[Senate Hearing 109-534]
[From the U.S. Government Publishing Office]



                                                        S. Hrg. 109-534
 
           SAGO MINE DISASTER AND AN OVERVIEW OF MINE SAFETY

=======================================================================

                                HEARING

                                before a

                          SUBCOMMITTEE OF THE

            COMMITTEE ON APPROPRIATIONS UNITED STATES SENATE

                       ONE HUNDRED NINTH CONGRESS

                             SECOND SESSION

                               __________

                            SPECIAL HEARING

                    JANUARY 23, 2006--WASHINGTON, DC

                               __________

         Printed for the use of the Committee on Appropriations


  Available via the World Wide Web: http://www.gpoaccess.gov/congress/
                               index.html



                    U.S. GOVERNMENT PRINTING OFFICE
27-461                      WASHINGTON : 2006
_____________________________________________________________________________
For Sale by the Superintendent of Documents, U.S. Government Printing Office
Internet: bookstore.gpo.gov  Phone: toll free (866) 512-1800; (202) 512ï¿½091800  
Fax: (202) 512ï¿½092250 Mail: Stop SSOP, Washington, DC 20402ï¿½090001

                               __________

                      COMMITTEE ON APPROPRIATIONS

                  THAD COCHRAN, Mississippi, Chairman
TED STEVENS, Alaska                  ROBERT C. BYRD, West Virginia
ARLEN SPECTER, Pennsylvania          DANIEL K. INOUYE, Hawaii
PETE V. DOMENICI, New Mexico         PATRICK J. LEAHY, Vermont
CHRISTOPHER S. BOND, Missouri        TOM HARKIN, Iowa
MITCH McCONNELL, Kentucky            BARBARA A. MIKULSKI, Maryland
CONRAD BURNS, Montana                HARRY REID, Nevada
RICHARD C. SHELBY, Alabama           HERB KOHL, Wisconsin
JUDD GREGG, New Hampshire            PATTY MURRAY, Washington
ROBERT F. BENNETT, Utah              BYRON L. DORGAN, North Dakota
LARRY CRAIG, Idaho                   DIANNE FEINSTEIN, California
KAY BAILEY HUTCHISON, Texas          RICHARD J. DURBIN, Illinois
MIKE DeWINE, Ohio                    TIM JOHNSON, South Dakota
SAM BROWNBACK, Kansas                MARY L. LANDRIEU, Louisiana
WAYNE ALLARD, Colorado
                    J. Keith Kennedy, Staff Director
              Terrence E. Sauvain, Minority Staff Director
                                 ------                                

 Subcommittee on Departments of Labor, Health and Human Services, and 
                    Education, and Related Agencies

                 ARLEN SPECTER, Pennsylvania, Chairman
THAD COCHRAN, Mississippi            TOM HARKIN, Iowa
JUDD GREGG, New Hampshire            DANIEL K. INOUYE, Hawaii
LARRY CRAIG, Idaho                   HARRY REID, Nevada
KAY BAILEY HUTCHISON, Texas          HERB KOHL, Wisconsin
TED STEVENS, Alaska                  PATTY MURRAY, Washington
MIKE DeWINE, Ohio                    MARY L. LANDRIEU, Louisiana
RICHARD C. SHELBY, Alabama           RICHARD J. DURBIN, Illinois
                                     ROBERT C. BYRD, West Virginia (Ex 
                                         officio)
                           Professional Staff
                            Bettilou Taylor
                              Jim Sourwine
                              Mark Laisch
                         Sudip Shrikant Parikh
                              Candice Ngo
                             Lisa Bernhardt
                              Rachel Jones
                        Ellen Murray (Minority)
                         Erik Fatemi (Minority)
                      Adrienne Hallett (Minority)


                            C O N T E N T S

                              ----------                              
                                                                   Page

Opening statement of Senator Arlen Specter.......................     1
Statement of Senator Robert C. Byrd..............................     2
Statement of Senator Tom Harkin..................................     5
Statement of Senator Mike DeWine.................................     7
Statement of Hon. David G. Dye, Acting Assistant Secretary, Mine 
  Safety and Health Administration, Department of Labor..........     8
    Prepared statement...........................................     9
Statement of Ray McKinney, Administrator for Coal Mine Safety and 
  Health, Mine Safety and Health Administation, Department of 
  Labor..........................................................    11
Statement of Cecil E. Roberts, International President, United 
  Mine Workers of America, Fairfax, Virginia.....................    21
    Prepared statement...........................................    22
Statement of Bennett K. Hatfield, President and CEO, 
  International Coal Group, Ashland, Kentucky....................    28
    Prepared statement...........................................    29
Statement of Chris R. Hamilton, Senior Vice President, West 
  Virginia Coal Association, Charleston, West Virginia...........    32
    Prepared statement...........................................    34
Statement of Davitt McAteer, Vice President of Sponsored 
  Programs, Wheeling Jesuit University, Shepherdstown, West 
  Virginia.......................................................    38
    Prepared statement...........................................    39
Statement of Bruce Watzman, Vice President, Safety and Health, 
  National Mining Association, Washington, DC....................    43
    Prepared statement...........................................    45
Prepared statement from the National Stone Sand & Gravel 
  Association....................................................    63
Responses of David G. Dye to committee questions.................    65
Responses of Bennett K. Hatfield to committee questions..........    78
Responses of Cecil E. Roberts to committee questions.............    80


