proceeding until the completion of the state proceedings, observing that a tentative trial date had already been set by the state court. Federal litigation of the same issues would therefore be duplicatiye and wasteful. He rejected Calvert’s contention that the court should proceed with the entire case because of its exclusive jurisdiction under the 1934 Act, noting that the state court was bound to provide the equitable relief sought by Calvert by recognizing a valid Rule 10b—5 claim as a defense to the state action.1 Only Calvert’s claim for damages under Rule 10b-5 was subject to the exclusive jurisdiction of the federal court. Petitioner therefore stayed all aspects of Calvert’s federal action subject to the concurrent jurisdiction of both courts, recognizing “only Calvert’s very limited claim for 1 Calvert’s answer in the state action explicitly contended that it was “entitled to rescission of its purchase of the aforesaid security” because of the alleged Rule 10b-5 violation. App. to Pet. for Cert. D-5. It sought identical equitable relief in its federal complaint. Id., at E-6. See Weiner v. Shearson, Hammill & Co., 521 F. 2d 817, 822 (CA9 1975); Aetna State Bank v. Altheimer, 430 F. 2d 750, 754 (CA7 1970). 660 OCTOBER TERM, 1977 Opinion of Rehnquist, J. 437U.S. monetary damages under the 1934 Securities Act as a viable claim in this court.” App. to Pet. for Cert. B-9. On May 9, 1975, Judge Will heard oral argument on the basic question of whether Calvert’s interest in the reinsurance pool is a security within the meaning of the 1934 Act. He has not yet rendered a decision on that issue.2 Judge Will rejected two motions to reconsider his stay order and refused to certify an interlocutory appeal pursuant to 28 U. S. C. § 1292 (b). On May 26, 1976, Calvert petitioned the Court of Appeals for the Seventh Circuit for a writ of mandamus directing Judge Will to proceed to adjudicate its Rule 10b-5 claims.3 Nearly 14 months later, on August 15, 1977, the Court of Appeals granted the petition and directed Judge Will to “proceed immediately with Calvert’s claim for damages and equitable relief under the Securities Exchange Act of 1934.” 560 F. 2d, at 797.4 2 The state court, however, has reached a decision on the issue. The Circuit Court concluded that the agreement was not a security, and therefore struck the federal issues from Calvert’s answer and counterclaim. On an interlocutory appeal the Illinois Appellate Court affirmed, holding that the agreement was not a security within the meaning of either the 1933 or the 1934 Act and that, in any event, § 2 (b) of the McCarran-Ferguson Act, 15 U. S. C. § 1012 (b) (1976 ed.), exempted insurance from the reach of the federal securities laws. American Mutual Reinsurance Co. v. Calvert Fire Ins. Co., 52 Ill. App. 3d 922, 367 N. E. 2d 104 (1977), pet. for leave to appeal denied, No. 50,085 (Jan. 26, 1978), cert, denied, 436 U. S. 906 (1978). 3 As already noted, the stay order did not apply to Calvert’s claim for damages under Rule 10b-5. Judge Will had stayed Calvert’s claim for equitable relief because the state court had jurisdiction to rescind the agreement by recognition of a Rule 10b-5 defense. The petition did not seek to require Judge Will to proceed with the state-law claims or the federal claim based on the 1933 Act. 560 F. 2d 792, 794 n. 2. 4 Although Calvert’s petition addressed only its Rule 10b-5 claims, the court went on to note: “The logic behind our holding in this case supports the conclusion that the stay of 1933 Act claims, as well as the 1934 Act claims, was improper.” 560 F. 2d, at 797 n. 6. WILL v. CALVERT FIRE INS. CO. 661 655 Opinion of Rehnquist, J. We granted certiorari to consider Judge Will’s contention that the issuance of the writ of mandamus impermissibly interfered with the discretion of a district court to control its own docket. 434 U. S. 1008 (1978). II The correct disposition of this case hinges in large part on the appropriate standard of inquiry to be employed by a court of appeals in determining whether to issue a writ of mandamus to a district court. On direct appeal, a court of appeals has broad authority to “modify, vacate, set aside or reverse” an order of a district court, and it may direct such further action on remand “as may be just under the circumstances.” 28 U. S. C. § 2106. By contrast, under the All Writs Act, 28 U. S. C. § 1651 (a), courts of appeals may issue a writ of mandamus only when “necessary or appropriate in aid of their respective jurisdictions.” Whereas a simple showing of error may suffice to obtain a reversal on direct appeal, to issue a writ of mandamus under such circumstances “would undermine the settled limitations upon the power of an appellate court to review interlocutory orders.” Will v. United States, 389 U. S.90, 98 n. 6 (1967). As we have repeatedly reaffirmed in cases such as Kerr v. United States District Court, 426 U. S. 394, 402 (1976), and Bankers Life & Cas. Co. v. Holland, 346 U. S. 379, 382 (1953), the “traditional use of the writ in aid of appellate jurisdiction both at common law and in the federal courts has been to confine an inferior court to a lawful exercise of its prescribed jurisdiction or to compel it to exercise its authority when it is its duty to do so.” Roche v. Evaporated Milk Assn., 319 U. S. 21, 26 (1943). Calvert makes no contention that petitioner has exceeded the bounds of his jurisdiction. Rather, it contends that the District Court, in entering the stay order, has refused “to exercise its authority when it is its duty to do so.” Ibid. There can be no doubt that, where a district 662 437 U.S. OCTOBER TERM, 1977 Opinion of Rehnquist, J. court persistently and without reason refuses to adjudicate a case properly before it, the court of appeals may issue the writ “in order that [it] may exercise the jurisdiction of review given by law.” Insurance Co. v. Comstock, 16 Wall. 258, 270 (1873). “Otherwise the appellate jurisdiction could be defeated and the purpose of the statute authorizing the writ thwarted by unauthorized action of the district court obstructing the appeal.” Roche, supra, at 25.5 To say that a court of appeals has the power to direct a district court to proceed to judgment in a pending case “when it is its duty to do so,” 319 U. S., at 26, states the standard but does not decide this or any other particular case. It is essential that the moving party satisfy “the burden of showing that its right to issuance of the writ is ‘clear and indisputable.’ ” Bankers Life & Cas. Co., supra, at 384, quoting United States v. Duell, 172 U. S. 576, 582 (1899). Judge Will urges that Calvert does not have a “clear and indisputable” right to the adjudication of its claims in the District Court without regard to the concurrent state proceedings. To that issue we now must turn. Ill It is well established that “the pendency of an action in the state court is no bar to proceedings concerning the same matter in the Federal court having jurisdiction.” McClellan v. Carland, 217 U. S. 268, 282 (1910). It is equally well settled that a district court is “under no compulsion to exercise that jurisdiction,” Brillhart v. Excess Ins. Co., 316 U. S. 491, 5 A classic example of the proper issuance of the writ to protect eventual appellate jurisdiction is T her mt ron Products, Inc. v. Hermansdorjer, 423 U. S. 336 (1976), in which a case had been remanded to the state courts on grounds utterly unauthorized by the controlling statute. The dissenters in that case urged that Congress had intended to bar all review of remand orders, not that mandamus would have been inappropriate absent such a bar. Id., at 354 (Rehnquist, J., joined by Burger, C. J., and Stewart, J., dissenting). WILL v. CALVERT FIRE INS. CO. 663 655 Opinion of Rehnquist, J. 494 (1942), where the controversy may be settled more expeditiously in the state court. Although most of our decisions discussing the propriety of stays or dismissals of duplicative actions have concerned conflicts of jurisdiction between two federal district courts, e. g., Kerotest Mfg. Co., v. C-O-Two Fire Equipment Co., 342 U. S. 180 (1952); Landis v. North American Co., 299 U. S. 248 (1936), we have recognized the relevance of those cases in the analogous circumstances presented here. See Colorado River, 424 U. S., at 817-819. In both situations, the decision is largely committed to the “carefully considered judgment,” id., at 818, of the district court. This power has not always been so clear. In McClellan, on facts similar to those presented here, this Court indicated that the writ might properly issue where the District Court had stayed its proceedings in deference to concurrent state proceedings.6 Such an automatic exercise of authority may well have been appropriate in a day when Congress had authorized fewer claims for relief in the federal courts, so that duplicative litigation and the concomitant tension between state and federal courts could rarely result. However, as the overlap between state claims and federal claims increased, this Court soon recognized that situations would often arise when it would be appropriate to defer to the state courts. “Ordinarily it would be uneconomical as well as vexatious for a federal court to proceed in a declaratory judgment suit where another suit is pending in a state court presenting the same issues, not governed by federal law, 6 This Court there held, not that the writ should issue, but that the Court of Appeals should have required the District Judge to show cause why the writ should not issue. Judge Carland presented an affidavit to this Court attempting to defend his stay order on the basis of substantially completed state proceedings. As that affidavit was not in the record before the Court of Appeals, this Court did not “pass upon the sufficiency of those proceedings to authorize the orders in question,” 217 U. S., at 283, but directed the Court of Appeals to do so in the first instance. 664 437 U.S. OCTOBER TERM, 1977 Opinion of Rehnquist, J. between the same parties. Gratuitous interference with the orderly and comprehensive disposition of a state court litigation should be avoided.” Brillhart, supra, at 495. The decision in such circumstances is largely committed to the discretion of the district court. 316 U. S., at 494. Furthermore, Colorado River, supra, at 820, established that such deference may be equally appropriate even when matters of substantive federal law are involved in the case. It is true that Colorado River emphasized “the virtually unflagging obligation of the federal courts to exercise the jurisdiction given them.” 424 U. S., at 817. That language underscores our conviction that a district court should exercise its discretion with this factor in mind, but it in no way undermines the conclusion of Brillhart that the decision whether to defer to the concurrent jurisdiction of a state court is, in the last analysis, a matter committed to the district court’s discretion. Seizing upon the phrase “unflagging obligation” in an opinion which upheld the correctness of a district court’s final decision to dismiss because of concurrent jurisdiction does little to bolster a claim for the extraordinary writ of mandamus in a case such as this where the District Court has rendered no final decision. We think it of considerably more importance than did the Court of Appeals that Colorado River came before the Court of Appeals on appeal pursuant to 28 U. S. C. § 1291 following outright dismissal of the action by the District Court, rather than through an effort on the part of the federal-court plaintiff to seek mandamus. Calvert contends here, and the Court of Appeals for the Seventh Circuit agreed, that Judge Will’s order deferring the federal proceedings was “equivalent to a dismissal.” 560 F. 2d, at 796. We are loath to rest our analysis on this ubiquitous phrase, for if used carelessly or without a precise definition it may impede rather than assist sound resolution of the underlying legal issue. WILL v. CALVERT FIRE INS. CO. 665 655 Opinion of Rehnquist, J. Obviously, if Judge Will had dismissed Calvert’s action Calvert could have appealed the order of dismissal to the Court of Appeals, which could have required such action of Judge Will “as may be just under the circumstances.” 28 U. S. C. § 2106. Since he did not dismiss the action, Calvert remained free to urge reconsideration of his decision to defer based on new information as to the progress of the state case; to this extent, at least, deferral was not “equivalent to a dismissal.” There are sound reasons for our reiteration of the rule that a district court’s decision to defer proceedings because of concurrent state litigation is generally committed to the discretion of that court. No one can seriously contend that a busy federal trial judge, confronted both with competing demands on his time for matters properly within his jurisdiction and with inevitable scheduling difficulties because of the unavailability of lawyers, parties, and witnesses, is not entrusted with a wide latitude in setting his own calendar. Had Judge Will simply decided on his own initiative to defer setting this case for trial until the state proceedings were completed, his action would have been the “equivalent” of granting the motion of American Mutual to defer, yet such action would at best have afforded Calvert a highly dubious claim for mandamus. We think the fact that the judge accomplished this same result by ruling favorably on a party’s motion to defer does not change the underlying legal question. Although the District Court’s exercise of its discretion may be subject to review and modification in a proper interlocutory appeal, cf. Landis, 299 U. S., at 256-259, we are convinced that it ought not to be overridden by a writ of mandamus.7 Where 7 Although in at least one instance we approved the issuance of the writ upon a mere showing of abuse of discretion, La Buy v. Howes Leather Co., 352 U. S. 249, 257 (1957), we warned soon thereafter against the dangers of such a practice. “Courts faced with petitions for the peremptory writs must be careful lest they suffer themselves to be misled by labels such as ‘abuse of discretion’ and ‘want of power’ into interlocutory review of non- 666 OCTOBER TERM, 1977 Opinion of Rehnquist, J. 437U.S. a matter is committed to the discretion of a district court, it cannot be said that a litigant’s right to a particular result is “clear and indisputable.” 8 Calvert contends that a district court is without power to stay proceedings, in deference to a contemporaneous state action, where the federal courts have exclusive jurisdiction over the issue presented. Whether or not this is so, petitioner has not purported to stay consideration of Calvert’s claim for damages under the Securities Exchange Act of 1934, which is the only issue which may not be concurrently resolved by both courts.9 It is true that petitioner has not yet ruled upon this claim. Where a district court obstinately refuses to adjudicate a matter properly before it, a court of appeals may issue the writ to correct “unauthorized action of the district appealable orders on the mere ground that they may be erroneous.” Will v. United States, 389 U. S. 90, 98 n. 6 (1967). Beacon Theatres, Inc. v. Westover, 359 U. S. 500 (1959), is not to the contrary. Both the Court and the dissenters agreed that mandamus should issue to protect a clear right to a jury trial. Id., at 511; ibid. (Stewart, J., dissenting). The Court simply concluded that it was “not permissible,” id., at 508, for the District Court to postpone a jury trial until after most of the relevant issues had been settled in an equitable action before the court. Here, we have repeatedly recognized that it is permissible for a district court to defer to the concurrent jurisdiction of a state court. 8 That a litigant’s right to proceed with a duplicative action in a federal court can never be said to be “clear and indisputable” is made all the more apparent by our holding earlier this Term in General Atomic Co. v. Felter, 434 U. S. 12 (1977), that a state court lacks the power to restrain vexatious litigation in the federal courts. There, we reaffirmed the principle that “[f]ederal courts are fully capable of preventing their mis-use for purposes of harassment.” Id., at 19. 9 The only other issue encompassed by the writ was Calvert’s Rule 10b-5 claim for equitable relief. It is not disputed here that the state court has jurisdiction to rescind the agreement as Calvert requests. That being conceded, we find no merit in Calvert’s further argument that the statutory grant of exclusive jurisdiction in any way distinguishes this aspect of the case from our earlier decisions in which both the state and federal courts had power to grant the desired relief. WILL v. CALVERT FIRE INS. CO. 667 655 Blackmun, J., concurring in judgment court obstructing the appeal.” Roche, 319 U. S., at 25, citing Ex parte United States, 287 U. S. 241 (1932). Calvert, however, has neither alleged nor proved such a heedless refusal to proceed as a basis for the issuance of the writ here. Its petition offers only the bare allegation that Judge Will “in effect” abated the damages claim in deference to the state proceedings. App. 12. Judge Will has never issued such an order, and the sparse record before us will not support any such inference. So far as appears, the delay in adjudicating the damages claim is simply a product of the normal excessive load of business in the District Court, compounded by “the unfortunate consequence of making the judge a litigant” in this mandamus proceeding. Ex parte Fahey, 332 U. S. 258, 260 (1947). The judgment of the Court of Appeals is therefore Reversed. Mr. Justice Blackmun, concurring in the judgment. The plurality’s opinion, ante, at 662-663, appears to me to indicate that it now regards as fully compatible the Court’s decisions in Brillhart v. Excess Ins. Co., 316 U. S. 491 (1942), a diversity case, and Colorado River Water Conservation Dist. v. United States, 424 U. S. 800 (1976), a federal-issue case. I am not at all sure that this is so. I—as were Mr. Justice Stewart and Mr. Justice Stevens—was in dissent in Colorado River, and if the holding in that case is what I think it is, and if one assumes, as I do not, that Brillhart has any application here, the Court cut back on Mr. Justice Frankfurter’s rather sweeping language in Brillhart, 316 U. S., at 494-495.* *“Although the District Court had jurisdiction of the suit under the Federal Declaratory Judgments Act, it was under no compulsion to exercise that jurisdiction. The petitioner’s motion to dismiss the bill was addressed to the discretion of the court. . . . The motion rested upon the claim that, since another proceeding was pending in a state court in which all the matters in controversy between the parties could be fully adjudicated, a declaratory judgment in the federal court was unwarranted. The correctness of this claim was certainly relevant in determining whether 668 OCTOBER TERM, 1977 Brennan, J., dissenting 437 U. S. Because Judge Will’s stay order was issued prior to this Court’s decision in Colorado River, and he therefore did not have such guidance as that case affords in the area, I join in the Court’s reversal of the Court of Appeals’ issuance of a writ of mandamus. The issuance was premature. The Court of Appeals should have done no more than require reconsideration of the case by Judge Will in light of Colorado River. Mr. Chief Justice Burger, dissenting. I am in general agreement with Mr. Justice Brennan’s dissenting opinion. I write separately only to emphasize that I consider it unnecessary to determine in the context of this case whether it would ever be appropriate to give res judicata effect to a state-court judgment implicating a claim over which the federal courts have been given exclusive jurisdiction. Our concern here is simply with the propriety of a federal court’s delaying adjudication of such a claim in deference to a statecourt proceeding. As Mr. Justice Brennan correctly notes, whatever the proper resolution of the res judicata issue, a federal court remains under an obligation to expeditiously consider and resolve those claims which Congress explicitly reserved to the federal courts. With this minor caveat, I join Mr. Justice Brennan in his dissent. Mr. Justice Brennan, with whom The Chief Justice, Mr. Justice Marshall, and Mr. Justice Powell join, dissenting. This case falls within none of the three general abstention categories, and the opinion of my Brother Rehnquist there- the District Court should assume jurisdiction and proceed to determine the rights of the parties. Ordinarily it would be uneconomical as well as vexatious for a federal court to proceed in a declaratory judgment suit where another suit is pending in a state court presenting the same issues, not governed by federal law, between the same parties. Gratuitous interference with the orderly and comprehensive disposition of a state court litigation should be avoided.” WILL v. CALVERT FIRE INS. CO. 669 655 Brennan, J., dissenting fore strains to bring it within the principles that govern in a very narrow class of “exceptional” situations that involve “the contemporaneous exercise of concurrent jurisdictions.” Colorado River Water Conservation Dist. v. United States, 424 U. S. 800, 813-818 (1976). In so straining, the opinion reaches a result supported by neither policy nor precedent, ignores difficult legal issues, misapprehends the significance of the proceedings below, and casts doubt upon a decision that has stood unquestioned for nearly 70 years. Moreover, there lurks an ominous potential for the abdication of federal-court jurisdiction in the opinion’s disturbing indifference to “the virtually unflagging obligation of the federal courts to exercise the jurisdiction given them,” id., at 817—for obedience to that obligation becomes all the more important when, as here, Congress has made that jurisdiction exclusive. I dissent. I Because this case came to the Court of Appeals on respondent Calvert Fire Insurance Co.’s motion for a writ of mandamus to compel Judge Will to adjudicate its claims for damages and equitable relief under the Securities Exchange Act of 1934 (1934 Act), I agree with my Brother Rehnquist that it is essential to determine precisely what obligation the District Court had to adjudicate respondent’s 1934 Act claims. That, however, is as far as my agreement goes. On the same day Calvert filed its answer to the state suit instituted against it—an answer containing a defense under the 1934 Act that the state court was required to recognize under the Supremacy Clause—it commenced an action in Federal District Court seeking relief under the 1934 Act, the Securities Act of 1933, and various state provisions. The District Court stayed all claims alleged in this complaint, other than Calvert’s claim for money damages under Rule 10b-5 of the 1934 Act, pending the outcome of the state suit. Although the District Court did not formally stay the Rule 10b-5 damages claim and heard oral argument on the primary . 670 OCTOBER TERM, 1977 Brennan, J., dissenting 437U.S. issue underlying the claim—whether a participatory interest in a reinsurance pool is a “security”—the District Court has yet to rule on this issue, so Calvert’s Rule 10b-5 damages claim, like the rest of its federal suit, remains in suspension. Section 27 of the 1934 Act, 15 U. S. C. § 78aa (1976 ed.), gives the federal courts exclusive jurisdiction over claims arising under the Act. This jurisdictional grant evinces a legislative desire for the uniform determination of such claims by tribunals expert in the administration of federal laws and sensitive to the national concerns underlying them. When Congress thus mandates that only federal courts shall exercise jurisdiction to adjudicate specified claims, the “well established” principle1—accepted by my Brother Rehnquist, ante, at 662—of McClellan v. Carland, 217 U. S. 268, 282 (1910), that “the pendency of an action in the state court is no bar to proceedings concerning the same matter in the Federal court having jurisdiction,” governs a multo fortiori. Yet, relying on the completely inapposite case of Brillhart v. Excess Insurance Co., 316 U. S. 491 (1942), the opinion of my Brother Rehnquist disregards the McClellan principle and all but ignores the analysis set forth in Colorado River Water Conservation Dist. v. United States, supra, our most recent pronouncement on a district court’s authority to defer to a contemporaneous state proceeding. In Brillhart, the District Court dismissed a diversity suit for a declaratory judgment because of the pendency in state court of a suit between the same parties and involving the same subject matter. The Court of Appeals reversed, holding that the dismissal was an abuse of discretion. In reversing the Court of Appeals, this Court reasoned: “Although the District Court had jurisdiction of the suit under the Federal Declaratory Judgments Act, it 1See, e. g., Thermtron Products, Inc. v. Hermansdorfer, 423 U. S. 336, 344-345 (1976); Meredith v. Winter Haven, 320 U. S. 228, 234-235 (1943). WILL v. CALVERT FIRE INS. CO. 671 655 Brennan, J., dissenting was under no compulsion to exercise that jurisdiction. The petitioner’s motion to dismiss the bill was addressed to the discretion of the court. Aetna Casualty Co. v. Quarles, 92 F. 2d 321; Maryland Casualty Co. v. Consumers Finance Service, 101 F. 2d 514; American Automobile Ins. Co. v. Freundt, 103 F. 2d 613 .... The motion rested upon the claim that, since another proceeding was pending in a state court in which all the matters in controversy between the parties could be fully adjudicated, a declaratory judgment in the federal court was unwarranted. The correctness of this claim was certainly relevant in determining whether the District Court should assume jurisdiction and proceed to determine the rights of the parties. Ordinarily it would be uneconomical as well as vexatious for a federal court to proceed in a declaratory judgment suit where another suit is pending in a state court presenting the same issues, not governed by federal law, between the same parties.” Brillhart v. Excess Insurance Co., supra, at 494-495 (emphasis added). As is readily apparent, crucial to this Court’s approval of the District Court’s dismissal of the suit in Brillhart were two factors absent here. First, because the federal suit was founded on diversity, state rather than federal law would govern the outcome of the federal suit. Second, and more significantly, the federal suit was for a declaratory judgment. Under the terms of the provision empowering federal courts to entertain declaratory judgment suits, 28 U. S. C. § 2201, the assumption of jurisdiction over such suits is discretionary. That section provides: “In a case of actual controversy within its jurisdiction . . . any court of the United States, upon the filing of an appropriate pleading, may declare the rights and other legal relations of any interested party seeking such declaration . . . .” (Emphasis added.) It was primarily because federal jurisdiction over declaratory judgment suits is 672 OCTOBER TERM, 1977 Brennan, J., dissenting 437U.S. discretionary that Brillhart found the District Court’s deference to state-court proceedings permissible. This is clear from the lower court cases approvingly cited by Brillhart—American Automobile Insurance Co. v. Freundt, 103 F. 2d 613 (CA7 1939); Maryland Casualty Co. v. Consumers Finance Service, 101 F. 2d 514 (CA3 1938); and Aetna Casualty Co. v. Quarles, 92 F. 2d 321 (CA4 1937)—all of which emphasized that a district court’s discretion to dismiss a federal declaratory judgment suit in favor of a pending state suit is a product of the permissive nature of declaratory judgment jurisdiction.2 Obviously neither the logic nor the holding of Brillhart is pertinent where, as here, federal jurisdiction is not only non-discretionary, but exclusive. The unpersuasive grope for supporting precedent in which the opinion of my Brother Rehnquist engages is especially lamentable in light of our decision only two Terms ago in Colorado River Water Conservation Dist. v. United States. In Colorado River we addressed the precise issue presented here: the circumstances in which it is appropriate for a federal district court to stay a proceeding before it in deference to a parallel state-court proceeding in situations falling within none of the traditional categories for federal abstention. We explained that, in contrast to situations in which jurisdiction is concurrent in two or more federal courts, 2 These decisions recognized, however, that even where a federal suit seeks only declaratory relief, a district court does not have unbridled authority to dismiss the action in deference to a concurrent state suit. For example, the court in Maryland Casualty Co. v. Consumers Finance Service, 101 F. 2d, at 515, observed: “The granting of the remedy of a declaratory judgment is . . . discretionary with the court and it may be refused if it will not finally settle the rights of the parties or if it is being sought merely to determine issues. involved in cases already pending. Aetna Casualty & Surety Co. v. Quarles, 4 Cir., 92 F. 2d 321. It may not be refused, however, merely on the ground that another remedy is available ... or because of the pendency of another suit, if the controversy between the parties will not necessarily be determined in that suit.” WILL v. CALVERT FIRE INS. CO. 673 655 Brennan, J., dissenting where the action paralleling a federal suit is in a state court, the federal court’s power to dismiss the suit before it in deference to the parallel proceeding is limited by the “virtually unflagging obligation of the federal courts to exercise the jurisdiction given them.” 424 U. S., at 817. Because a federal district court’s power is so limited, the circumstances that justify federal-court inaction in deference to a state proceeding must be “exceptional.” Id., at 818. Just how “exceptional” such circumstances must be was made clear by our admonition that “the circumstances permitting the dismissal of a federal suit due to the presence of a concurrent state proceeding for reasons of wise judicial administration are considerably more limited than the circumstances appropriate for abstention.” Ibid. Since we had previously noted that “‘[abdication of the obligation to decide cases can be justified under [the abstention] doctrine only in the exceptional circumstances where the order to the parties to repair to the State court would clearly serve an important countervailing interest,’ ” id., at 813, quoting County of Allegheny v. Frank Mashuda Co., 360 U. S. 185, 188-189 (1959), the circumstances warranting dismissal “for reasons of wise judicial administration” must be rare indeed. Such rare circumstances were present in Colorado River. There, the decisive factor in favor of staying the concurrent federal proceedings was “[t]he clear federal policy,” evinced by the McCarran Amendment, of “avoid [ing the] piecemeal adjudication of water rights in a river system ... a policy that recognizes the availability of comprehensive state systems for adjudication of water rights as the means for achieving [this] goa[l].” 424 U. S., at 819. No comparable federal policy favoring unitary state adjudication exists here. In fact, as evinced by the exclusive jurisdiction of the federal courts to determine 1934 Act claims, the relevant federal policy here is the precise opposite of that found to require deference to the concurrent state proceeding in Colorado River. 674 437 U. S. OCTOBER TERM, 1977 Brennan, J., dissenting Ignoring wholesale the analytical framework set forth in Colorado River, whose vitality is not questioned, the opinion of my Brother Rehnquist seemingly focuses on one of the four secondary factors found to support the federal dismissal in that case—the fact that the state proceedings were initiated before the federal suit—and finds that factor sufficient to insulate the District Court’s actions here from mandamus review. Even putting aside the opinion’s case-reading errors— its flouting of McClellan, its misreliance on Brillhart, and its misapplication of Colorado River—and analyzing this case on the opinion’s own erroneous terms, the conclusion is still compelled that the District Court had no authority to stay Calvert’s 1934 Act claims. Quite conveniently, the opinion of my Brother Rehnquist avoids any discussion of the possible res judicata or collateral-estoppel effects the state court’s determination of Calvert’s 1934 Act defense would have on Calvert’s 1934 Act claims for affirmative relief in federal court.3 To be sure, the preclusive effect of a state-court determination of a claim within the exclusive jurisdiction of the federal courts is an unresolved and difficult issue. See generally Note, Res Judicata: Exclusive Federal Jurisdiction and the Effect of Prior State-Court Determinations, 53 Va. L. Rev. 1360 (1967). For myself, I confess to serious doubt that it is ever appropriate to accord res judicata effect to a state-court determination of a claim over which the federal courts have exclusive jurisdiction; for surely state-court determinations should not disable federal courts from ruling de novo on purely legal questions surrounding such federal claims. See Cotier v. Inter- 3 Because the Court of Appeals held that “the district court should not have deferred to the state court on grounds of federalism in light of Colorado River,” it found it unnecessary to “reach the difficult issue of whether the conclusion of the state proceedings would have a collateral estoppel effect on the Rule 10b-5 claim for damages over which the court had retained jurisdiction but declined to resolve.” 560 F. 2d 792, 797. WILL v. CALVERT FIRE INS. CO. 675 655 Brennan, J., dissenting County Orthopaedic Assn., 526 F. 2d 537 (CA3 1975); McGough v. First Arlington National Bank, 519 F. 2d 552 (CA7 1975); Clark v. Watchie, 513 F. 2d 994 (CA9 1975). As recognized by Judge Learned Hand in Lyons v. Westinghouse Electric Co., 222 F. 2d 184, 189 (CA2 1955), “the grant to the district courts of exclusive jurisdiction over the action . . . should be taken to imply an immunity of their decisions from any prejudgment elsewhere.” I recognize that it may make sense, for reasons of fairness and judicial economy, to give collateral-estoppel effect to specific findings of historical facts by a state court’s adjudicating an exclusively federal claim raised as a defense, see Granader v. Public Bank, 417 F. 2d 75 (CA6 1969), but there are reasons why even such a limited preclusive effect should not be given state-court determinations. It is at least arguable that, in creating and defining a particular federal claim, Congress assumed that the claim would be litigated only in the context of federal-court procedure—a fair assumption when the claim is within exclusive federal jurisdiction. For example, Congress may have thought the liberal federal discovery procedures crucial to the proper determination of the factual disputes underlying the federal claim. All this is not to say that I disagree with the refusal of the opinion of my Brother Rehnquist to decide what preclusive effects the state court’s determination of Calvert’s Rule 10b-5 defense would have in Calvert’s federal action, so much as it is to expose the opinion’s error in failing even to consider the res judicata/collateral estoppel problem in evaluating the District Court’s obligation to adjudicate Calvert’s Rule 10b-5 claim. In my view, regardless of whether the state-court judgment would be given res judicata or collateral-estoppel effect, it was incumbent upon the District Court—at least in the absence of other overriding reasons—expeditiously to adjudicate at least Calvert’s 1934 Act claims. If res judicata effect is accorded the prior state-court judgment, the exclusive jurisdic 676 OCTOBER TERM, 1977 Brennan, J., dissenting 437 U. S. tion given the federal courts over 1934 Act claims would be effectively thwarted, and the policy of uniform and effective federal administration and interpretation of the 1934 Act frustrated. A stay having so undesirable a consequence could possibly be justified only by compelling circumstances absent here. On the other hand, if the state-court adjudication is not given res judicata or collateral-estoppel effect, the 1934 Act claims will have to be adjudicated in federal court in any event, and there would be no reason for staying the federal action since nothing that transpires in the state proceedings would affect the adjudication of the federal claims. Thus, regardless of the proper disposition of the res judicata/collateral estoppel question, it is clear that a district court should not stay claims over which the federal courts have exclusive jurisdiction. See Cotier v. Inter-County Orthopaedic Assn., supra; Lecor, Inc. v. United States District Court, 502 F. 2d 104 (CA9 1974). II Whether evaluated under the “clear abuse of discretion” standard set forth in La Buy v. Howes Leather Co., 352 U. S. 249, 257 (1957), or under the prong of Will v. United States, 389 U. S. 90, 95 (1967), and Roche v. Evaporated Milk Assn., 319 U. S. 21, 26 (1943), that permits the use of mandamus “to compel [an inferior court] to exercise its authority when it is its duty to do so,” the issuance of the writ of mandamus by the Court of Appeals was proper; there is simply a complete dearth of “exceptional” circumstances countervailing the District Court’s “unflagging obligation” to exercise its exclusive jurisdiction. The opinion of my Brother Rehnquist asserts, however, that the District Court “has not purported to stay consideration of Calvert’s claim for damages under the Securities Exchange Act of 1934,” but rather has simply “not yet ruled upon this claim.” Ante, at 666. While technically accurate, this characterization of the status of the proceedings below utterly ignores two important facts that shed more than WILL v. CALVERT FIRE INS. CO. 677 655 Brennan, J., dissenting a little illumination on the true procedural posture of this case. First, at the time the Court of Appeals granted the writ, Calvert’s Rule 10b-5 damages action had been before Judge Will for more than 2% years without a ruling on the basic legal issue underlying the claim. Second, and for me dispositive, the District Court indicated that it would give the state court’s determination that the disputed transaction did not involve a “security” within the meaning of the 1934 Act res judicata effect, App. to Brief for Respondent Calvert Fire Insurance Co. E-l, thereby depriving Calvert of a federal-court determination of a legal issue within the exclusive jurisdiction of the federal courts. This Court has held that mandamus will lie to correct a district court’s improper deference to pending state-court proceedings, McClellan v. Carland, 217 U. S. 268 (1910), and to preserve a proper federal-court determination of a federal issue, Beacon Theatres, Inc. n. Westover, 359 U. S. 500 (1959). Where, as here, both of these justifications are present, the propriety of the issuance of the writ cannot be questioned. I would affirm the Court of Appeals. 678 OCTOBER TERM, 1977 Syllabus 437 U. S. HUTTO ET AL. v. FINNEY et al. CERTIORARI TO THE UNITED STATES COURT OF APPEALS FOR THE EIGHTH CIRCUIT No. 76-1660. Argued February 21, 1978—Decided June 23, 1978 After finding in respondent prison inmates’ action against petitioner prison officials that conditions in the Arkansas prison system constituted cruel and unusual punishment in violation of the Eighth and Fourteenth Amendments, the District Court entered a series of detailed remedial orders. On appeal to the Court of Appeals, petitioners challenged two aspects of that relief: (1) an order placing a maximum limit of 30 days on confinement in punitive isolation; and (2) an award of attorney’s fees to be paid out of Department of Correction funds, based on the District Court’s finding that petitioners had acted in bad faith in failing to cure the previously identified violations. The Court of Appeals affirmed and assessed an additional attorney’s fee to cover services on appeal. Held: 1. The District Court did not err in including the 30-day limitation on sentences to isolation as part of its comprehensive remedy to correct the constitutional violations. Where the question before the court was whether these past constitutional violations had been remedied, it was entitled to consider the severity of the violations in assessing the constitutionality of conditions in the isolation cells, the length of time each inmate spent in isolation being simply one consideration among many. Pp. 685-688. 2. The District Court’s award of attorney’s fees to be paid out of Department of Correction funds is adequately supported by its finding that petitioners had acted in bad faith, and does not violate the Eleventh Amendment. The award served the same purpose as a remedial fine imposed for civil contempt, and vindicated the court’s authority over a recalcitrant litigant. There being no reason to distinguish the award from any other penalty imposed to enforce a prospective injunction, the Eleventh Amendment’s substantive protections do not prevent the award against the Department’s officers in their official capacities, and the fact that the order directed the award to be paid out of Department funds rather than being assessed against petitioners in their official capacities, does not constitute reversible error. Pp. 689-693. 3. The Civil Rights Attorney’s Fees Awards Act of 1976, which provides that “[i]n any action” to enforce certain civil rights laws (including the law under which this action was brought), federal courts HUTTO v. FINNEY 679 678 Syllabus may award prevailing parties reasonable attorney’s fees “as part of the costs,” supports the additional award of attorney’s fees by the Court of Appeals. Pp. 693-700. (a) The Act’s broad language and the fact that it primarily applies to laws specifically passed to restrain unlawful state action, as well as the Act’s legislative history, make it clear that Congress, when it passed the Act, intended to exercise its power to set aside the States’ immunity from retroactive relief in order to enforce the Fourteenth Amendment, and to authorize fee awards payable by the States when their officials are sued in their official capacities. Pp. 693-694. (b) Costs have traditionally been awarded against States without regard for the States’ Eleventh Amendment immunity, and it is much too late to single out attorney’s fees as the one kind of litigation cost whose recovery may not be authorized by Congress without an express statutory waiver of States’ immunity. Pp. 694—698. (c) The fact that neither the State nor the Department of Correction was expressly named as a defendant, does not preclude the Court of Appeals’ award, since, although the Eleventh Amendment prevented respondents from suing the State by name, their injunctive suit against petitioner prison officials was, for all practical purposes, brought against the State, so that absent any indication that petitioners acted in bad faith before the Court of Appeals, the Department of Correction is the entity intended by Congress to bear the burden of the award. Pp. 699-700. 