           SAGO MINE DISASTER AND AN OVERVIEW OF MINE SAFETY

                              ----------                              


                        MONDAY, JANUARY 23, 2006

                           U.S. Senate,    
    Subcommittee on Labor, Health and Human
         Services, Education, and Related Agencies,
                               Committee on Appropriations,
                                                    Washington, DC.
    The subcommittee met at 11 a.m., in room SH-216, Hart 
Senate Office Building, Senator Arlen Specter (chairman) 
presiding.
    Present: Senators Specter, DeWine, Harkin, and Byrd.


               OPENING STATEMENT OF SENATOR ARLEN SPECTER


    Senator Specter. Good morning, ladies and gentlemen. The 
Appropriations Subcommittee on Labor, Health and Human Services 
and Education will now proceed with this hearing, which will 
inquire into mine safety on coal mines.
    This hearing is prompted by the disaster at the Sago Mine 
on January 2. There have since been two more fatal accidents in 
West Virginia. I thank the distinguished ranking member of the 
full committee former president pro tempore Senator Byrd for 
his requesting the hearing. This subcommittee has been active 
on the investigation of mine disasters in the past, notably the 
Quecreek, which occurred in Pennsylvania in the year 2002, and 
there are very, very important questions on the issue on what 
has happened at these mine incidents, focusing on Sago.
    The Pittsburgh Post Gazette set the stage for our 
consideration here in a very poignant statement as follows: 
``The rest of the world will move on--in their reference to the 
Sago Mine Accident. In the weeks and months to come, there will 
be other disasters, other wars, other political scandals. But 
for the families of the 12 men who died in the mine at 
Tallmansville, West Virginia, for the one who survived and 
their relatives and friends, for the investigator searching for 
the cause of the mine explosion, for the people of these coal-
rich hills 100 miles south of Pittsburgh, Sago will be a daily 
litany. Some questions about the January 2 accident may never 
be answered, but there is understanding to be gained by 
reconstructing what happened''.
    We all know that mining is a very dangerous business. As I 
noted a moment ago, two underground coal mine incidents this 
January after the Sago disaster. On January 10, there was a 
miner killed in Kentucky after a roof cave-in. On January 19, 
two miners became trapped in the Aracoma Mine in Melville, West 
Virginia. Both of them were fatalities. Last year, the safest 
year on record, there were 22 fatalities in underground coal 
mines in 20 separate accidents. Four deaths in my home State, 
Pennsylvania, three in West Virginia, eight in Kentucky and 
seven in other States.
    The Sago Mine had some 208 citations, orders and safeguards 
issued against it last year, 2005, with nearly half of these 
violations categorized as ``significant and substantial''. The 
budget for mine and health safety has increased by some 42 
percent over the past 10 years, but these increases have barely 
kept up with inflationary costs. This has forced the agency to 
reduce staffing, we're told, by some 183 positions over that 
period of time. Coal mine staffing has declined by 217 
enforcement personel in the last 5 years. One of the grave 
difficulties we have had in maintaining funding for mine safety 
is the fact that the Department of Labor, which this committee 
has jurisdiction over for appropriations purposes, is grouped 
with the Department of Education and the Department of Health 
and Human Services, and there are many demands on this 
subcommittee. Senator Harkin, the distinguished ranking member, 
Senator Byrd, and I, as well as others on this subcommittee 
have advocated a more realistic budget through this 
subcommittee, which has jurisdiction over worker safety, 
jurisdiction over education and jurisdiction over healthcare. 
We were underfunded in the last budget resolution, and then we 
sustained a 1 percent across-the-board cut, and I've already 
put the leadership on notice in the Senate that I'm not going 
to support a budget resolution that does not adequately fund 
this department next year or not adequately fund this 
subcommittee next year. Of course, I hadn't anticipated the 
Sago Mine Disaster, but we saw from Quecreek, and we've seen 
from other disasters the importance of having adequate funding. 
Now, let me yield at this time, it's always a difficult matter 
whether to recognize the ranking member or the senior member, 
but I've already had a note here from the generous ranking 
member, Senator Harkin, asking or instructing that I recognize 
Senator Byrd first. Senator Byrd, will you accept that 
recognition?