548 F. 2d 740, affirmed. Stevens, J., delivered the opinion of the Court, in which Brennan, Stewart, Marshall, and Blackmun, JJ., joined, in Part I of which White, J., joined, and in Parts I and II-A of which Burger, C. J., and Powell, J., joined. Brennan, J., filed a concurring opinion, post, p. 700. Powell, J., filed an opinion concurring in part and dissenting in part, in which Burger, C. J., joined, and in the dissenting portion of which White and Rehnquist, JJ., joined, post, p. 704. Rehnquist, J., filed a dissenting opinion, in Part II of which White, J., joined, post, p. 710. Garner L. Taylor, Jr., Assistant Attorney General of Arkansas, argued the cause for petitioners. On the brief were Bill Clinton, Attorney General, and Robert Alston Newcomb. Philip E. Kaplan argued the cause for respondents. With him on the brief were Jack Holt, Jr., Philip E. McMath, Jack 680 OCTOBER TERM, 1977 Opinion of the Court 437 U. S. Greenberg, James M. Nabrit III, Charles Stephen Ralston, Stanley Bass, Eric Schnapper, and Lynn Walker* Mr. Justice Stevens delivered the opinion of the Courts After finding that conditions in the Arkansas penal system constituted cruel and unusual punishment, the District Court entered a series of detailed remedial orders. On appeal to the United States Court of Appeals for the Eighth Circuit, petitioners1 challenged two aspects of that relief: (1) an order placing a maximum limit of 30 days on confinement in punitive isolation; and (2) an award of attorney’s fees to be paid out of Department of Correction funds. The Court of *Briefs of amici curiae urging reversal were filed by Evelle J. Younger, Attorney General, Jack R. Winkler, Chief Assistant Attorney General, Edward P. O’Brien, Assistant Attorney General, and Gloria F. DeHart and Patrick G. Golden, Deputy Attorneys General, for the State of California; by Richard C. Turner, Attorney General, Stephen C. Robinson, Special Assistant Attorney General, and Theodore R. Boecker and Frederick M. Haskins, Assistant Attorneys General, for the State of Iowa; and by Robert P. Kane, Attorney General, and Melvin R. Shuster and J. Justin Blewitt, Jr., Deputy Attorneys General, for the Commonwealth of Pennsylvania. Briefs of amici curiae urging affirmance were filed by Solicitor General McCree, Assistant Attorney General Days, Walter W. Barnett, and Dennis J. Dimsey for the United States; by Bruce J. Ennis, Burt Neuborne, and Richard Emery for the American Civil Liberties Union et al.; and by Charles A. Bane, Thomas D. Barr, Armand Derjner, Paul R. Dimond, Norman Redlich, Robert A. Murphy, Norman J. Chachkin, Richard S. Kohn, David M. Lipman, and William E. Caldwell for the Lawyers’ Committee for Civil Rights Under Law. Briefs of amici curiae were filed by A. F. Summer, Attorney General, and P. Roger Googe, Jr., and Peter M. Stockett, Jr., Assistant Attorneys General, for the State of Mississippi; and by John L. Hill, Attorney General, David M. Kendall, First Assistant Attorney General, and Joe B. Dibrell, Richel Rivers, and Nancy Simonson, Assistant Attorneys General, for the State of Texas. +Mr. Justice White joins only Part I of this opinion. 1 Petitioners are the Commissioner of Correction and members of the Arkansas Board of Correction. HUTTO v. FINNEY 681 678 Opinion of the Court Appeals affirmed and assessed an additional attorney’s fee to cover services on appeal. 548 F. 2d 740 (1977). We granted certiorari, 434 U. S. 901, and now affirm. This litigation began in 1969; it is a sequel to two earlier cases holding that conditions in the Arkansas prison system violated the Eighth and Fourteenth Amendments.2 Only a brief summary of the facts is necessary to explain the basis for the remedial orders. The routine conditions that the ordinary Arkansas convict had to endure were characterized by the District Court as “a dark and evil world completely alien to the free world.” Holt v. Sarver, 309 F. Supp. 362, 381 (ED Ark. 1970) {Holt II). That characterization was amply supported by the evidence.3 2 This case began as Holt v. Sarver, 300 F. Supp. 825 (ED Ark. 1969) (Holt I). The two earlier cases were Talley n. Stephens, 247 F. Supp. 683 (ED Ark. 1965), and Jackson v. Bishop, 268 F. Supp. 804 (ED Ark. 1967), vacated, 404 F. 2d 571 (CA8 1968). Judge Henley decided the first of these cases in 1965, when he was Chief Judge of the Eastern District of Arkansas. Although appointed to the Court of Appeals for the Eighth Circuit in 1975, he was specially designated to continue to hear this case as a District Judge. 3 The administrators of Arkansas’ prison system evidently tried to operate their prisons at a profit. See Talley v. Stephens, supra, at 688. Cummins Farm, the institution at the center of this litigation, required its 1,000 inmates to work in the fields 10 hours a day, six days a week, using mule-drawn tools and tending crops by hand. 247 F. Supp., at 688. The inmates were sometimes required to run to and from the fields, with a guard in an automobile or on horseback driving them on. Holt v. Hutto, 363 F. Supp. 194, 213 (ED Ark. 1973) (Holt III). They worked in all sorts of weather, so long as the temperature was above freezing, sometimes in unsuitably light clothing or without shoes. Holt II, 309 F. Supp., at 370. The inmates slept together in large, 100-man barracks, and some convicts, known as “creepers,” would slip from their beds to crawl along the floor, stalking their sleeping enemies. In one 18-month period, there were 17 stabbings, all but 1 occurring in the barracks. Holt I, supra, at 830-831. Homosexual rape was so common and uncontrolled that some potential victims dared not sleep; instead they would leave their beds and 682 OCTOBER TERM, 1977 Opinion of the Court 437U.S. The punishments for misconduct not serious enough to result in punitive isolation were cruel,4 unusual,5 and unpredictable.6 It is the discipline known as “punitive isolation” that is most relevant for present purposes. Confinement in punitive isolation was for an indeterminate period of time. An average of 4, and sometimes as many as 10 or 11, prisoners were crowded into windowless 8'xlO' cells containing no furniture other than a source of water and a toilet that could only be flushed from outside the cell. Holt v. Sarver, 300 F. Supp. 825, 831-832 (ED Ark. 1969) {Holt I}. At night the prisoners were given mattresses to spread on the floor. Although some prisoners suffered from infectious diseases such as hepatitis and venereal disease, mattresses were removed and jumbled together each morning, spend the night clinging to the bars nearest the guards’ station. Holt II, supra, at 377. 4 Inmates were lashed with a wooden-handled leather strap five feet long and four inches wide. Talley n. Stephens, supra, at 687. Although it was not official policy to do so, some inmates were apparently whipped for minor offenses until their skin was bloody and bruised. Jackson v. Bishop, supra, at 810-811. 5 The “Tucker telephone,” a hand-cranked device, was used to adminis-ter electrical shocks to various sensitive parts of an inmate’s body. Jackson v. Bishop, supra, at 812. 6 Most of the guards were simply inmates who had been issued guns. Holt II, supra, at 373. Although it had 1,000 prisoners, Cummins employed only eight guards who were not themselves convicts. Only two nonconvict guards kept watch over the 1,000 men at night. 309 F. Supp., at 373. While the “trusties” maintained an appearance of order, they took a high toll from the other prisoners. Inmates could obtain access to medical treatment only if they bribed the trusty in charge of sick call. As the District Court found, it was “within the power of a trusty guard to murder another inmate with practical impunity,” because trusties with weapons were authorized to use deadly force against escapees. Id., at 374. “Accidental shootings” also occurred; and one trusty fired his shotgun into a crowded barracks because the inmates would not turn off their TV. Ibid. Another trusty beat an inmate so badly the victim required partial dentures. Talley v. Stephens, supra, at 689. HUTTO v. FINNEY 683 678 Opinion of the Court then returned to the cells at random in the evening. Id., at 832. Prisoners in isolation received fewer than 1,000 calories a day;7 their meals consisted primarily of 4-inch squares of “grue,” a substance created by mashing meat, potatoes, oleo, syrup, vegetables, eggs, and seasoning into a paste and baking the mixture in a pan. Ibid. After finding the conditions of confinement unconstitutional, the District Court did not immediately impose a detailed remedy of its own. Instead, it directed the Department of Correction to “make a substantial start” on improving conditions and to file reports on its progress. Holt I, supra, at 833-834. When the Department’s progress proved unsatisfactory, a second hearing was held. The District Court found some improvements, but concluded that prison conditions remained unconstitutional. Holt II, 309 F. Supp., at 383. Again the court offered prison administrators an opportunity to devise a plan of their own for remedying the constitutional violations, but this time the court issued guidelines, identifying four areas of change that would cure the worst evils: improving conditions in the isolation cells, increasing inmate safety, eliminating the barracks sleeping arrangements, and putting an end to the trusty system. Id., at 385. The Department was ordered to move as rapidly as funds became available. Ibid. After this order was affirmed on appeal, Holt v. Sarver, 442 F. 2d 304 (CA8 1971), more hearings were held in 1972 and 1973 to review the Department’s progress. Finding substantial improvements, the District Court concluded that continuing supervision was no longer necessary. The court held, 7 A daily allowance of 2,700 calories is recommended for the average male between 23 and 50. National Academy of Sciences, Recommended Dietary Allowances, Appendix (8th rev. ed. 1974). Prisoners in punitive isolation are less active than the average person; but a mature man who spends 12 hours a day lying down and 12 hours a day simply sitting or standing consumes approximately 2,000 calories a day. Id., at 27. 684 437 U. S. OCTOBER TERM, 1977 Opinion of the Court however, that its prior decrees would remain in effect and noted that sanctions, as well as an award of costs and attorney’s fees, would be imposed if violations occurred. Holt v. Hutto, 363 F. Supp. 194, 217 (ED Ark. 1973) (Holt III). The Court of Appeals reversed the District Court’s decision to withdraw its supervisory jurisdiction, Finney v. Arkansas Board of Correction, 505 F. 2d 194 (CA8 1974), and the District Court held a fourth set of hearings. 410 F. Supp. 251 (ED Ark. 1976). It found that, in some respects, conditions had seriously deteriorated since 1973, when the court had withdrawn its supervisory jurisdiction. Cummins Farm, which the court had condemned as overcrowded in 1970 because it housed 1,000 inmates, now had a population of about 1,500. Id., at 254-255. The situation in the punitive isolation cells was particularly disturbing. The court concluded that either it had misjudged conditions in these cells in 1973 or conditions had become much worse since then. Id., at 275. There were twice as many prisoners as beds in some cells. And because inmates in punitive isolation are often violently antisocial, overcrowding led to persecution of the weaker prisoners. The “grue” diet was still in use, and practically all inmates were losing weight on it. The cells had been vandalized to a “very substantial” extent. Id., at 276. Because of their inadequate numbers, guards assigned to the punitive isolation cells frequently resorted to physical violence, using nightsticks and Mace in their efforts to maintain order. Prisoners were sometimes left in isolation for months, their release depending on “their attitudes as appraised by prison personnel.” Id., at 275. The court concluded that the constitutional violations identified earlier had not been cured. It entered an order that placed limits on the number of men that could be confined in one cell, required that each have a bunk, discontinued the “grue” diet, and set 30 days as the maximum isolation sentence. The District Court gave detailed consideration to HUTTO v. FINNEY 685 678 Opinion of the Court the matter of fees and expenses, made an express finding that petitioners had acted in'bad faith, and awarded counsel “a fee of $20,000.00 to be paid out of Department of Correction funds.” Id., at 285. The Court of Appeals affirmed and assessed an additional $2,500 to cover fees and expenses on appeal. 548 F. 2d, at 743. I The Eighth Amendment’s ban on inflicting cruel and unusual punishments, made applicable to the States by the Fourteenth Amendment, “proscribe [s] more than physically barbarous punishments.” Estelle v. Gamble, 429 U. S. 97, 102. It prohibits penalties that are grossly disproportionate to the offense, Weems v. United States, 217 U. S. 349, 367, as well as those that transgress today’s “ ‘broad and idealistic concepts of dignity, civilized standards, humanity, and decency.’ ” Estelle v. Gamble, supra, at 102, quoting Jackson v. Bishop, 404 F. 2d 571, 579 (CAS 1968). Confinement in a prison or in an isolation cell is a form of punishment subject to scrutiny under Eighth Amendment standards. Petitioners do not challenge this proposition; nor do they disagree with the District Court’s original conclusion that conditions in Arkansas’ prisons, including its punitive isolation cells, constituted cruel and unusual punishment. Rather, petitioners single out that portion of the District Court’s most recent order that forbids the Department to sentence inmates to more than 30 days in punitive isolation. Petitioners assume that the District Court held that indeterminate sentences to punitive isolation always constitute cruel and unusual punishment. This assumption misreads the District Court’s holding. Read in its entirety, the District Court’s opinion makes it abundantly clear that the length of isolation sentences was not considered in a vacuum. In the court’s words, punitive isolation “is not necessarily unconstitutional, but it may be, depending on the duration of the confinement and the con 686 OCTOBER TERM, 1977 Opinion of the Court 437U.S. ditions thereof.” 410 F. Supp., at 275.8 It is perfectly obvious that every decision to rem6ve a particular inmate from the general prison population for an indeterminate period could not be characterized as cruel and unusual. If new conditions of confinement are not materially different from those affecting other prisoners, a transfer for the duration of a prisoner’s sentence might be completely unobjectionable and well within the authority of the prison administrator. Cf. Meachum v. Fano, 427 U. S. 215. It is equally plain, however, that the length of confinement cannot be ignored in deciding whether the confinement meets constitutional standards. A filthy, overcrowded cell and a diet of 8 The Department reads the following sentence in the District Court’s 76-page opinion as an unqualified holding that any indeterminate sentence to solitary confinement is unconstitutional: “The court holds that the policy of sentencing inmates to indeterminate periods of confinement, in punitive isolation is unreasonable and unconstitutional.” 410 F. Supp., at 278. But in the context of its full opinion, we think it quite clear that the court was describing the specific conditions found in the Arkansas penal system. Indeed, in the same paragraph it noted that “segregated confinement under maximum security conditions is one thing; segregated confinement under the punitive conditions that have been described is quite another thing.” Ibid, (emphasis in original). The Department also suggests that the District Court made rehabilitation a constitutional requirement. The court did note its agreement with an expert witness who testified “that punitive isolation as it exists at Cummins today serves no rehabilitative purpose, and that it is counterproductive.” Id., at 277. The court went on to say that punitive isolation “makes bad men worse. It must be changed.” Ibid. We agree with the Department’s contention that the Constitution does not require that every aspect of prison discipline serve a rehabilitative purpose. Novak v. Beto, 453 F. 2d 661, 670-671 (CA5 1971); Nadeau v. Helgemoe, 561 F. 2d 411, 415-416 (CAI 1977). But the District Court did not impose a new legal test. Its remarks form the transition from a detailed description of conditions in the isolation cells to a traditional legal analysis of those conditions. The quoted passage simply summarized the facts and presaged the legal conclusion to come. HUTTO v. FINNEY 687 678 Opinion Of the Court “grue” might be tolerable for a few days and intolerably cruel for weeks or months. The question before the trial court was whether past constitutional violations had been remedied. The court was entitled to consider the severity of those violations in assessing the constitutionality of conditions in the isolation cells. The court took note of the inmates’ diet, the continued overcrowding, the rampant violence, the vandalized cells, and the “lack of professionalism and good judgment on the part of maximum security personnel.” 410 F. Supp., at 277 and 278. The length of time each inmate spent in isolation was simply one consideration among many. We find no error in the court’s conclusion that, taken as a whole, conditions in the isolation cells continued to violate the prohibition against cruel and unusual punishment. In fashioning a remedy, the District Court had ample authority to go beyond earlier orders and to address each element contributing to the violation. The District Court had given the Department repeated opportunities to remedy the cruel and unusual conditions in the isolation cells. If petitioners had fully complied with the court’s earlier orders, the present time limit might well have been unnecessary. But taking the long and unhappy history of the litigation into account, the court was justified in entering a comprehensive order to insure against the risk of inadequate compliance.9 9 As we explained in Milliken v. Bradley, 433 U. S. 267, 281, state and local authorities have primary responsibility for curing constitutional violations. “If, however ‘[those] authorities fail in their affirmative obligations . . . judicial authority may be invoked.’ Swann [v. Charlotte-Mecklenburg Board of Education, 402 U. S. 1,] 15. Once invoked, ‘the scope of a district court’s equitable powers to remedy past wrongs is broad, for breadth and flexibility are inherent in equitable remedies.’ ” Ibid. In this case, the District Court was not remedying the present effects of a violation in the past. It was seeking to bring an ongoing violation to an immediate halt. Cooperation on the part of Department officials and compliance with other aspects of the decree may justify elirni- 688 OCTOBER TERM, 1977 Opinion of the Court 437 U. S. The order is supported by the interdependence of the conditions producing the violation. The vandalized cells and the atmosphere of violence were attributable, in part, to overcrowding and to deep-seated enmities growing out of months of constant daily friction.10 The 30-day limit will help to correct these conditions.11 Moreover, the limit presents little danger of interference with prison administration, for the Commissioner of Correction himself stated that prisoners should not ordinarily be held in punitive isolation for more than 14 days. Id., at 278. Finally, the exercise of discretion in this case is entitled to special deference because of the trial judge’s years of experience with the problem at hand and his recognition of the limits on a federal court’s authority in a case of this kind.12 Like the Court of Appeals, we find no error in the inclusion of a 30-day limitation on sentences to punitive isolation as a part of the District Court’s comprehensive remedy. nation of this added safeguard in the future, but it is entirely appropriate for the District Court to postpone any such determination until the Department’s progress can be evaluated. w The District Court noted “that as a class the inmates of the punitive cells hate those in charge of them, and that they may harbor particular hatreds against prison employees who have been in charge of the same inmates for a substantial period of time.” 410 F. Supp., at 277. 11 As early as 1969, the District Court had identified shorter sentences as a possible remedy for overcrowding in the isolation cells. Holt I, 300 F. Supp., at 834. The limit imposed in 1976 was a mechanical—and therefore an easily enforced—method of minimizing overcrowding, with its attendant vandalism and unsanitary conditions. 12 See, e. g., Holt II, 309 F. Supp., at 369: “The Court, however, is limited in its inquiry to the question of whether or not the constitutional rights of inmates are being invaded and with whether the Penitentiary itself is unconstitutional. The Court is not judicially concerned with questions which in the last analysis are addressed to legislative and administrative judgment. A practice that may be bad from the standpoint of penology may not necessarily be forbidden by the Constitution.” HUTTO v. FINNEY 689 678 Opinion of the Court II The Attorney General of Arkansas, whose office has represented petitioners throughout this litigation, contends that any award of fees is prohibited by the Eleventh Amendment. He also argues that the Court of Appeals incorrectly held that fees were authorized by the Civil Rights Attorney’s Fees Awards Act of 1976. We hold that the District Court’s award is adequately supported by its finding of bad faith and that the Act supports the additional award by the Court of Appeals. A. The District Court Award Although the Attorney General argues that the finding of bad faith does not overcome the State’s Eleventh Amendment protection, he does not question the accuracy of the finding made by the District Court and approved by the Court of Appeals.13 Nor does he question the settled rule that a losing litigant’s bad faith may justify an allowance of fees to the prevailing party.14 He merely argues that the order requir 13 In affirming the award, the Court of Appeals relied chiefly on the Civil Rights Attorney’s Fees Awards Act of 1976, but it also noted expressly that “the record fully supports the finding of the district court that the conduct of the state officials justified the award under the bad faith exception enumerated in Alyeska [Pipeline Service Co. v. Wilderness Society, 421 U. S. 240].” 548 F. 2d 740, 742 n. 6. 14 An equity court has the unquestioned power to award attorney’s fees against a party who shows bad faith by delaying or disrupting the litigation or by hampering enforcement of a court order. Alyeska Pipeline Service Co. v. Wilderness Society, 421 U. S. 240, 258-259; Christiansburg Garment Co. v. EEOC, 434 U. S. 412; Straub v. Vaisman & Co., Inc., 540 F. 2d 591, 598-600 (CA3 1976); cf. Fed. Rule Civ. Proc. 56 (g) (attorney’s fees to be awarded against party filing summary judgment affidavits “in bad faith or solely for the purpose of delay”); Fed. Rule Civ. Proc. 37 (a) (4) (motions to compel discovery; prevailing party may recover attorney’s fees). The award vindicates judicial authority without resort to the more drastic sanctions available for contempt of court and makes the prevailing party whole for expenses caused by his opponent’s obstinacy. Cf. First Nat. Bank v. Dunham, 471 F. 2d 712 (CA8 1973). Of course, fees can also be awarded as part of a civil contempt penalty. See, e. g., Toledo Scale Co. 690 OCTOBER TERM, 1977 Opinion of the Court 437U.S. ing that the fees be paid from public funds violates the Eleventh Amendment. In the landmark decision in Ex parte Young, 209 U. S. 123, the Court held that, although prohibited from giving orders directly to a State, federal courts could enjoin state officials in their official capacities. And in Edelman v. Jordan, 415 U. S. 651, when the Court held that the Amendment grants the States an immunity from retroactive monetary relief, it reaffirmed the principle that state officers are not immune from prospective injunctive relief. Aware that the difference between retroactive and prospective relief “will not in many instances be that between day and night,” id., at 667, the Court emphasized in Edelman that the distinction did not immunize the States from their obligation to obey costly federal-court orders. The cost of compliance is “ancillary” to the prospective order enforcing federal law. Id., at 668.15 The line between retroactive and prospective relief cannot be so rigid that it defeats the effective enforcement of prospective relief. The present case requires application of that principle. In exercising their prospective powers under Ex parte Young and Edelman v. J or dan, federal courts are not reduced to issuing injunctions against state officers and hoping for compliance. Once issued, an injunction may be enforced. Many of the court’s most effective enforcement weapons involve financial penalties. A criminal contempt prosecution for “resistance to [the court’s] lawful . . . order” may result in a jail term or a fine. 18 U. S. C. § 401 (1976 ed.). Civil contempt proceedings may yield a conditional jail term or fine. United States v. v. Computing Scale Co., 261 U. S. 399; Signal Delivery Service, Inc. v. Highway Truck Drivers, 68 F. R. D. 318 (ED Pa. 1975). 15 “Ancillary” costs may be very large indeed. Last Term, for example, this Court rejected an Eleventh Amendment defense and approved an injunction ordering a State to pay almost $6 million to help defray the costs of desegregating the Detroit school system. Milliken v. Bradley, 433 U. S., at 293 (Powell, J., concurring in judgment). HUTTO v. FINNEY 691 678 Opinion of the Court Mine Workers, 330 U. S. 258, 305. Civil contempt may also be punished by a remedial fine, which compensates the party who won the injunction for the effects of his opponent’s noncompliance. Id., at 304; Gompers v. Bucks Stove & Range Co., 221 U. S. 418. If a state agency refuses to adhere to a court order, a financial penalty may be the most effective means of insuring compliance. The principles of federalism that inform Eleventh Amendment doctrine surely do not require federal courts to enforce their decrees only by sending high state officials to jail.16 The less intrusive power to impose a fine is properly treated as ancillary to the federal court’s power to impose injunctive relief. In this case, the award of attorney’s fees for bad faith served the same purpose as a remedial fine imposed for civil contempt. It vindicated the District Court’s authority over a recalcitrant litigant. Compensation was not the sole motive for the award; in setting the amount of the fee, the court said that it would “make no effort to adequately compensate counsel for the work that they have done or for the time that they have spent on the case.” 410 F. Supp., at 285. The court did allow a “substantial” fee, however, because “the allowance thereof may incline the Department to act in such a manner that further protracted litigation about the prisons will not be necessary.” Ibid.11 We see no reason to distin 16 See Note, Attorneys’ Fees and the Eleventh Amendment, 88 Harv. L. Rev. 1875, 1892 (1975). 17 That the award had a compensatory effect does not in any event distinguish it from a fine for civil contempt, which also compensates a private party for the consequences of a contemnor’s disobedience. Gompers v. Bucks Stove & Range Co., 221 U. S. 418. Moreover, the Court has approved federal rulings requiring a State to support programs that compensate for past misdeeds, saying: “That the programs are also ‘compensatory’ in nature does not change the fact that they are part of a plan that operates prospectively to bring about the delayed benefits of a unitary school system. We therefore hold that such prospective relief is not barred by the Eleventh Amendment.” Milliken v. Bradley, supra, at 290 (emphasis in original). The award of attorney’s fees against a State disregarding a 692 OCTOBER TERM, 1977 Opinion of the Court 437U.S. guish this award from any other penalty imposed to enforce a prospective injunction.18 Hence the substantive protections of the Eleventh Amendment do not prevent an award of attorney’s fees against the Department’s officers in their official capacities. Instead of assessing the award against the defendants in their official capacities, the District Court directed that the fees are “to be paid out of Department of Correction funds.” Ibid. Although the Attorney General objects to the form of the order,19 no useful purpose would be served by requiring that it be recast in different language. We have previously approved directives that were comparable in their actual impact on the State without pausing to attach significance to the language used by the District Court.20 Even if it might have federal order stands on the same footing; like other enforcement powers, it is integral to the court’s grant of prospective relief. 18 The Attorney General has not argued that this award was so large or so unexpected that it interfered with the State’s budgeting process. Although the Eleventh Amendment does not prohibit attorney’s fees awards for bad faith, it may counsel moderation in determining the size of the award or in giving the State time to adjust its budget before paying the full amount of the fee. Cf. Edelman v. Jordan, 415 U. S. 651, 666 n. 11. In this case, however, the timing of the award has not been put in issue; nor has the State claimed that the award was larger than necessary to enforce the court’s prior orders. 10 We do not understand the Attorney General to urge that the fees should have been awarded against the officers personally; that would be a remarkable way to treat individuals who have relied on the Attorney General to represent their interests throughout this litigation. 20 In Milliken n. Bradley, supra, we affirmed an order requiring a state treasurer to pay a substantial sum to another litigant, even though the District Court’s opinion explicitly recognized that “this remedial decree will be paid for by the taxpayers of the City of Detroit and the State of Michigan,” App. to Pet. for Cert, in Milliken n. Bradley, 0. T. 1976, No. 76-447, pp. 116a-117a, and even though the Court of Appeals, in affirming, stated that “the District Court ordered that the State and Detroit Board each pay one-half the costs” of relief. Bradley v. Milliken, 540 F. 2d 229, 245 (CA6 1976). HUTTO v. FINNEY 693 678 Opinion of the Court been better form to omit the reference to the Department of Correction, the use of that language is surely not reversible error. B. The Court of Appeals Award Petitioners, as the losing litigants in the Court of Appeals, were ordered to pay an additional $2,500 to counsel for the prevailing parties “for their services on this appeal.” 548 F. 2d, at 743. The order does not expressly direct the Department of Correction to pay the award, but since petitioners are sued in their official capacities, and since they are represented by the Attorney General, it is obvious that the award will be paid with state funds. It is also clear that this order is not supported by any finding of bad faith. It is founded instead on the provisions of the Civil Rights Attorney’s Fees Awards Act of 1976. Pub. L. No. 94-559, 90 Stat. 2641, 42 U. S. C. § 1988 (1976 ed.). The Act declares that, in suits under 42 U. S. C. § 1983 and certain other statutes, federal courts may award prevailing parties reasonable attorney’s fees “as part of the costs.”21 As this Court made clear in Fitzpatrick v. Bitzer, 427 U. S. 445, Congress has plenary power to set aside the States’ immunity from retroactive relief in order to enforce the Fourteenth Amendment. When it passed the Act, Congress undoubtedly intended to exercise that power and to authorize fee awards 21 The Act declares: “In any action or proceeding to enforce a provision of §§ 1977, 1978, 1979, 1980, and 1981 of the Revised Statutes [42 U. S. C. §§ 1981-1983, 1985, 1986], title IX of Public Law 92-318 [20 U. S. C. § 1681 et seq. (1976 ed.)], or in any civil action or proceeding, by or on behalf of the United States of America, to enforce, or charging a violation of, a provision of the United States Internal Revenue Code [26 U. S. C. § 1 et seq. (1976 ed.)], or title VI of the Civil Rights Act of 1964 [42 U. S. C. §2000d et seg.], 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.” 90 Stat. 2641. 694 OCTOBER TERM, 1977 Opinion of the Court 437U.S. payable by the States when their officials are sued in their official capacities. The Act itself could not be broader. It applies to “any” action brought to enforce certain civil rights laws. It contains no hint of an exception for States defending injunction actions; indeed, the Act primarily applies to laws passed specifically to restrain state action. See, e. g., 42 U. S. C. § 1983. The legislative history is equally plain: “[I]t is intended that the attorneys’ fees, like other items of costs, will be collected either directly from the official, in his official capacity, from funds of his agency or under his control, or from the State or local government (whether or not the agency or government is a named party).” S. Rep. No. 94—1011, p. 5 (1976) (footnotes omitted). The House Report is in accord: “The greater resources available to governments provide an ample base from which fees can be awarded to the prevailing plaintiff in suits against governmental officials or entities.” H. R. Rep. No. 94r-1558, p. 7 (1976). The Report adds in a footnote that: “Of course, the 11th Amendment is not a bar to the awarding of counsel fees against state governments. Fitzpatrick v. Bitzer.” Id., at 7 n. 14. Congress’ intent was expressed in deeds as well as words. It rejected at least two attempts to amend the Act and immunize state and local governments from awards.22 The Attorney General does not quarrel with the rule established in Fitzpatrick v. Bitzer, supra. Rather, he argues that these plain indications of legislative intent are not enough. In his view, Congress must enact express statutory language making the States liable if it wishes to abrogate their immunity.23 The Attorney General points out that this Court has 22See 122 Cong. Rec. 31832-31835 (1976) (amendment of Sen. Helms); id., at 32296 and 32396-32397 (amendment of Sen. Allen). See aso id., at 32931 (amendment of Sen. William Scott). 23 The Attorney General also contends that the fee award should not apply to cases, such as this one, that were pending when the Act was passed HUTTO v. FINNEY 695 678 Opinion of the Court sometimes refused to impose retroactive liability on the States in the absence of an extraordinarily explicit statutory mandate. See Employees v. Missouri Public Health & Welfare Dept., 411 U. S. 279; see also Edelman v. Jordan, 415 U. S. 651. But these cases concern retroactive liability for prelitigation conduct rather than expenses incurred in litigation seeking only prospective relief. The Act imposes attorney’s fees “as part of the costs.” Costs have traditionally been awarded without regard for the States’ Eleventh Amendment immunity. The practice of awarding costs against the States goes back to 1849 in this Court. See Missouri v. Iowa, I How. 660, 681; North Dakota v. Minnesota, 263 U. S. 583 (collecting cases). The Court has never viewed the Eleventh Amendment as barring such awards, even in suits between States and individual litigants.24 in 1976. But the legislative history of the Act, as well as this Court’s general practice, defeats this argument. The House Report declared: ‘Tn accordance with applicable decisions of the Supreme Court, the bill is intended to apply to all cases pending on the date of enactment . . . .” H. R. Rep. No. 94-1558, p. 4 n. 6 (1976). See also Bradley n. Richmond School Board, 416 U. S. 696. 24 While the decisions allowing the award of costs against States antedate the line drawn between retroactive and prospective relief in Edelman v. Jordan, 415 U. S. 651, such awards do not seriously strain that distinction. Unlike ordinary “retroactive” relief such as damages or restitution, an award of costs does not compensate the plaintiff for the injury that first brought him into court. Instead, the award reimburses him for a portion of the expenses he incurred in seeking prospective relief. (An award of costs will almost invariably be incidental to an award of prospective relief, for costs are generally awarded only to prevailing parties, see Fed. Rule Civ. Proc. 54 (d), and only prospective relief can be successfully pursued by an individual in a suit against a State.) Moreover, like the power to award attorney’s fees for litigating in bad faith, the power to assess costs is an important and well-recognized tool used to restrain the behavior of parties during litigation. See, e. g., Rule 37 (b) (costs may be awarded for failure to obey discovery order); Rule 30 (g) (costs may be awarded for failure to attend deposition or for failure to serve subpoena). When 696 OCTOBER TERM, 1977 Opinion of the Court 437U.S. In Fairmont Creamery Co. v. Minnesota, 275 U. S. 70, the State challenged this Court’s award of costs, but we squarely rejected the State’s claim of immunity. Far from requiring an explicit abrogation of state immunity, we relied on a statutory mandate that was entirely silent on the question of state liability.25 The power to make the award was supported by “the inherent authority of the Court in the orderly administration of justice as between all parties litigant.” Id., at 74. A federal court’s interest in orderly, expeditious proceedings “justifies [it] in treating the state just as any other litigant and in imposing costs upon it” when an award is called for. Id., at 77.2* Just as a federal court may treat a State like any other litigant when it assesses costs, so also may Congress amend its definition of taxable costs and have the amended class of costs apply to the States, as it does to all other litigants, without expressly stating that it intends to abrogate the States’ Eleventh Amendment immunity. For it would be absurd to require an express a State defends a suit for prospective relief, it is not exempt from the ordinary discipline of the courtroom. 25 “If specific statutory authority [for an award of costs] is needed, it is found in § 254 of the Judicial Code .... It provides that there shall be ‘taxed against the losing party in each and every cause pending in the Supreme Court’ the cost of printing the record, except when the judgment is against the United States. This exception of the United States in the section with its emphatic inclusion of every other litigant shows that a state as litigant must pay the costs of printing, if it loses, in every case, civil or criminal. These costs constitute a large part of all the costs. The section certainly constitutes pro tanto statutory authority to impose costs generally against a state if defeated.” 275 U. S., at 77. 26 Because the interest in orderly and evenhanded justice is equally pressing in lower courts, Fairmont Creamery has been widely understood as foreclosing any Eleventh Amendment objection to assessing costs against a State in all federal courts. See, e. g., Skehan v. Board of Trustees, 538 F. 2d 53, 58 (CA3 1976) (en banc); Utah v. United States, 304 F. 2d 23 (CAIO 1962); United States ex rel. Griffin v. McMann, 310 F. Supp. 72 (EDNY 1970). HUTTO v. FINNEY 697 678 Opinion of the Court reference to state litigants whenever a filing fee, or a new item, such as an expert witness’ fee, is added to the category of taxable costs.27 There is ample precedent for Congress’ decision to authorize an award of attorney’s fees as an item of costs. In England, costs “as between solicitor and client,” Sprague v. Ticonic Nat. Bank, 307 U. S. 161, 167, are routinely taxed today, and have been awarded since 1278. Alyeska Pipeline Service Co. v. Wilderness Society, 421 U. S. 240,247 n. 18. In America, although fees are not routinely awarded, there are a large number of statutory and common-law situations in which allowable costs include counsel fees.28 Indeed, the federal statutory definition of costs, which was enacted before the Civil War and which remains in effect today, includes certain fixed attorney’s fees as recoverable costs.29 In Fairmont Creamery itself, the Court awarded these statutory attorney’s fees against the 27 This conclusion is consistent with the reasons for requiring a formal indication of Congress’ intent to abrogate the States’ Eleventh Amendment immunity. The requirement insures that Congress has not imposed “enormous fiscal burdens on the States” without careful thought. Employees v. Missouri Public Health & Welfare Dept., 411 U. S. 279, 284. See Tribe, Intergovernmental Immunities in Litigation, Taxation and Regulation, 89 Harv. L. Rev. 682, 695 (1976). But an award of costs—limited as it is to partially compensating a successful litigant for the expense of his suit—could hardly create any such hardship for a State. Thus we do not suggest that our analysis would be the same if Congress were to expand the concept of costs beyond the traditional category of litigation expenses. 28 In 1975, we listed 29 statutes allowing federal courts to award attorney’s fees in certain suits. See Alyeska Pipeline Service Co. v. Wilderness Society, 421 U. S., at 260-261, n. 33. Some of these statutes define attorney’s fees as an element of costs, while others separate fees from other taxable costs. Compare 42 U. S. C. § 2000a-3 (b) with 29 U. S. C. § 216(b) (1970 ed., Supp. V). 29 See 28 U. S. C. § 1923 (a) ($100 in fees for admiralty appeals involving more than $5,000). Inflation has now made the awards merely nominal, but the principle of allowing such awards against all parties has undiminished force. 698 OCTOBER TERM, 1977 Opinion of the Court 437U.S. State of Minnesota along with other taxable costs,30 even though the governing statute said nothing about state liability. It is much too late to single out attorney’s fees as the one kind of litigation cost whose recovery may not be authorized by Congress without an express statutory waiver of the States’ immunity.31 30 File of the Clerk of this Court in Fairmont Creamery Co. v. Minnesota, 0. T. 1926, No. 725. 31 The Attorney General argues that the statute itself must expressly abrogate the States’ immunity from retroactive liability, relying on Employees v. Missouri Public Health & Welfare Dept., supra. Even if we were not dealing with an item such as costs, this reliance would be misplaced. In Employees, the Court refused to permit individual backpay suits against state institutions because the Court "found not a word in the history of the [statute] to indicate a purpose of Congress to make it possible for a citizen of that State or another State to sue the State in the federal courts.” 411 U. S., at 285. The Court was careful to add, moreover, that its reading of the law did not make the statute’s inclusion of state institutions meaningless. Because the Secretary of Labor was empowered to bring suit against violators, the amendment covering state institutions gave him authority to enforce the statute against them. Id., at 285-286. The present Act, in contrast, has a history focusing directly on the question of state liability; Congress considered and firmly rejected the suggestion that States should be immune from fee awards. Moreover, the Act is not part of an intricate regulatory scheme offering alternative methods of obtaining relief. If the Act does not impose liability for attorney’s fees on the States, it has no meaning with respect to them. Finally, the claims asserted in Employees and in Edelman n. Jordan, 415 U. S. 651, were based on a statute rooted in Congress’ Art. I power. See Employees, supra, at 281 (claim based on Fair Labor Standards Act, 29 U. S. C. § 201 et seq.) ; Edelman v. Jordan, supra, at 674 (underlying claim based on Social Security Act provisions dealing with aid to aged, blind, and disabled, 42 U. S. C. §§ 1381-1385). In this case, as in Fitzpatrick n. Bitzer, 427 U. S. 445, the claim is based on a statute enacted to enforce the Fourteenth Amendment. As we pointed out in Fitzpatrick: “[T]he Eleventh Amendment, and the principle of state sovereignty which it embodies ... are necessarily limited by the enforcement provisions of § 5 of the Fourteenth Amendment. . . . When Congress acts pursuant to HUTTO v. FINNEY 699 678 Opinion of the Court Finally, the Attorney General argues that, even if attorney’s fees may be awarded against a State, they should not be awarded in this case, because neither the State nor the Department is expressly named as a defendant. Although the Eleventh Amendment prevented respondents from suing the State by name, their injunctive suit against prison officials was, for all practical purposes, brought against the State. The actions of the Attorney General himself show that. His office has defended this action since it began. See Holt I, 300 F. Supp., at 826. The State apparently paid earlier fee awards; and it was the State’s lawyers who decided to bring this appeal, thereby risking another award.32 § 5, not only is it exercising legislative authority that is plenary within the terms of the constitutional grant, it is exercising that authority under one section of a constitutional Amendment whose other sections by their own terms embody limitations on state authority.” Id., at 456. Cf. National League of Cities v. Usery, 426 U. S. 833, 852 n. 17. Applying the standard appropriate in a case brought to enforce the Fourteenth Amendment, we have no doubt that the Act is clear enough to authorize the award of attorney’s fees payable by the State. 32 The Attorney General is hardly in a position to argue that the fee awards should be borne not by the State, but by individual officers who have relied on his office to protect their interests throughout the litigation. Nonetheless, our dissenting Brethren would apparently force these officers to bear the award alone. The Act authorizes an attorney’s fee award even though the appeal was not taken in bad faith; no one denies that. The Court of Appeals’ award is thus proper, and the only question is who will pay it. In the dissenters’ view, the Eleventh Amendment protects the State from liability. But the State’s immunity does not extend to the individual officers. The dissenters would apparently leave the officers to pay the award; whether the officials would be reimbursed is a decision that “may . . . safely be left to the State involved.” Post, at 716 (Rehnquist, J., dissenting). This is manifestly unfair when, as here, the individual officers have no personal interest in the conduct of the State’s litigation, and it defies this Court’s insistence in a related context that imposing personal liability in the absence of bad faith may cause state officers to “exercise their discretion with undue timidity.” Wood v. Strickland, 420 U. S. 308, 321. 700 OCTOBER TERM, 1977 Brennan, J., concurring 437U.S. Like the Attorney General, Congress recognized that suits brought against individual officers for injunctive relief are for all practical purposes suits against the State itself. The legislative history makes it clear that in such suits attorney’s fee awards should generally be obtained “either directly from the official, in his official capacity, from funds of his agency or under his control, or from the State or local government (whether or not the agency or government is a named party).” S. Rep. No. 94-1011, p. 5 (1976). Awards against the official in his individual capacity, in contrast, were not to be affected by the statute; in injunctive suits they would continue to be awarded only “under the traditional bad faith standard recognized by the Supreme Court in Aly e ska.” Id., at 5 n. 7. There is no indication in this case that the named defendants litigated in bad faith before the Court of Appeals. Consequently, the Department of Correction is the entity intended by Congress to bear the burden of the counsel-fees award. The judgment of the Court of Appeals is accordingly affirmed. It is so ordered. Mr. Justice Brennan, concurring. I join fully in the opinion of the Court and write separately only to answer points made by Mr. Justice Powell. I agree with the Court that there is no reason in this case to decide more than whether 42 U. S. C. § 1988 (1976 ed.), itself authorizes awards of attorney’s fees against the States. Mr. Justice Powell takes the view, however, that unless 42 U. S. C. § 1983 also authorizes damages awards against the States, the requirements of the Eleventh Amendment are not met. Citing Edelman v. Jordan, 415 U. S. 651 (1974), he concludes that § 1983 does not authorize damages awards against the State and, accordingly, that § 1988 does not either. There are a number of difficulties with this syllogism, but the most striking is its reliance on Edelman v. Jordan, a case whose foundations would seem to have been seriously under HUTTO v. FINNEY 701 678 Brennan, J., concurring mined by our later holdings in Fitzpatrick v. Bitzer, 427 U. S. 445 (1976), and Monell v. New York City Dept, of Social Services, 436 U. S. 658 (1978). It cannot be gainsaid that this Court in Edelman rejected the argument that 42 U. S. C. § 1983 “was intended to create a waiver of a State’s Eleventh Amendment immunity merely because an action could be brought under that section against state officers, rather than against the State itself.” 415 U. S., at 676-677. When Edelman was decided, we had affirmed monetary awards against the States only when they had consented to suit or had waived their Eleventh Amendment immunity. See, e. g., Petty v. Tennessee-Missouri Bridge Comm’n, 359 U. S. 275 (1959); Par den v. Terminal R. Co., 377 U. S. 184 (1964); Employees v. Missouri Public Health & Welfare Dept., 411 U. S. 279 (1973). In Edelman, we summarized the rule of our cases as follows: The “question of waiver or consent under the Eleventh Amendment was found in [our] cases to turn on whether Congress had intended to abrogate the immunity in question, and whether the State by its participation in [a regulated activity] authorized by Congress had in effect consented to the abrogation of [Eleventh Amendment] immunity.” 415 U. S., at 672. At the very least, such consent could not be found unless Congress had authorized suits against “a class of defendants which literally includes States.” Ibid. It was a short jump from that proposition, to the conclusion that § 1983—which was then thought to include only natural persons among those who could be party defendants, see Monroe v. Pape, 365 U. S. 167, 187-191 (1961)—was not in the class of statutes that might lead to a waiver of Eleventh Amendment immunity. This is best summed up by Mr. Justice Rehnquist, the author of Edelman, in his opinion for the Court in Fitzpatrick v. Bitzer, supra: “We concluded that none of the statutes relied upon by plaintiffs in Edelman contained any authorization by 702 OCTOBER TERM, 1977 Brennan, J., concurring 437U.S. Congress to join a State as defendant. The Civil Rights Act of 1871, 42 U. S. C. § 1983, had been held in Monroe v. Pape, 365 U. S. 167, 187-191 (1961), to exclude cities and other municipal corporations from its ambit; that being the case, it could not have been intended to include States as parties defendant.” 427 U. S., at 452. But time has not stood still. Two Terms ago, we decided Fitzpatrick v. Bitzer, which for the first time in the recent history of the Court asked us to decide “the question of the relationship between the Eleventh Amendment and the enforcement power granted to Congress under § 5 of the Fourteenth Amendment.”1 Id., at 456. There we concluded that “the Eleventh Amendment, and the principle of state sovereignty which it embodies, . . . are necessarily limited by the enforcement provisions of § 5 of the Fourteenth Amendment.” Ibid. (Citation omitted.) And we went on to hold: “Congress may, in determining what is ‘appropriate legislation’ for the purpose of enforcing the provisions of the Fourteenth Amendment, provide for private suits against States or state officials which are constitutionally impermissible in other contexts.” Ibid. Then, in Monell v. New York City Dept, of Social Services, supra, decided only weeks ago, we held that the Congress which passed the Civil Rights Act of 1871, now § 1983—a statute enacted pursuant to § 5 of the Fourteenth Amendment, see 436 U. S., at 665—“did intend municipalities and other local government units to be included among those persons to whom § 1983 applies.” Id., at 690. This holding alone would appear to be enough to vitiate the vitality of Fitzpatrick’s explanation of Edelman.2 1 As Fitzpatrick noted, this issue had been before the Court in Ex parte Virginia, 100 U. S. 339 (1880). 2 It can also be questioned whether, had Congress meant to exempt municipalities from liability under § 1983, it would necessarily follow that HUTTO v. FINNEY 703 678 Brennan, J., concurring Moreover, central to the holding in Monell was the conclusion that the Act of Feb. 25, 1871, ch. 71, § 2, 16 Stat. 431, provided a definition of the word “person” used to describe the class of defendants in § 1983 suits. 436 U. S., at 688. Although we did not in Monell have to consider whether § 1983 as properly construed makes States liable in damages for their constitutional violations, the conclusion seems inescapable that, at the very least, § 1983 includes among possible defendants “a class . . . which literally includes States.” Edelman v. Jordan, 415 U. S., at 672. This follows immediately from the language of the Act of Feb. 25,1871: “[I]n all acts hereafter passed . . . the word ‘person’ may extend and be applied to bodies politic and corporate . . . unless the context shows that such words were intended to be used in a more limited sense . . . .” The phrase “bodies politic and corporate” is now, and certainly would have been in 1871, a synonym for the word “State.” See, e. g., United States v. Maurice, 26 F. Cas. 1211, 1216 (No. 15,747) (CC Va. 1823) (Marshall, C. J.) (“The United States is a government and, consequently, a body politic and corporate”). See also Pfizer Inc. v. Government of India, 434 U. S. 308 (1978). Given our holding in Monell, the essential premise of our Edelman holding—that no statute involved in Edelman authorized suit against “a class of defendants which literally includes States,” 415 U. S., at 672—would clearly appear to be no longer true. Moreover, given Fitzpatrick’s holding that Congress has plenary power to make States liable in damages when it acts pursuant to § 5 of the Fourteenth Amendment, it is surely at least an open question whether § 1983 properly construed does not make the States liable for relief of all kinds, notwithstanding the Eleventh Amendment. Whether this is Congress also meant to exempt States. See Monell y. New York City Dept, of Social Services, 436 U. S. 658, 673-674, n. 30 (1978). 704 OCTOBER TERM, 1977 Opinion of Powell, J. 437 U. S. in fact so, must of course await consideration in an appropriate case.3 Mr. Justice Powell, with whom The Chief Justice joins, concurring in part and dissenting in part.* While I join Parts I1 and II-A of the Court’s opinion, I cannot subscribe to Part II-B’s reading of the Eleventh Amendment as permitting counsel-fee awards against the State on the authority of a statute that concededly does not effect “an express statutory waiver of the States’ immunity.” Ante, at 698. Edelman v. Jordan, 415 U. S. 651, 676-677 (1974), rejected the argument that 42 U. S. C. § 1983 “was intended to create a waiver of the State’s Eleventh Amendment immunity merely because an action could be brought under that section against state officers, rather than against the State itself.” In a § 1983 3 As I understand Mr. Justice Powell’s objection to the Court’s opinion, it rests squarely on the proposition that a clear statement to make States liable for damages cannot be found in legislative history but only on the face of a statute. See post, at 705-706. In § 1983 and the Act of Feb. 25, 1871, we have a statute that on its face applies to state defendants, but now Mr. Justice Powell tells us that this is not enough because there is still an absence of “congressional purpose in 1871 to abrogate the protections of the Eleventh Amendment.” Post, at 709 n. 6. I suppose that this means either that no statute can meet the Eleventh Amendment clear-statement test or, alternatively, that Mr. Justice Powell has some undisclosed rule as to when legislative history may be taken into account that works only to defeat state liability. *Mr. Justice White and Mr. Justice Rehnquist join this opinion to the extent it dissents from the opinion and judgment of the Court. 1 The principles emphasized by Mr. Justice Rehnquist, post, at 711, as to the limitation of equitable remedies are settled. See DayUm Board of Education v. Brinkman, 433 U. S. 406 (1977); Milliken n. Bradley, 433 U. S. 267 (1977). On the extraordinary facts of this case, however, I agree with the Court that the 30-day limitation on punitive isolation was within the bounds of the District Court’s discretion in fashioning appropriate relief. It also is evident from the Court’s opinion, see ante, at 688, that this limitation will have only a minimal effect on prison administration, an area of responsibility primarily reserved to the States. HUTTO v. FINNEY 705 678 Opinion of Powell, J. action “a federal court’s remedial power, consistent with the Eleventh Amendment, is necessarily limited to prospective injunctive relief, . . . and may not include a retroactive award which requires the payment of funds from the state treasury.” 415 U. S., at 677 (citations omitted). There is no indication in the language of the Civil Rights Attorney’s Fees Awards Act of 1976 (Act), Pub. L. No. 94—559, 90 Stat. 2641, 42 U. S. C. § 1988 (1976 ed.), that Congress sought to overrule that holding.2 In this case, as in Edelman, “the threshold fact of congressional authorization to sue a class of defendants which literally includes States is wholly absent.” 415 U. S., at 672 (emphasis supplied). Absent such authorization, grounded in statutory language sufficiently clear to alert every voting Member of Congress of the constitutional implications of particular legislation, we undermine the values of federalism served by the Eleventh Amendment by inferring from congressional silence an intent to “place new or even enormous fiscal burdens on the States.” Employees v. Missouri Public Health & Welfare Dept., 411 U. S. 279, 284 (1973). The Court notes that the Committee Reports and the defeat of two proposed amendments indicate a purpose to authorize counsel-fee awards against the States. Ante, at 694. That evidence might provide persuasive support for a finding of “waiver” if this case involved “a congressional enactment which by its terms authorized suit by designated plaintiffs against a general class of defendants which literally included 2 In Mondi v. New York City Dept, of Social Services, 436 U. S. 658 (1978), the Court held that “the legislative history of the Civil Rights Act of 1871 compels the conclusion that Congress did intend municipalities and other local government units to be included among those persons to whom § 1983 applies.” Id., at 690. We noted, however, that there was no “basis for concluding that the Eleventh Amendment is a bar to municipal liability,” and that our holding was “limited to local government units which are not considered part of the State for Eleventh Amendment purposes.” Id., at 690, and n. 54 (emphasis in original). 706 OCTOBER TERM, 1977 Opinion of Powell, J. 437U.S. States or state instrumentalities.” Edelman, supra, at 672. Compare Fitzpatrick v. Bitzer, 427 U. S. 445, 452 (1976), with Employees, supra, at 283, 284-2S5.3 But in this sensitive area of conflicting interests of constitutional dimension, we should not permit items of legislative history to substitute for explicit statutory language. The Court should be “hesitant to presume general congressional awareness,” SEC v. Sloan, 436 U. S. 103, 121 (1978), of Eleventh Amendment consequences of a statute that does not make express provision for monetary recovery against the States.4 3 Although Fitzpatrick states that the “prerequisite” of “congressional authorization ... to sue the State as employer” was found “wanting in Employees,” 427 U. S., at 452, this reference is to the Court’s conclusion in Employees that notwithstanding the literal inclusion of the States as statutory employers, in certain contexts, there was “not a word in the history of the [statute] to indicate a purpose of Congress to make it possible for a citizen of that State or another State to sue the State in the federal courts.” 411 U. S., at 285. See Edelman, 415 U. S., at 672. While it has been suggested that “[t]he legislative changes that made state governments liable under Title VII closely paralleled the changes that made state governments liable under the Fair Labor Standards Act,” Baker, Federalism and the Eleventh Amendment, 48 U. Colo. L. Rev. 139, 171 n. 152 (1977), comparing Fitzpatrick, 427 U. S., at 449 n. 2, with Employees, 411 U. S., at 282-283, the statute considered in Fitzpatrick made explicit reference to the availability of a private action against state and local governments in the event the Equal Employment Opportunity Commission or the Attorney General failed to bring suit or effect a conciliation agreement. Equal Opportunity Employment Act of 1972, 86 Stat. 104, 42 U. S. C. §2000e-5 (f)(1) (1970 ed., Supp. V); see H. R. Rep. No. 92-238, pp. 17-19 (1971); S. Rep. No. 92-415, pp. 9-11 (1971); S. Conf. Rep. No. 92-681, pp. 17-18 (1972); H. R. Conf. Rep. No. 92-899, pp. 17-18 (1972). 4 “By making a law unenforceable against the states unless a contrary intent were apparent in the language of the statute, the clear statement rule . . . ensure [s] that attempts to limit state power [are] unmistakable, thereby structuring the legislative process to allow the centrifugal forces in Congress the greatest opportunity to protect the states’ interests.” Tribe, Intergovernmental Immunities in Litigation, Taxation, and Regu HUTTO v. FINNEY 707 678 Opinion of Powell, J. The Court maintains that the Act presents a special case because (i) it imposes attorney’s fees as an element of costs that traditionally have been awarded without regard to the States’ constitutional immunity from monetary liability, and (ii) Congress acted pursuant to its enforcement power under § 5 of the Fourteenth Amendment, as contrasted with its power under more general grants such as the Commerce Clause. I find neither ground a persuasive justification for dilution of the “clear statement” rule. Notwithstanding the limitations of the Court’s first ground of justification, see ante, at 697 n. 27,1 am unwilling to ignore otherwise applicable principles simply because the statute in question imposes substantial monetary liability as an element of “costs.” Counsel fees traditionally have not been part of the routine litigation expenses assessed against parties in American courts. Cf. Alyeska Pipeline Service Co. v. Wilderness Society, 421 U. S. 240 (1975); Arcambel v. Wiseman, 3 Dall. 306 (1796). Quite unlike those routine expenses, an award of counsel fees may involve substantial sums and is not a charge intimately related to the mechanics of the litigation. I therefore cannot accept the Court’s assumption that counselfee awards are part of “the ordinary discipline of the courtroom.” Ante, at 696 n. 24.5 lation: Separation of Powers Issues in Controversies About Federalism, 89 Harv. L. Rev. 682, 695 (1976) (emphasis supplied). 5 The Court places undue reliance on Fairmont Creamery Co. v. Minnesota, 275 U. S. 70 (1927), in support of its holding. That decision holds that no common-law bar of sovereign immunity prevents the imposition of costs against the State “when [it is] a party to litigation in this Court . . . .” Id., at 74. In addition to the fact that the State was a party in the litigation, and that there is no discussion of counsel fees, Fairmont Creamery “did not mention the eleventh amendment. Furthermore, the Court had held long before that when an individual appeals a case initiated by a state to the Supreme Court, that appeal does not fall within the eleventh amendment’s prohibition of suit 'commenced or prosecuted against’ the states.” Note, Attorneys’ Fees and the Eleventh Amendment, 88 Harv. L. Rev. 1875, 1890 (1975). 708 OCTOBER TERM, 1977 Opinion of Powell, J. 437 U. S. Moreover, counsel-fee awards cannot be viewed as having the kind of “ancillary effect on the state treasury,” Edelman, 415 U. S., at 668, that avoids the need for an explicit waiver of Eleventh Amendment protections. As with damages and restitutory relief, an award of counsel fees could impose a substantial burden on the State to make unbudgeted disbursements to satisfy an obligation stemming from past (as opposed to post-litigation) activities. It stretches the rationale of Edelman beyond recognition to characterize such awards as “the necessary result of compliance with decrees which by their terms [are] prospective in nature.” Ibid. In the case of a purely prospective decree, budgeting can take account of the expenditures entailed in compliance, and the State retains some flexibility in implementing the decree, which may reduce the impact on the state fisc. In some situations fiscal considerations may induce the State to curtail the activity triggering the constitutional obligation. Here, in contrast, the State must satisfy a potentially substantial liability without the measure of flexibility that would be available with respect to prospective relief. The Court’s second ground for application of a diluted “clear statement” rule stems from language in Fitzpatrick recognizing that “[w]hen Congress acts pursuant to § 5” of the Fourteenth Amendment, “it is exercising [legislative] authority under one section of a constitutional Amendment whose other sections by their own terms embody limitations on state authority,” 427 U. S., at 456. I do not view this language as overruling, by implication, Edelman’s holding that no waiver is present in § 19836—the quintessential Fourteenth Amend 6 Mr. Justice Brennan’s concurring opinion asserts that the Court’s holding in Edelman has been undermined, sub silentio, by Fitzpatrick and the re-examination of the legislative history of § 1983 undertaken in MoneU. The language in question from Fitzpatrick was not essential to the Court’s holding in that case. Moreover, this position ignores the fact that Edelman rests squarely on the Eleventh Amendment immunity, without HUTTO v. FINNEY 709 678 Opinion of Powell, J. ment measure—or disturbing the vitality of the “threshold [requirement] of congressional authorization to sue a class of defendants which literally includes States,” 415 U. S., at 672.7 adverting in terms to the treatment of the legislative history in Monroe v. Pape, 365 U. S. 167 (1961). And there is nothing in Monroe itself that supports the proposition that § 1983 was “thought to include only natural persons among those who could be party defendants .... Ante, at 701. The Monroe Court held that because the 1871 Congress entertained doubts as to its “power ... to impose civil liability on municipalities,” the Court could not “believe that the word ‘person’ was used in this particular Act to include them.” 365 U. S., at 190, 191. As the decision in Monett itself illustrates, see n. 2, supra, the statutory issue of municipal liability is quite independent of the question of the State’s constitutional immunity. Mr. Justice Brennan’s opinion appears to dispense with the “clear statement” requirement altogether, a position that the Court does not embrace today. It relies on the reference to “bodies politic” in the Dictionary Act,” Act of Feb. 25, 1871, 16 Stat. 431, as adequate to override the States’ constitutional immunity, even though there is no evidence of a congressional purpose in 1871 to abrogate the protections of the Eleventh Amendment. But the Court’s rulings in Edelman and Employees are rendered obsolete if provisions like the “Dictionary Act” are all that is necessary to expose the States to monetary liability. After a century of § 1983 jurisprudence, in which States were not thought to be liable in damages, Edelman made clear that the 1871 measure does not override the Eleventh Amendment. I would give force to our prior Eleventh Amendment decisions by requiring explicit legislation on the point. 7 The Court suggests that the “dissenting Brethren would apparently force [the individual] officers to bear the award alone.” Ante, at 699 n. 32. It is not clear to me that this issue, not fairly embraced within the questions presented, is before us. Moreover, there is no suggestion in the opinion below that the Court of Appeals intended that its award of fees for “services on this appeal” would be paid by the individual petitioners, in the event the Eleventh Amendment were found to bar an award against the Department of Correction. See 548 F. 2d 740, 742—743 (1977). But even if the question properly were before this Court, there is nothing in the Act that requires the routine imposition of counsel-fee liability on anyone. As we noted in Monett, the Act “allows prevailing parties (in the discretion of the court) in § 1983 suits to obtain attorney’s fees from the losing parties . . . .” 436 U. S., at 698-699 (emphasis supplied). Congress deliberately rejected a mandatory statute, in favor of “a more moderate 710 OCTOBER TERM, 1977 Rehnquist, J., dissenting 437U.S. Because explicit authorization “to join a State as defendant,” Fitzpatrick, 427 U. S., at 452, is absent here, and because every part of the Act can be given meaning without ascribing to Congress an intention to override the Eleventh Amendment immunity,8 I dissent from Part II-B of the Court’s opinion. Mr. Justice Rehnquist, dissenting.* The Court’s affirmance of a District Court’s injunction against a prison practice which has not been shown to violate the Constitution can only be considered an aberration in light of decisions as recently as last Term carefully defining the remedial discretion of the federal courts. Dayton Board of Education v. Brinkman, 433 U. S. 406 (1977); Milliken v. Bradley, 433 U. S. 267 (1977) (Milliken H). Nor are any of the several theories which the Court advances in support of its affirmance of the assessment of attorney’s fees against the taxpayers of Arkansas sufficiently convincing to overcome the prohibition of the Eleventh Amendment. Accordingly, I dissent. approach [which left] the matter to the discretion of the judge, guided of course by the case law interpreting similar attorney’s fee provisions.” H. R. Rep. No. 94-1558, p. 8 (1976). Whether or not the standard of cases like Wood v. Strickland, 420 U. S. 308 (1975), was rejected with respect to counsel-fee liability, see H. R. Rep. No. 94-1558, supra, at 9, and n. 17, neither the Act nor its legislative history prevents a court from taking into account the personal culpability of the individual officer where an award against the government entity would be barred by the Eleventh Amendment. 81 do not understand the Court’s observation that “(i]f the Act does not impose liability for attorney’s fees on the States, it has no meaning with respect to them.” Ante, at 698 n. 31. Significantly, the Court does not say that any part of the Act would be rendered meaningless without finding an Eleventh Amendment waiver. Cf. Employees, 411 U S at 285-286. *Mr. Justice White joins Part II of this opinion. HUTTO v. FINNEY 711 678 Rehnquist, J., dissenting I No person of ordinary feeling could fail to be moved by the Court’s recitation of the conditions formerly prevailing in the Arkansas prison system. Yet I fear that the Court has allowed itself to be moved beyond the well-established bounds limiting the exercise of remedial authority by the federal district courts. The purpose and extent of that discretion in another context were carefully defined by the Court’s opinion last Term in Milliken II, supra, at 280-281: “In the first place, like other equitable remedies, the nature of the desegregation remedy is to be determined by the nature and scope of the constitutional violation. Swann v. Charlotte-Mecklenburg Board of Education, 402 U. S. [1,] 16 [(1971)]. The remedy must therefore be related to The condition alleged to offend the Constitution . . . Milliken [v. Bradley}, 418 U. S. [717,] 738 [(1974)]. Second, the decree must indeed be remedial in nature, that is, it must be designed as nearly as possible ‘to restore the victims of discriminatory conduct to the position they would have occupied in the absence of such conduct.’ Id., at 746. Third, the federal courts in devising a remedy must take into account the interests of state and local authorities in managing their own affairs, consistent with the Constitution.” (Footnotes omitted.)1 1 The Court suggests, ante, at 687 n. 9, that its holding is consistent with Milliken II, because it “was not remedying the present effects of a violation in the past. It was seeking to bring an ongoing violation to an immediate halt.” This suggestion is wide of the mark. Whether exercising its authority to “remed [y] the present effects of a violation in the past,” or “seeking to bring an ongoing violation to an immediate halt,” the court’s remedial authority remains circumscribed by the language quoted in the text from Milliken II. If anything, less ingenuity and discretion would appear to be required to “bring an ongoing violation to an immediate halt” than in “remedying the present effects of a violation in the past.” The difficulty with the Court’s position is that it quite properly refrains 712 OCTOBER TERM, 1977 Rehnquist, J., dissenting 437U.S. The District Court’s order limiting the maximum period of punitive isolation to 30 days in no way relates to any condition found offensive to the Constitution. It is, when stripped of descriptive verbiage, a prophylactic rule, doubtless well designed to assure a more humane prison system in Arkansas, but not complying with the limitations set forth in Milliken II, supra. Petitioners do not dispute the District Court’s conclusion that the overcrowded conditions and the inadequate diet provided for those prisoners in punitive isolation offended the Constitution, but the District Court has ordered a cessation of those practices. The District Court found that the confinement of two prisoners in a single cell on a restricted diet for 30 days did not violate the Eighth Amendment. 410 F. Supp. 251, 278 (ED Ark. 1970). While the Court today remarks that “the length of confinement cannot be ignored,” ante, at 686, it does not find that confinement under the conditions described by the District Court becomes unconstitutional on the 31st day. It must seek other justifications for its affirmance of that portion of the District Court’s order. Certainly the provision is not remedial in the sense that it “restore [s] the victims of discriminatory conduct to the position they would have occupied in the absence of such conduct.” Milliken v. Bradley, 418 U. S. 717, 746 (1974) {Milliken I). The sole effect of the provision is to grant future offenders against prison discipline greater benefits than the Constitution requires; it does nothing to remedy the plight of past victims of conditions which may well have been unconstitutional. A prison is unlike a school system, in which students in the later grades may receive special instruction to compensate for discrimination to which they were subjected in the from characterizing solitary confinement for a period in excess of 30 days as a cruel and unusual punishment; but given this position, a “remedial” order that no such solitary confinement may take place is necessarily of a prophylactic nature, and not essential to “bring an ongoing violation to an immediate halt.” HUTTO v. FINNEY 713 678 Rehnquist, J., dissenting earlier grades. Milliken II, supra, at 281—283. Nor has it been shown that petitioners’ conduct had any collateral effect upon private actions for which the District Court may seek to compensate so as to eliminate the continuing effect of past unconstitutional conduct. See Swann v. Charlotte-Mecklenburg Board of Education, 402 U. S. 1, 28 (1971). Even where such remedial relief is justified, a district court may go no further than is necessary to eliminate the consequences of official unconstitutional conduct. Day ton, supra, at 419-420; Pasadena Board of Education n. Spangler, 427 U. S. 424, 435-437 (1976); Swann, supra, at 31-32. The Court’s only asserted justification for its affirmance of the decree, despite its dissimilarity to remedial decrees in other contexts, is that it is “a mechanical—and therefore an easily enforced—method of minimizing overcrowding.” Ante, at 688 n. 11. This conclusion fails adequately to take into account the third consideration cited in Milliken II: “the interests of state and local authorities in managing their own affairs, consistent with the Constitution.” 433 U. S., at 281. The prohibition against extended punitive isolation, a practice which has not been shown to be inconsistent with the Constitution, can only be defended because of the difficulty of policing the District Court’s explicit injunction against the overcrowding and inadequate diet which have been found to be violative of the Constitution. But even if such an expansion of remedial authority could be justified in a case where the defendants had been repeatedly contumacious, this is not such a case. The District Court’s dissatisfaction with petitioners’ performance under its earlier direction to “make a substantial start,” Holt v. Sarver, 300 F. Supp. 825, 833 (ED Ark. 1969), on alleviating unconstitutional conditions cannot support an inference that petitioners are prepared to defy the specific orders now laid down by the District Court and not challenged by the petitioners. A proper respect for “the interests of state and local authorities in managing their own 714 OCTOBER TERM, 1977 Rehnquist, J., dissenting 437U.S. affairs,” Milliken II, 433 U. S., at 281, requires the opposite conclusion.2 The District Court’s order enjoins a practice which has not been found inconsistent with the Constitution. The only-ground for the injunction, therefore, is the prophylactic one of assuring that no unconstitutional conduct will occur in the future. In a unitary system of prison management there would be much to be said for such a rule, but neither this Court nor any other federal court is entrusted with such a management role under the Constitution. II The Court advances separate theories to support the separate awards of attorney’s fees in this case. First, the Court holds that the taxpayers of Arkansas may be held responsible for the bad faith of their officials in the litigation before the District Court. Second, it concludes that the award of fees in the Court of Appeals, where there was no bad faith, is authorized by the Civil Rights Attorney’s Fees Awards Act of 1976. Pub. L. No. 94-559, 90 Stat. 2641, 42 U. S. C. § 1988 (1976 ed.). The first holding results in a totally unnecessary intrusion upon the State’s conduct of its own affairs, and the second is not supportable under this Court’s earlier decisions outlining congressional authority to abrogate the protections of the Eleventh Amendment. A Petitioners do not contest the District Court’s finding that they acted in bad faith. For this reason, the Court has no 21 reserve judgment on whether such a precautionary order would be justified where state officials have been shown to have violated previous remedial orders. I also note the similarity between this decree and the “no majority of any minority” requirement which was found impermissible in Pasadena Board of Education v. Spangler, 427 U. S. 424 (1976), even though it too might have been defended on the theory that it was an easily enforceable mechanism for preventing future acts of official discrimination. HUTTO v. FINNEY 715 678 Rehnquist, J., dissenting occasion to address the nature of the showing necessary to support an award of attorney’s fees for bad faith under Alyeska Pipeline Service Co. v. Wilderness Society, 421 U. S. 240, 258-259 (1975). The only issue before us is whether a proper finding of bad faith on the part of state officials will support an award of attorney’s fees directly against the state treasury under the ancillary-effect doctrine of Edelman n. Jordan, 415 U. S. 651, 668 (1974). The ancillary-effect doctrine recognized in Edelman is a necessary concomitant of a federal court’s authority to require state officials to conform their conduct to the dictates of the Constitution. “State officials, in order to shape their official conduct to the mandate of the Court’s decrees, would more likely have to spend money from the state treasury than if they had been left free to pursue their previous course of conduct.” Id., at 668. The Court today suggests that a federal court may impose a retroactive financial penalty upon a State when it fails to comply with prospective relief previously and validly ordered. “If a state agency refuses to adhere to a court order, a financial penalty may be the most effective means of insuring compliance.” Ante, at 691. This application of the ancillary-effect doctrine has never before been recognized by this Court, and there is no need to do so in this case, since it has not been shown that these petitioners have “refuse[d] to adhere to a court order.” A State’s jealous defense of its authority to operate its own correctional system cannot casually be equated with contempt of court.3 3 In any event, it is apparent that the District Court did not consider its order a form of retroactive discipline supporting its previous orders. The court concluded that the allowance of the fee “may incline the Department to act in such a manner that further protracted litigation about the prisons will not be necessary.” 410 F. Supp. 251, 285 (ED Ark. 1976). It does not appear to me that the court’s desire to weaken petitioners’ future resistance is a legitimate use of the Alyeska doctrine permitting the award of attorney’s fees for past acts of bad faith. 716 OCTOBER TERM, 1977 Rehnquist, J., dissenting 437U.S. Even were I to agree with the Court that petitioners had willfully defied federal decrees, I could not conclude that the award of fees against the taxpayers of Arkansas would be justified, since there is a less intrusive means of insuring respondents’ right to relief. It is sufficient to order an award of fees against those defendants, acting in their official capacity, who are personally responsible for the recalcitrance which the District Court wishes to penalize. There is no reason for the federal courts to engage in speculation as to whether the imposition of a fine against the State is “less intrusive” than “sending high state officials to jail.” Ibid. So long as the rights of the plaintiffs and the authority of the District Court are amply vindicated by an award of fees, it should be a matter of no concern to the court whether those fees are paid by state officials personally or by the State itself. The Arkansas Legislature has already made statutory provision for deciding when its officials shall be reimbursed by the State for judgments ordered by the federal courts. 1977 Ark. Gen. Act No. 543. The Court presents no persuasive reason for its conclusion that the decision of who must pay such fees may not safely be left to the State involved. It insists, ante, at 699 n. 32, that it is “manifestly unfair” to leave the individual state officers to pay the award of counsel fees rather than permitting their collection directly from the state treasury. But petitioners do not contest the District Court’s finding that they acted in bad faith, and thus the Court’s insistence that it is “unfair” to impose attorney’s fees on them individually rings somewhat hollow.4 Even in a case where the equities were more strongly in favor of the individual state officials (as opposed to the State as an entity) than they are in this case, 4 It is true that fees may be awarded under 42 U. S. C. § 1988 (1976 ed.) even in the absence of bad faith. But that statute leaves the decision to award fees to the discretion of the district court, which may be expected to alleviate any possible unfairness. HUTTO v. FINNEY 717 678 Rehnquist, J., dissenting the possibility of individual liability in damages of a state official where the State itself could not be held liable is as old as Ex parte Young, 209 U. S. 123 (1908), and has been repeatedly reaffirmed by decisions of this Court. Great Northern Life Insurance Co. v. Read, 322 U. S. 47 (1944); Ford Motor Co. v. Department of Treasury, 323 U. S. 459 (1945); Edelman v. Jordan, supra. Since the Court evidences no disagreement with this line of cases, its assertion of ‘unfairness is not only doubtful in fact but also irrelevant as a matter of law. Likewise, the Court’s fear that imposition of liability would inhibit state officials in the fearless exercise of their duties may be remedied, if deemed desirable, by legislation in each of the various States similar to that which Arkansas has already enacted. B For the reasons stated in the dissenting portion of my Brother Powell’s opinion, which I join, I do not agree that the Civil Rights Attorney’s Fees Awards Act of 1976 can be considered a valid congressional abrogation of the State’s Eleventh Amendment immunity. I have in addition serious reservations about the lack of any analysis accompanying the Court’s transposition of the holding of Fitzpatrick n. Bitzer, 427 U. S. 445 (1976), to this case. In Fitzpatrick, we held that under § 5 of the Fourteenth Amendment Congress could explicitly allow for recovery against state agencies without violating the Eleventh Amendment. But in Fitzpatrick, supra, there was conceded to be a violation of the Equal Protection Clause which is contained in haec verba in the language of the Fourteenth Amendment itself. In this case the claimed constitutional violation is the infliction of cruel and unusual punishment, which is expressly prohibited by the Eighth but not by the Fourteenth Amendment. While the Court has held that the Fourteenth Amendment “incorporates” the prohibition against cruel and unusual punishment, it is not at all clear to me that it follows that Congress has the same enforcement power 718 OCTOBER TERM, 1977 Rehnquist, J., dissenting 437 U.S. under § 5 with respect to a constitutional provision which has merely been judicially “incorporated” into the Fourteenth Amendment that it has with respect to a provision which was placed in that Amendment by the drafters. I would therefore reverse the judgment of the Court of Appeals in its entirety. Reporter’s Note The next page is purposely numbered 901. The numbers between 718 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 pubheation of the preliminary prints of the United States Reports. ORDERS FROM JUNE 14 THROUGH JUNE 19, 1978 June 14, 1978 Dismissal Under Rule 60 No. 77-6373. Eaton v. United States. C. A. 5th Cir. Certiorari dismissed under this Court’s Rule 60. Reported below: 567 F. 2d 389. June 19, 1978 Dismissal Under Rule 60 No. 77-6547. Wedel v. United States. C. A. 10th Cir. Certiorari dismissed under this Court’s Rule 60. Appeals Dismissed No. 77-1460. Niagara Mohawk Power Corp. v. Public Service Commission of New York. Appeal from App. Div., Sup. Ct. N. Y., 3d Jud. Dept., dismissed for want of jurisdiction. Treating the papers whereon the appeal was taken as a petition for writ of certiorari, certiorari denied. Reported below: 59 App. Div. 2d 73, 397 N. Y. S. 2d 210. No. 77-6796. Godbout v. Norton. 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. Reported below: 262 N. W. 2d 374. Miscellaneous Orders No. A-1029 (77-1674). Field v. United States. C. A. 2d Cir. Application for bail denied without prejudice to an application in the District Court. See 18 U. S. C. §§ 3148, 3146; Fed. Rule App. Proc. 9 (b); United States v. Bowdach, 561 F. 2d 1160, 1167, and n. 2 (CA5 1977). 901 902 OCTOBER TERM, 1977 June 19, 1978 437 U.S. No. A-1054 (Nos. 77-452, 77-457, 77-551, and 77-602). Mobil Alaska Pipeline Co. v. United States et al. ; Exxon Pipeline Co. v. United States et al. ; BP Pipelines, Inc. v. United States et al.; and ARCO Pipe Line Co. v. United States et al., 436 U. S. 631. Application for an order directing compliance with conditions of stay received and presented to Mr. Justice Brennan, and by him referred to the Court. It is ordered that the judgment of this Court in the aboveentitled cases be issued forthwith. No. 77-891. Beal, Secretary of Welfare of Pennsylvania, et al. v. Franklin et al. D. C. E. D. Pa. [Probable jurisdiction noted, 435 U. S. 913.] Motion for appointment of Alan Ernest, Esquire, as guardian ad litem for unborn children denied. No. 77-1388. Massachusetts v. White. Sup. Jud. Ct. Mass. [Certiorari granted, 436 U. S. 925.] Motion of respondent for leave to proceed further herein in forma pauperis granted. No. 77—6670. Carter v. Roberts, U. S. District Judge. Motion for leave to file petition for writ of mandamus denied. Probable Jurisdiction Noted No. 77-1715. Secretary of Public Welfare of Pennsylvania et al. v. Institutionalized Juveniles et al. Appeal from D. C. E. D. Pa. Motion of appellees for leave to proceed in forma pauperis granted. Probable jurisdiction noted. Case set for oral argument with No. 75-1690, Parham n. J. L. [probable jurisdiction noted, 431 U. S. 936; restored to calendar, 434 U. S. 1031]. Certiorari Granted No. 77—1547. Douglas Oil Company of California et al. v. Petrol Stops Northwest et al. C. A. 9th Cir. Certiorari granted. Reported below: 571 F. 2d 1127. ORDERS 903 437 U.S. June 19, 1978 No. 77-874. Alexander et al. v. United States Department of Housing and Urban Development et al. C. A. 7th Cir,; and No. 77-1463. Harris, Secretary of Housing and Urban Development, et al. v. Cole et al. C. A. D. C. Cir. Certiorari granted, cases consolidated, and a total of one hour allotted for oral argument. Reported below: No. 77-874, 555 F. 2d 166; No. 77-1463, 187 U. S. App. D. C. 156, 571 F. 2d 590. No. 77-1553. County of Los Angeles et al. v. Davis et al. C. A. 9th Cir. Certiorari granted. Reported below: 566 F. 2d 1334. Certiorari Denied. (See also Nos. 77-1460 and 77-6796, supra.) No. 77-1137. Neville v. Friedman, Judge, et al. Sup. Ct. Ill. Certiorari denied. Reported below: 67 Ill. 2d 488, 367 N. E. 2d 1341. No. 77-1225. Seidel v. Texas. Ct. Crim. App. Tex. Certiorari denied. Reported below: 557 S. W. 2d 311. No. 77-1354. Furrer et ux. v. Commissioner of Internal Revenue. C. A. 9th Cir. Certiorari denied. Reported below: 566 F. 2d 1115. No. 77-1389. Cadillac Overall Supply Co. v. United States. C. A. 5th Cir. Certiorari denied. Reported below: 568 F. 2d 1078. No. 77-1394. Quick Pak, Inc. v. National Labor Relations Board. C. A. 6th Cir. Certiorari denied. Reported below: 570 F. 2d 649. No. 77-1397. Abrams v. United States. C. A. 5th Cir. Certiorari denied. Reported below: 568 F. 2d 411. 904 OCTOBER TERM, 1977 June 19, 1978 437 U.S. No. 77-1399. Mechanic’s Building & Loan Co. v. Federal Home Loan Bank Board et al. C. A. 6th Cir. Certiorari denied. Reported below: 571 F. 2d 582. No. 77-1401. Goodwin v. Briggs et al. C. A. D. C. Cir. Certiorari denied. Reported below: 186 U. S. App. D. C. 179, 569 F. 2d 10. No. 77-1420. Chicago Health Clubs, Inc. v. National Labor Relations Board. C. A. 7th Cir. Certiorari denied. Reported below: 567 F. 2d 331. No. 77-1423. Pavone v. United States. C. A. 7th Cir. Certiorari denied. Reported below: 570 F. 2d 674. No. 77-1428. Yesterday’s Children et al. v. Kennedy, Director, Department of Children and Family Services of Illinois, et al. C. A. 7th Cir. Certiorari denied. Reported below: 569 F. 2d 431. No. 77-1448. Thomas v. United States. C. A. 9th Cir. Certiorari denied. Reported below: 571 F. 2d 589. No. 77-1452. Lewin et ux. v. Commissioner of Internal Revenue. C. A. 7th Cir. Certiorari denied. Reported below: 569 F. 2d 444. No. 77-1458. MAPCO, Inc., et al. v. Carter, President of the United States, et al. Temp. Emerg. Ct. App. Certiorari denied. Reported below: 573 F. 2d 1268. No. 77-1488. Perlman et al. v. California. Ct. App. Cal., 2d App. Dist. Certiorari denied. No. 77-1508. Gust v. United States Customs Service. C. A. 6th Cir. Certiorari denied. Reported below: 571 F. 2d 581. No. 77-1525. Kaufman v. Association of the Bar of the City of New York. C. A. 2d Cir. Certiorari denied. Reported below: 573 F. 2d 1291. ORDERS 905 437 U.S. June 19, 1978 No. 77-1526. Golz v. Illinois. App. Ct. Ill., 1st Dist. Certiorari denied. Reported below: 53 Ill. App. 3d 654, 368 N. E. 2d 1069. No. 77-1527. Laje v. R. E. Thomason General Hospital. C. A. 5th Cir. Certiorari denied. Reported below: 564 F. 2d 1159. No. 77-1530. Harbin v. Interlake Steamship Co.; and No. 77-1536. Inteblake Steamship Co. v. Harbin. C. A. 6th Cir. Certiorari denied. Reported below: 570 F. 2d 99. No. 77-1531. Associated Milk Producers, Inc. v. United States et al. C. A. 5th Cir. Certiorari denied. Reported below: 566 F. 2d 1293. No. 77-1537. Volkswagen werk AG et al. v. Herman, U. S. District Judge, et al. C. A. 3d Cir. Certiorari denied. No. 77-1544. Joyce Beverages, Inc., et al. v. Joyce et al. C. A. 2d Cir. Certiorari denied. Reported below: 571 F. 2d 703. No. 77-1549. Johnson et al. v. General Motors Assembly Division, General Motors Corp. Ct. App. Ga. Certiorari denied. Reported below: 144 Ga. App. 305, 241 S. E. 2d 30. No. 77-1560. Parish of East Baton Rouge v. Pierson et al. Ct. App. La., 1st Cir. Certiorari denied. Reported below: 353 So. 2d 726. No. 77-1561. O’Connor et al. v. City of Louisville Fire Fighters Pension Fund. Sup. Ct. Ky. Certiorari denied. Reported below: 561 S. W. 2d 675. No. 77-1563. PPX Enterprises, Inc. v. Scepter Records, Inc. Ct. App. N. Y. Certiorari denied. Reported below: 43 N. Y. 2d 972, 375 N. E. 2d 731. 906 OCTOBER TERM, 1977 June 19, 1978 437 U.S. No. 77-1572. Rogers v. Florida. Dist. Ct. App. Fla., 2d Dist. Certiorari denied. Reported below: 349 So. 2d 239. No. 77-1579. Fennell v. Butler et al. C. A. 8th Cir. Certiorari denied. Reported below: 570 F. 2d 263. No. 77-1623. Griffin v. United States. C. A. 7th Cir. Certiorari denied. Reported below: 575 F. 2d 1341. No. 77-5985. Cane v. Kentucky. Ct. App. Ky. Certiorari denied. Reported below: 556 S. W. 2d 902. No. 77-6339. Smith v. United States. C. A. 9th Cir. Certiorari denied. Reported below: 570 F. 2d 351. No. 77-6395. Guzman v. United States; and No. 77-6577. Bensor v. United States. C. A. 5th Cir. Certiorari denied. Reported below: 566 F. 2d 560. No. 77-6428. Hannah v. New Jersey. C. A. 3d Cir. Certiorari denied. Reported below: 568 F. 2d 771. No. 77-6429. Griffin v. San Bernardino Police Department et al. C. A. 9th Cir. Certiorari denied. No. 77-6463. Buttorff et al. v. United States. C. A. 8th Cir. Certiorari denied. Reported below: 572 F. 2d 619. No. 77-6489. Davis et al. v. United States. C. A. 5th Cir. Certiorari denied. Reported below: 566 F. 2d 105. No. 77-6501. White v. United States; and No. 77-6544. Diaz v. United States. C. A. 9th Cir. Certiorari denied. Reported below: No. 77-6501, 570 F. 2d 354; No. 77-6544, 570 F. 2d 352. No. 77-6520. Johnson v. United States. C. A. 5th Cir. Certiorari denied. Reported below: 569 F. 2d 269. No. 77-6531. Cameron v. United States. C. A. 9th Cir. Certiorari denied. Reported below: 570 F. 2d 352. ORDERS 907 437 U.S. June 19, 1978 No. 77-6548. Morgan et al. v. Jackson, Corrections Director, et al. C. A. D. C. Cir. Certiorari denied. No. 77-6551. Hamilton v. Delaware. Sup. Ct. Del. Certiorari denied. Reported below: 382 A. 2d 249. No. 77-6559. Jackson v. United States. C. A. 7th Cir. Certiorari denied. Reported below: 569 F. 2d 1003. No. 77-6576. Johnson v. United States. C. A. 9th Cir. Certiorari denied. Reported below: 572 F. 2d 227. No. 77-6581. Ferri v. United States et al. C. A. D. C. Cir. Certiorari denied. No. 77-6586. Cheshire v. United States. C. A. 5th Cir. Certiorari denied. Reported below: 569 F. 2d 887. No. 77-6591. Ferrara v. United States. C. A. 8th Cir. Certiorari denied. Reported below: 571 F. 2d 428. No. 77-6614. Palmer v. United States. C. A. 3d Cir. Certiorari denied. Reported below: 574 F. 2d 164. No. 77-6627. King v. United States. C. A. 8th Cir. Certiorari denied. Reported below: 572 F. 2d 1274. No. 77-6636. Zepeda-Santana v. United States. C. A. 5th Cir. Certiorari denied. Reported below: 569 F. 2d 1386. No. 77-6640. Williams v. Louisiana et al. Sup. Ct. La. Certiorari denied. Reported below: 353 So. 2d 1299. No. 77-6648. Nasim v. Commissioner of Internal Revenue. C. A. 4th Cir. Certiorari denied. Reported below: 565 F. 2d 157. No. 77-6656. Bonnell v. Black, Reformatory Superintendent. C, A. 6th Cir. Certiorari denied. No. 77-6658. Brewster v. New York. App. Div., Sup. Ct. N. Y., 1st Jud. Dept. Certiorari denied. Reported below: 60 App. Div. 2d 796, 390 N. Y. S. 2d 958. 908 OCTOBER TERM, 1977 June 19, 1978 437 U.S. No. 77-6660. Taylor v. Poehling, Assistant Circuit Attorney, City of St. Louis. C. A. 8th Cir. Certiorari' denied. No. 77-6664. Chambers v. Oklahoma. Ct. Crim. App. Okla. Certiorari denied. No. 77-6665. Rachal v. Superintendent of M. C. I. at Walpole. C. A. 1st Cir. Certiorari denied. No. 77-6668. Gray et al. v. California; and No. 77-6682. Mertz v. California. Ct. App. Cal., 2d App. Dist. Certiorari denied. No. 77-6669. Williams v. Ct. Crim, App. Ala Certiorari denied. Reported below: 354 So. 2d 829. No. 77-6677. Sullivan v. California. Ct. App. Cal., 1st App. Dist. Certiorari denied. No. 77-6690. Meader v. Montana. Sup. Ct. Mont. Certiorari denied. No. 77-6693. Radisich v. Radisich et al. Ct. App. Cal., 1st App. Dist. Certiorari denied. No. 77-6695. Watson v. Maryland. Ct. App. Md. Certiorari denied. Reported below: 282 Md. 73, 382 A. 2d 574. No. 77-6786. Bradford v. United States. C. A. 5th Cir. Certiorari denied. Reported below: 571 F. 2d 1351. No. 77-6791. Skidmore v. United States. C. A. 6th Cir. Certiorari denied. Reported below: 575 F. 2d 1338. No. 77-6800. Kirk v. United States. C. A. 4th Cir. Certiorari denied. No. 77-6803. Marquez v. United States. C. A. 7th Cir. Certiorari denied. Reported below: 574 F. 2d 931. ORDERS 909 437 U.S. June 19, 1978 No. 77-1196. Sherwin et al. v. United States. C. A. 9th Cir. Certiorari denied. Reported below: 572 F. 2d 196. Mr. Justice Brennan, with whom Mr. Justice Stewart and Mr. Justice Marshall join, dissenting. Petitioners were convicted in the United States District Court for the Central District of California of multiple counts of knowingly transporting allegedly obscene materials in interstate commerce by common carrier in violation of 18 U. S. C. §§ 14621 and 14652 (1976 ed.). Although it overturned convictions on some counts, the Court of Appeals for the Ninth Circuit affirmed petitioners’ convictions on three counts and also petitioner Sherwin’s conviction for conspiracy under 18 U. S. C. §371 (1976 ed.). Petitioners ask this Court to consider whether “a standard of scienter which authorizes obscenity convictions on mere knowledge of the ‘sexual orientation’ of material impermissibly chill [s] the dissemination of expression protected under the First Amendment of the United States Constitution.” Pet. for Cert. 2. This question is much the same as that presented in Ballew v. Georgia, 435 U. S. 223 (1978), Sewell v. 1 “Whoever brings into the United States, or any place subject to the jurisdiction thereof, or knowingly uses any express company or other common carrier, for carriage in interstate or foreign commerce— “(a) any obscene, lewd, lascivious, or filthy book, pamphlet, picture, motion-picture film, paper, letter, writing, print, or other matter of indecent character . . . “Shall be fined not more than $5,000 or imprisoned not more than five years, or both . . . .” 2 “Whoever knowingly transports in interstate or foreign commerce for the purpose of sale or distribution any obscene, lewd, lascivious, or filthy book, pamphlet, picture, film, paper, letter, writing, print, silhouette, drawing, figure, image, cast, phonograph recording, electrical transcription or other article capable of producing sound or any other matter of indecent or immoral character, shall be fined not more than $5,000 or imprisoned not more than five years, or both.” 910 OCTOBER TERM, 1977 June 19, 1978 437 U.S. Georgia, 435 U. S. 982 (1978), Robinson v. Georgia, 435 U. S. 991 (1978), and Teal v. Georgia, 435 U. S. 989 (1978). For the reasons stated in my dissent from denial of certiorari in Sewell, supra, at 982, I would hear oral argument on this issue. Barring this, I would summarily reverse petitioners’ convictions. See, e. g., United States v. Orito, 413 U. S. 139, 147 (1973) (Brennan, J., dissenting); Christian v. United States, 432 U. S. 910 (1977) (Brennan, J., dissenting from denial of certiorari); Danley v. United States, 424 U. S. 929 (1976) (same); Kutler v. United States, 423 U. S. 959 (1975) (same). No. 77-1342. Perrin et al. v. Dunn et al. C. A. 1st Cir. Motions of respondents for leave to proceed in forma pauperis granted. Certiorari denied. Reported below: 570 F. 2d 21. No. 77—1381. Drummond et ux. v. Fulton County Department of Family and Children’s Services et al. ; and Nov 77—6454. Hill v. Fulton County Department of Family and Children’s Services et al. C. A. 5th Cir. Certiorari denied. Mr. Justice Brennan and Mr. Justice White would grant certiorari. Reported below: 563 F. 2d 1200. No. 77-1505. Essex County Welfare Board v. Department of Institutions and Agencies et al. Sup. Ct. N. J. Motion of respondent Irene Stowers for leave to proceed in forma pauperis granted. Certiorari denied. Reported below: 75 N. J. 232, 381 A. 2d 349. No. 77-1532. Hall, Corrections Commissioner, et al. v. Morgan. C. A. 1st Cir. Motion of respondent for leave to proceed in forma pauperis granted. Certiorari denied. Reported below: 569 F. 2d 1161. ORDERS 911 437 U.S. June 19, 1978 No. 77-1534. South Central Bell Telephone Co. v. Louisiana Public Service Commission. Sup. Ct. La. Certiorari denied. Mr. Justice White and Mr. Justice Powell would grant certiorari. Reported below: 352 So. 2d 964. No. 77-1548. Sheet Metal Workers’ International Association, Local No. 3 v. Siebler Heating & Air Conditioning, Inc., et al. C. A. 8th Cir. Certiorari denied. Mr. Justice White would grant certiorari. Reported below: 563 F. 2d 366. No. 77-6539. Denney v. Texas. Ct. Crim. App. Tex. Certiorari denied. Reported below: 558 S. W. 2d 467. Mr. Justice Brennan and Mr. 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 sentence in this case. No. 77-6616. Layton v. Pogue, Warden. C. A. 9th Cir. Certiorari denied. Mr. Justice Brennan, Mr. Justice White, and Mr. Justice Marshall would grant certiorari. Reported below: 568 F. 2d 777. Rehearing Denied No. 77-1091. Epstein v. Civil Service Commission et al., 435 U. S. 911; No. 77-1190. All Island Delivery Service, Inc., et al. v. United States et al., 435 U. S. 1007; No. 77-1357. Coleman v. Virginia, 435 U. S. 997; No. 77-6235. Bretz v. Crist, Warden, 436 U. S. 908; and No. 77-6303. Banks et al. v. United States Department of Housing and Urban Development et al., 436 U. S. 908. Petitions for rehearing denied. 912 OCTOBER TERM, 1977 June 19, 1978 437 U.S. No. 77-6409. Conrad v. Commerce Bank of Kansas City, 436 U. S. 901; No. 77-6457. Bamond v. New York et al., 436 U. S. 910; and No. 77-6657. Conrad v. First State Bank & Trust Co., 436 U. S. 916. Petitions for rehearing denied. INDEX ACCESS TO WITNESSES’ STATEMENTS IN UNFAIR LABOR PRACTICE PROCEEDINGS. See Freedom of Information Act. ACCUSED’S TERMINATION OF TRIAL. See Constitutional Law, II, 1. ACQUITTALS. See Constitutional Law, II, 2, 3. ADMINISTRATIVE PROCEDURE. See Freedom of Information Act. ADMISSIBILITY OF EVIDENCE. See Constitutional Law, III, 1. ALARM LIMITS. See Patents. “ALASKA HIRE” STATUTE. See Constitutional Law, V; Mootness. ANTITRUST ACTS. Clayton Act—Statute of limitations—Interstate Commerce Commission proceeding—Government’s intervention.—Clayton Act’s statute of limitations was not tolled under § 5 (i) of Act by filing of Government’s petition to intervene in ICC proceeding instituted by motor carrier. Greyhound Corp. v. Mt. Hood Stages, Inc., p. 322. APPEALS. See also Constitutional Law, II, 1, 2. 1. Order denying class certification—Appealability.—“Collateral order” exception to “final decision” requirement of 28 U. S. C. § 1291 does not apply to a prejudgment order denying class certification, nor does “death knell” doctrine support appellate jurisdiction of such an order. Cooper & Lybrand v. Livesay, p. 463. 2. Order denying class certification—Appealability.—Order denying class certification was not appealable under 28 U. S. C. § 1292 (a) (1). Gardner v. Westinghouse Broadcasting Co., p. 478. ARKANSAS. See Civil Rights; Civil Rights Attorney’s Fees Awards Act of 1976; Constitutional Law, IV. AT-LARGE ELECTIONS. See Elections. ATTACHMENT OF JEOPARDY. See Constitutional Law, II, 4. ATTORNEY’S FEES. See Civil Rights Attorney’s Fees Awards Act of 1976; Constitutional Law, IV. BANNING OF OUT-OF-STATE WASTE. See Constitutional Law, I, 2. 913 914 INDEX BOUNTY OR GRANT PAID BY FOREIGN COUNTRY FOR EXPORTATION. See Tariff Act of 1930. BURDEN ON INTERSTATE COMMERCE. See Constitutional Law, I, 3. CARRIERS. See Antitrust Acts. CASE OR CONTROVERSY. See Mootness. CATALYTIC CONVERSION. See Patents. CHOCTAW INDIANS. See Indians. CITY COUNCILS. See Elections. CIVIL RIGHTS. See also Civil Rights Attorney’s Fees Awards Act of 1976. Prisons—Remedy to correct constitutional violations.—District Court did nor err in including 30-day limitation on sentences to isolation as part of its comprehensive remedy to correct constitutional violations in Arkansas prison system. Hutto v. Finney, p. 678. CIVIL RIGHTS ATTORNEY’S FEES AWARDS ACT OF 1976. Action to remedy unconstitutional prison conditions—Court of Appeals’ additional award of attorney’s fees.—Act supports Court of Appeals’ additional award of attorney’s fees to cover services rendered prison inmates on prison officials unsuccessful appeal from adverse judgment in inmates’ action to remedy unconstitutional conditions in Arkansas prison system. Hutto v. Finney, p. 678. CLASS ACTIONS. See Appeals; Federal Rules of Civil Procedure. CLASS CERTIFICATION. See Appeals, CLAYTON ACT. See Antitrust Acts; Federal-State Relations, 2. “COLLATERAL ORDER’’ EXCEPTION. See Appeals, 1. COMMERCE CLAUSE. See Constitutional Law, I. CONSTITUTIONAL LAW. See also Criminal Law. I. Commerce Clause. 1. State income taxes—Apportionment of interstate corporation’s income—Single-factor formula.—.Iowa’s single-factor sales formula for appor-lomng an interstate corporation’s income for state income tax purposes is not invalid under Commerce Clause. Moorman Mfg. Co. v. Bair, p. 267. 2 State prohibition againct importation of waste.—New Jersey statute prohibiting importation of solid or liquid waste originating or collected outside State violates Commerce Clause. Philadelphia v. New Jersey, INDEX 915 CONSTITUTIONAL LAW—Continued. 3. State regulation of retail gasoline marketing.—Maryland statute prohibiting oil producers or refiners from operating retail gasoline stations in State does not violate Commerce Clause. Exxon Corp. v. Governor of Maryland, p. 117. II. Double Jeopardy. 1. Defendant’s termination of trial—Government appeal not barred.— Government’s appeal from defendant’s successful effort to have trial terminated without submission to judge or jury as to his guilt or innocence does not offend Double Jeopardy Clause and hence is not barred by Criminal Appeals Act. United States v. Scott, p. 82. 2. Erroneous exclusion of evidence—Acquittal—Government appeal barred.—Government’s appeal from midtrial ruling excluding evidence and from subsequent acquittal is barred by Double Jeopardy Clause of Fifth Amendment even though midtrial ruling was erroneous. Sanabria v. United States, p. 54. 3. Reversal of conviction for insufficiency of evidence—Preclusion of second trial.—Double Jeopardy Clause of Fifth Amendment precludes second trial once reviewing court has found evidence insufficient to sustain guilty verdict, and only “just” remedy is entry of a judgment of acquittal. Burks v. United States, p. 1. 4. Time when jeopardy attaches—Empaneling and swearing of jury.— Federal rule that jeopardy attaches in a jury trial when jury is empaneled and sworn is an integral part of Fifth Amendment guarantee against double jeopardy made applicable to States by Fourteenth Amendment, and hence Montana statute providing that jeopardy does not attach until first witness is sworn cannot constitutionally be applied in a jury trial. Crist v. Bretz, p. 28. III. Due Process. 1. Accused’s statements—Voluntariness—Admissibility for impeachment purposes.—Due process requires that statements obtained from accused in hospital not be used to impeach his credibility at his state trial on murder and other charges, where it appears that they were not “the product of [his] free and rational choice.” Mincey v. Arizona, p. 385. 2. State income taxes—Apportionment of interstate corporation’s income—Single-factor formula.—Iowa’s single-factor sales formula for apportioning an interstate corporation’s income for state income tax purposes is not invalid under Due Process Clause. Moorman Mfg. Co. v. Bair, p. 267. 3. State regulation of retail gasoline marketing.—Maryland statute prohibiting oil producers’ or refiners’ operation of retail gasoline stations in State and regulating pricing practices does not violate Due Process Clause. Exxon Corp. v. Governor of Maryland, p. 117. 916 INDEX CONSTITUTIONAL LAW—Continued. IV. Eleventh Amendment. Action to remedy unconstitutional prison conditions—Attorney’s fees award.—In action wherein inmates were granted relief against unconstitutional conditions in Arkansas prison system, District Court’s award of attorney’s fees to be paid out of Department of Correction funds is adequately supported by its finding that defendant prison officials had acted in bad faith in failing to cure previously identified constitutional violations, and does not violate Eleventh Amendment. Hutto v. Finney, p. 678. V. Privileges and Immunities Clause. “Alaska Hire” statute—Employment preference for residents.—“Alaska Hire” statute requiring oil and gas leases, easements or right-of-way permits for pipelines, and unitization agreements to contain requirement that Alaska residents be hired in preference to nonresidents violates Privileges and Immunities Clause. Hicklin v. Orbeck, p. 518. VI. Searches and Seizures. Warrantless search of homicide scene.—“Murder scene” exception created by Arizona Supreme Court to warrant requirement is inconsistent with Fourth and Fourteenth Amendments, and warrantless search of accused’s apartment was not constitutionally permissible simply because a homicide had occurred there. Mincey v. Arizona, p. 385. CONSUMER ELECTRONIC PRODUCTS. See Tariff Act of 1930. CORPORATE INCOME TAXES. See Constitutional Law, I, 1; IH, 2. COSTS OF SENDING NOTICES IN CLASS ACTIONS. See Federal Rules of Civil Procedure. COUNTERVAILING DUTIES ON IMPORTS. See Tariff Act of 1930. COURT-IMPOSED REAPPORTIONMENT PLANS. See Elections. COURTS OF APPEALS. See Civil Rights Attorney’s Fees Awards Act of 1976; Elections; Mandamus; National Labor Relations Act, 1. CRIMINAL APPEALS ACT. See Constitutional Law, II, 1. CRIMINAL LAW. See also Constitutional Law, II; III, 1; VI; Indians. Ambiguity as to grounds for new trial—Remand—Double jeopardy determinations.—United States Court of Appeals’ judgment upholding denial of habeas corpus relief to state prisoner against double jeopardy contentions is remanded so that ambiguity as to grounds for Florida Supreme Court’s action in ordering a new trial can be resolved in light of instant opinion and Burks v. United States, ante, p. 1. Greene v. Massey, p. 19. CROSSING OF PICKET LINES. See National Labor Relations Act, 2. INDEX 917 DALLAS, TEX. See Elections. DAMS. See Endangered Species Act of 1973. “DEATH KNELL” DOCTRINE. See Appeals, 1. DENIAL OF CLASS CERTIFICATION. See Appeals. DISCIPLINING OF UNION MEMBERS. See National Labor Relations Act, 2. DISCLOSURE OF WITNESSES’ STATEMENTS IN UNFAIR LABOR PRACTICE PROCEEDINGS. See Freedom of Information Act. DISCRIMINATION AGAINST INTERSTATE COMMERCE. See Constitutional Law, I, 2, 3. DISCRIMINATION AGAINST NONRESIDENTS OF STATE. See Constitutional Law, V. DISCRIMINATION IN VOTING. See Elections. DISCRIMINATORY TAXES. See Federal-State Relations, 1. DISTRIBUTION OF LABOR UNION LITERATURE. See National Labor Relations Act, 1, 3. DISTRICT COURTS. See Civil Rights; Constitutional Law, IV; Elections; Federal Rules of Civil Procedure; Internal Revenue Code; Jurisdiction; Mandamus. DIVERSITY JURISDICTION. See Jurisdiction. DOUBLE JEOPARDY. See Constitutional Law, II; Criminal Law. DUE PROCESS. See Constitutional Law, III. DURATIONAL RESIDENCY REQUIREMENTS FOR EMPLOY- MENT. See Mootness. DUTIES ON IMPORTS. See Tariff Act of 1930. ELECTIONS. Legislative reapportionment plan—Standards for evaluating.—Court of Appeals’ judgment holding that District Court, in approving Dallas, Tex., reapportionment plan (later enacted as ordinance) providing for eight City Council members to be elected from single-member districts and remaining three to be elected at large, erred in evaluating plan only under constitutional standards without also applying rule requiring judicially imposed reapportionment plans, absent exceptional circumstances, to employ only single-member districts, is reversed and case is remanded. Wise v. Lipscomb, p. 535. ELECTRONIC PRODUCTS. See Tariff Act of 1930. ELEVENTH AMENDMENT. See Constitutional Law, IV. 918 INDEX EMPANELING AND SWEARING OF JURY AS TIME WHEN JEOPARDY ATTACHES. See Constitutional Law, II, 4. EMPLOYEE SOLICITATION BY LABOR UNIONS. See National Labor Relations Act, 1. EMPLOYER AND EMPLOYEES. See Constitutional Law, V; Mootness; National Labor Relations Act, 1, 3. EMPLOYMENT PREFERENCES FOR STATE RESIDENTS. See Constitutional Law, V; Mootness. ENDANGERED SPECIES ACT OF 1973. Snail darter—Prohibition against completion of dam.—Despite near completion of dam, Act prohibits impoundment, by dam, of reservoir in portion of river inhabited by snail darter, an endangered species under Act. TVA v. Hill, p. 153. ENFORCEMENT OF INTERNAL REVENUE SERVICE SUMMONSES. See Internal Revenue Code. ENFORCEMENT OF NATIONAL LABOR RELATIONS BOARD’S ORDERS. See National Labor Relations Act, 1. ENVIRONMENTAL LAW. See Endangered Species Act of 1973. ERRONEOUS EXCLUSION OF EVIDENCE. See Constitutional Law, II, 2. EXCISE TAXES. See Federal-State Relations, 1. EXEMPTION OF WITNESSES’ STATEMENTS FROM PREHEARING DISCLOSURE IN UNFAIR LABOR PRACTICE PROCEEDINGS. See Freedom of Information Act. EXEMPTION 7 (A) OF FREEDOM OF INFORMATION ACT. See Freedom of Information Act. EXPENSE OF SENDING NOTICES IN CLASS ACTIONS. See Federal Rules of Civil Procedure. EXPORTS. See Tariff Act of 1930. FEDERAL DAMS. See Endangered Species Act of 1973. FEDERAL JURISDICTION. See Indians; Jurisdiction. FEDERAL PROSECUTION OF INDIANS. See Indians. FEDERAL RULES OF CIVIL PROCEDURE. Rule 23 (d)—Class action—Compilation of class members—Order to defendants to assist—Cost allocation.—Rule 23 (d), not discovery rules, empowers District Court to direct defendants in class action to help compile list of members of plaintiff class, but District Court abused its dis- INDEX 919 FEDERAL RULES OF CIVIL PROCEDURE—Continued. cretion in requiring defendants to bear expense of identifying class members. Oppenheimer Fund, Inc. v. Sanders, p. 340. FEDERAL SAVINGS AND LOAN ASSOCIATIONS. See Federal- State Relations, 1. FEDERAL-STATE RELATIONS. See also Constitutional Law, I, 2; Indians; Mandamus. 1. State excise tax on federal savings and loan associations—Nondiscrim-inatory.—Massachusetts imposition of an excise tax on federal savings and loan associations as measured by their net operating income is not invalid as discriminating against such associations in violation of § 5 (h) of Home Owners’ Loan Act of 1933. First Federal S. & L. v. Massachusetts Tax Comm’n, p. 255. 2. State regulation of oil producers’ or refiners’ pricing practices—No pre-emption by federal laws.—Maryland statute requiring oil producers or refiners to extend “voluntary allowances” (temporary price reductions granted to independent retail service station dealers injured by local competitive price reductions) to all stations they supply is not pre-empted by § 2 (b) of Clayton Act, as amended by Robinson-Patman Act, or by Sherman Act. Exxon Corp. v. Governor of Maryland, p. 117. FEDERAL TAXES. See Internal Revenue Code. FIFTH AMENDMENT. See Constitutional Law, II; Criminal Law. FINAL DECISIONS. See Appeals, 1. FOREIGN PRODUCTS. See Tariff Act of 1930. FOURTEENTH AMENDMENT. See Constitutional Law, II, 4; III, 1; VI. FOURTH AMENDMENT. See Constitutional Law, VI. FREEDOM OF INFORMATION ACT. Unfair labor practice proceedings—Prehearing disclosure of witnesses’ statements.—Act does not require National Labor Relations Board to disclose witnesses’ statements prior to unfair labor practice hearing, and NLRB is entitled to withhold statements under Exemption 7 (A) of Act. NLRB v. Robbins Tire & Rubber Co., p. 214. GASOLINE STATIONS. See Constitutional Law, I, 3; III, 3; Federal- State Relations, 2. GOVERNMENT APPEALS. See Constitutional Law, II, 1, 2. HABEAS CORPUS. See Criminal Law. HOME OWNERS’ LOAN ACT OF 1933. See Federal-State Relations, 1. 920 INDEX HOSPITALS. See National Labor Relations Act, 1. IMPEACHMENT OF ACCUSED’S CREDIBILITY. See Constitutional Law, III, 1. IMPORTATION OF WASTE. See Constitutional Law, I, 2. IMPORTS. See Tariff Act of 1930. INCOME TAXES. See Constitutional Law, I, 1; III, 2. INDIANS. “Indian country”—State jurisdiction over offense.—Lands designated as a reservation for Choctaw Indians residing in central Mississippi are “Indian country,” as defined in 18 IT. S. C. § 1151 (1976 ed.) and as used in Major Crimes Act, and hence latter Act provided proper basis for federal prosecution of Choctaw Indian for offense occurring on such lands, and Mississippi had no power to prosecute him for same offense. United States v. John, p. 634. INJUNCTIONS. See Endangered Species Act of 1973. INTERFERENCE WITH UNFAIR LABOR PRACTICE PROCEED- INGS. See Freedom of Information Act. INTERLOCUTORY ORDERS REFUSING INJUNCTIONS. See Appeals, 2. INTERNAL REVENUE CODE. Summonses—District Court’s erroneous refusal to enforce.—District Court erred in refusing to enforce Internal Revenue Service summonses issued under § 7602 of Code on ground that they were not issued in good faith because they were issued “solely for the purpose of unearthing evidence of criminal conduct.” United States v. LaSalle National Bank, p. 298. INTERSTATE COMMERCE. See Constitutional Law, I; III, 2. INTERSTATE COMMERCE COMMISSION. See Antitrust Acts. INTERSTATE CORPORATIONS’ INCOME TAXES. See Constitutional Law, I, 1; III, 2. INTERVENTION BY UNITED STATES IN INTERSTATE COMMERCE COMMMISSION PROCEEDINGS. See Antitrust Acts. INVENTIONS. See Patents. INVOLUNTARY STATEMENTS BY ACCUSED. See Constitutional Law, III, 1. IOWA. See Constitutional Law, I, 1; III, 2. INDEX 921 ISOLATION SENTENCES. See Civil Rights. JAPAN. See Tariff Act of 1930. JUDGMENTS OP ACQUITTAL. See Constitutional Law, II, 3. JUDICIALLY IMPOSED REAPPORTIONMENT PLANS. See Elections. JURISDICTION. See also Appeals; Indians. Claim against third-party defendant—Lack of independent basis for jurisdiction.—In wrongful-death action in which federal jurisdiction was based on diversity of citizenship, District Court had no power to entertain plaintiff’s claim against third-party defendant as to which diversity jurisdiction was lacking. Owen Equipment & Erection Co. v. Kroger, p. 365. JUSTICIABILITY. See Mootness. LABOR UNIONS. See National Labor Relations Act. LEGISLATIVE REAPPORTIONMENT PLANS. See Elections. LIMITATION OF ACTIONS. See Antitrust Acts. LIMITATIONS ON SOLITARY CONFINEMENT. See Civil Rights. LIQUID WASTE DISPOSAL. See Constitutional Law, I, 2. LIST OF CLASS MEMBERS IN CLASS ACTION. See Federal Rules of Civil Procedure. LITERATURE DISTRIBUTION BY LABOR UNION. See National Labor Relations Act, 1, 3. LITTLE TENNESSEE RIVER. See Endangered Species Act of 1973. MAJOR CRIMES ACT. See Indians. MANDAMUS. Court of Appeals—Mandamus directing District Court to proceed— Concurrent state proceedings.—Court of Appeals’ judgment issuing writ of mandamus directing District Court to proceed to adjudicate claim based on Securities Exchange Act of 1934 despite pendency of a substantially identical proceeding between same parties in state courts, is reversed. Will v. Calvert Fire Ins. Co., p. 655. MARYLAND. See Constitutional Law, I, 3; III, 3; Federal-State Relations, 2. MASSACHUSETTS. See Federal-State Relations, 1. MATHEMATICAL FORMULAS AS PATENTABLE. See Patents. 922 INDEX METHOD FOR UPDATING ALARM LIMITS AS PATENTABLE. See Patents. MISSISSIPPI. See Indians. MONTANA. See Constitutional Law, II, 4. . MOOTNESS. Challenge to “Alaska Hire” statute—Effect of invalidation of durational residency requirement.—Challenge to "Alaska Hire” statute requiring oil and gas leases, easements or right-of-way permits for pipelines, and unitization agreements to contain requirement that Alaska residents be hired in preference to nonresidents, was not mooted by invalidation of 1-year durational residency requirement. Hicklin v. Orbeck, p. 518. MOTOR CARRIERS. See Antitrust Acts. MUNICIPAL CORPORATIONS. See Elections. “MURDER SCENE” EXCEPTION TO WARRANT REQUIREMENT. See Constitutional Law, VI. “MUTUAL AID OR PROTECTION’’ CLAUSE. See National Labor Relations Act, 3. NATIONAL LABOR RELATIONS ACT. 1. No-solicitation and no-distribution rule—Hospital cafeteria—Enforceability.—Court of Appeals did not err in enforcing National Labor Relations Board’s order to hospital to rescind its rule prohibiting employees from soliciting union support and distributing union literature during nonworking time in hospital cafeteria and coffeeshop. Beth Israel Hospital v. NLRB, p. 483. 2. Union—Disciplining of supervisory employee members—Unfair labor practice.—Labor union committed an unfair labor practice in violation of §8 (b)(1)(B) of Act when it disciplined members who were supervisory employees for crossing union’s picket line during a strike and performing their regular supervisory duties, which included adjustment of grievances. American Broadcasting Cos. v. Writers Guild, p. 411. 3. Union newsletter—Distribution in nonworking areas during nonworking time—Protection under Act.—Distribution of union newsletter in nonworking areas of employer’s property during nonworking time urging employees to oppose incorporation of state “right-to-work” statute into state constitution and criticizing Presidential veto of increase in federal minimum wage is protected under “mutual aid or protection” clause of § 7 of Act. Eastex, Inc. v. NLRB, p. 556. NATIONAL LABOR RELATIONS BOARD. See Freedom of Informa- tion Act; National Labor Relations Act, 1. INDEX 923 NEW JERSEY. See Constitutional Law, I, 2. NEWSLETTER DISTRIBUTION BY LABOR UNIONS. See National Labor Relations Act, 3. NEW TRIAL AFTER ACQUITTAL. See Constitutional Law, II, 2. NEW TRIAL AFTER REVERSAL OF CONVICTION FOR INSUFFI- CIENT EVIDENCE. See Constitutional Law, II, 3; Criminal Law. NONFEDERAL CLAIMS. See Jurisdiction. NO-SOLICITATION AND NO-DISTRIBUTION RULES. See National Labor Relations Act, 1, 3. NOTICE TO CLASS MEMBERS IN CLASS ACTIONS. See Federal Rules of Civil Procedure. OIL PRODUCERS OR REFINERS. See Constitutional Law, I, 3; III, 3; Federal-State Relations, 2. ORDERS DENYING CLASS CERTIFICATION. See Appeals. ORDERS REFUSING INJUNCTIONS. See Appeals, 2. OUT-OF-STATE WASTE. See Constitutional Law, I, 2. PATENTS. Method for updating alarm limits—Not patentable.—Method for updating alarm limits during catalytic conversion processes, in which only novel feature is a mathematical formula, is not patentable under § 101 of Patent Act. Parker v. Flook, p. 584. PICKET LINES. See National Labor Relations Act, 2. POTENTIAL WITNESSES’ STATEMENTS IN UNFAIR LABOR PRACTICE PROCEEDINGS. See Freedom of Information Act. PRE-EMPTION OF STATE LAW BY FEDERAL LAW. See Federal-State Relations, 2. PREHEARING DISCLOSURE OF WITNESSES’ STATEMENTS IN UNFAIR LABOR PRACTICE PROCEEDINGS. See Freedom of Information Act. PREJUDGMENT ORDERS DENYING CLASS CERTIFICATION. See Appeals. PRICE DISCRIMINATION. See Federal-State Relations, 2. PRISONS. See Civil Rights; Civil Rights Attorney’s Fees Awards Act of 1976; Constitutional Law, IV. PRIVILEGES AND IMMUNITIES CLAUSE. See Constitutional Law, V. 924 INDEX PROHIBITION AGAINST IMPORTATION OF WASTE. See Constitutional Law, I, 2. RACIAL DISCRIMINATION. See Elections. REAPPORTIONMENT PLANS. See Elections. REGULATION OF RETAIL GASOLINE MARKETING. See Constitutional Law, I, 3; III, 3; Federal-State Relations, 2. REMAND. See Criminal Law. REMEDIES FOR UNCONSTITUTIONAL PRISON CONDITIONS. See Civil Rights. RESERVOIRS. See Endangered Species Act of 1973. RESIDENCY REQUIREMENTS FOR EMPLOYMENT. See Constitutional Law, V; Mootness. RETAIL GASOLINE STATIONS. See Constitutional Law, I, 3; III, 3; Federal-State Relations, 2. RETRIAL AFTER ACQUITTAL. See Constitutional Law, II, 2. RETRIAL AFTER REVERSAL OF CONVICTION FOR INSUFFICIENT EVIDENCE. See Constitutional Law, II, 3; Criminal Law. REVERSAL OF CONVICTION FOR INSUFFICENT EVIDENCE. See Constitutional Law, II, 3; Criminal Law. ROBINSON-PATMAN ACT. See Federal-State Relations, 2. RULES OF CIVIL PROCEDURE. See Federal Rules of Civil Procedure. SAVINGS AND LOAN ASSOCIATIONS. See Federal-State Relations, 1. SEARCHES AND SEIZURES. See Constitutional Law, VI. SECOND TRIAL AFTER ACQUITTAL. See Constitutional Law, II, 2. SECOND TRIAL AFTER REVERSAL OF CONVICTION FOR INSUFFICIENT EVIDENCE. See Constitutional Law, II, 3; Criminal Law. SECRETARY OF THE INTERIOR. See Endangered Species Act of 1973. SECURITIES EXCHANGE ACT OF 1934. See Mandamus. SERVICE STATIONS. See Constitutional Law, I, 3; III, 3; Federal-State Relations, 2. SHERMAN ACT. See Federal-State Relations, 2. INDEX 925 SINGLE-FACTOR SALES FORMULA. See Constitutional Law, I, 1; III, 2. SINGLE-MEMBER DISTRICTS. See Elections. SNAIL DARTER. See Endangered Species Act of 1973. SOLICITATION OF EMPLOYEES BY LABOR UNIONS. See National Labor Relations Act, 1. SOLID WASTE DISPOSAL. See Constitutional Law, I, 2. SOLITARY CONFINEMENT. See Civil Rights. STATE EXCISE TAXES. See Federal-State Relations, 1. STATE INCOME TAXES. See Constitutional Law, I, 1; III, 2. STATE PROHIBITION AGAINST IMPORTATION OF WASTE. See Constitutional Law, I, 2. STATE PROSECUTION OF INDIANS. See Indians. STATE REGULATION OF RETAIL GASOLINE MARKETING. See Constitutional Law, I, 3; III, 3; Federal-State Relations, 2. STATUTES OF LIMITATIONS. See Antitrust Acts. STRIKES. See National Labor Relations Act, 2. SUMMONSES IN TAX INVESTIGATIONS. See Internal Revenue Code. SUPERVISORY EMPLOYEES. See National Labor Relations Act, 2. SWEARING OF FIRST WITNESS AS TIME WHEN JEOPARDY ATTACHES. See Constitutional Law, II, 4. TARIFF ACT OF 1930. Japan—“Bcrunty or grant”—Electronic products—No tax on exports.— Japan does not confer a “bounty or grant” within meaning of § 303 of Act on consumer electronic products by failing to impose a commodity tax on those products when they are exported to this country, while imposing tax on products when they are sold in Japan. Zenith Radio Corp. v. United States, p. 443. TAXES. See Constitutional Law, I, 1; III, 2; Federal-State Relations, 1; Internal Revenue Code. TAX-INVESTIGATION SUMMONSES. See Internal Revenue Code. TELLICO DAM. See Endangered Species Act of 1973. TENNESSEE VALLEY AUTHORITY. See Endangered Species Act of 1973. 926 INDEX TERMINATION OF TRIAL BY ACCUSED. See Constitutional Law, II, 1. THIRD-PARTY DEFENDANTS. See Jurisdiction. TIME WHEN JEOPARDY ATTACHES. See Constitutional Law, II, 4. TOLLING OF STATUTES OF LIMITATIONS. See Antitrust Acts. UNFAIR LABOR PRACTICES. See Freedom of Information Act; National Labor Relations Act. UNION NEW SLETTER. See National Labor Relations Act, 3. UNIONS. See National Labor Relations Act. UNITED STATES’ INTERVENTION IN INTERSTATE COMMERCE COMMISSION PROCEEDINGS. See Antitrust Acts. VOLUNTARINESS OF ACCUSED’S STATEMENTS. See Constitutional Law, III, 1. VOTING RIGHTS. See Elections. WARRANTLESS SEARCHES AND SEIZURES. See Constitutional Law, VI. WASTE DISPOSAL. See Constitutional Law, I, 2. WITNESSES’ STATEMENTS IN UNFAIR LABOR PRACTICE PROCEEDINGS. See Freedom of Information Act. WORDS AND PHRASES. 1. “Bounty or grant” §303, Tariff Act of 1930, 19 U. S. C. § 1303 (a) (1976 ed.). Zenith Radio Corp. v. United States, p. 443. 2. “Indian country.” 18 U. S. C. §§ 1151, 1153 (1976 ed.) (Major Crimes Act). United States v. John, p. 634. 3. “Instituted by the United States.” § 5 (i), Clayton Act, 15 U. S. C. § 16 (i) (1976 ed.). Greyhound Corp. v. Mt. Hood States, Inc., p. 322. 4. “Interfere with enforcement proceedings.” Exemption 7 (A) of Freedom of Information Act, 5 U. S. C. § 552 (b) (7) (A) (1976 ed.). NLRB v. Robbins Tire & Rubber Co., p. 214. WRITS OF MANDAMUS. See Mandamus. WRONGFUL-DEATH ACTIONS. See Jurisdiction.