                  STATEMENT OF SENATOR ROBERT C. BYRD


    Senator Byrd. Mr. Chairman, I do. First, let me compliment 
you, and Senator Harkin. Let me thank you and Senator Harkin 
for the great work you do on this committee, and the great work 
that you have done in the interest of safety and education and 
other matters that come under the jurisdiction of this 
committee. You have been very, very great and fine, generous, 
considerate in this particular situation. Each of us has coal 
mines in his or her State, and it is very important that the 
people out there understand that this subcommittee is going to 
do everything it can by way of funding and oversight to see 
that the safety and health of coal miners are not underfunded 
as they have been in the past. I'll get more into that. Shall I 
do that right now?
    Senator Specter. Please do, Senator Byrd.
    Senator Byrd. I thank you, Mr. Chairman. Fourteen men have 
died in the coal fields of West Virginia, 14 men in the span of 
3 weeks. These deaths, I believe, were entirely preventable, 
and we owe the families of these deceased and noble, and great, 
and brave men a hard look at what happened and why. I'm 
striving to be measured in my statements. I have deep feelings 
in this matter, but I'm going to try to be measured and careful 
in my questions, but why was it that the Federal agency charged 
with safety--the Federal agency charged with safety of this 
country's miners, failed to protect them. Why was it? Well, let 
me be clear.
    I have no complaint with the brave rescuers who risked 
their lives to try to save trapped miners, or with the rank and 
file former miners who work for MSHA in the field. I wish every 
American could know the total, the absolute total commitment of 
these folks of these people and understand the dangers faced by 
rescue teams, whether from secondary explosions like this one 
at the Sago Mine in Upshur County, West Virginia or the extreme 
heat of the raging underground fire at the Alma Mine in Logan 
County, West Virginia. I'm talking today. Let me be clear. I'm 
talking today about the leadership in MSHA's Washington office 
have decisions to cut the mining health and safety enforcement 
budget and staff, and abandon the critical mining regulations 
endangered of the safety of these miners of Buckhannon and 
Melville and Lord knows how many other communities in West 
Virginia and across the country.
    Federal and State investigators must explain what happened 
at the Sago Mine and at the Alma Mine and whether those 
tragedies could have been prevented. The Congress, however, has 
a broader responsibility. We must try to determine what is 
wrong at MSHA and contemplate how to make sure that the 
leadership of that Agency does its job. Why did it take 2 hours 
before MSHA was notified of the Sago Explosion and another 2 
hours before MSHA personnel arrived at the scene? Why another 
1.5 hours before rescue teams arrived and another 5 hours 
before the first team entered the Sago Mine? The Mine Act 
requires that rescue teams be readily available in the event of 
an emergency, and yet, it took 11 hours, 11 hours before the 
first team could even begin the rescue effort at Sago, 11 hours 
in the darkness. Where was the help? Is help coming? Our lights 
have gone out. It's dark. I can't see.
    It took 11 hours before the first team could even begin the 
rescue effort. Meanwhile, 12 coal miners, 12 at Sago, waited in 
the dark. The horrors, can one imagine? Twelve miners at Sago 
waited and waited and waited and waited in the dark 260 feet 
below the surface armed only with a 1-hour breathing device, 1 
hour of precious air, grossly inadequate for miners whom we 
know may have to barricade themselves and possibly wait many 
hours, even a day, even 2 days, even more for rescue. The 
miners could not communicate with the teams involved in the 
rescue effort underway, and the rescuers could not identify the 
location of the miners. How sad.
    Can you imagine the families of those miners waiting, 
waiting, waiting? Why, in an age of technological marvels, was 
the only apparent protection for these 12 souls trapped 
underground a brattice cloth intended to divert toxic gases 
from their barricade? Now, a short 2 weeks later, the Nation 
watched as similar horrors emerged from a second tragedy at the 
Alma Mine. Delayed notification to MSHA of 2.5 hours, no 
ability to locate the lost miners and no means for any 
survivors underground to communicate with rescue teams. Put 
yourself in that situation, how horrible that is, unbelievable 
in this age.
    The administration's budgets have translated into 4-year 
cuts in mine health and safety enforcement resources and 
personnel. Let me say that again. It's worth saying. The 
administration's budgets have translated into 4-year cuts in 
mine health and safety enforcement resources and personnel. 
This past week, or this past weekend, President Bush called my 
house to talk with me on another matter and in passing noted of 
the Alma Mine fire and asked if he could do anything. Here was 
the President, President Bush, asking me if he, the President, 
could do anything. I said, Yes. Stop cutting the coal health 
and safety enforcement budget. We need more enforcement, not 
less. We should be demanding a rapid response system in case of 
a mine emergency. We should be demanding new communications 
technology and emergency breathing equipment. We must look at 
the safety and health regulations that MSHA has on the books as 
well as those that have been discontinued. Those rules govern 
everything from mine rescue teams to the use of belt entries 
for ventilation. Let me say that again. Everything from mine 
rescue teams those rules govern to the use of belt entries for 
ventilation to inspection procedures, to emergency breathing 
equipment, to escape routes and rescues and refuges, anyone of 
which could figure into the disaster at the Sago and the Alma 
mines. How about that?
    What should be done about the pattern of violations that 
existed at the Sago Mine, and how has MSHA employed its 
statutory authority to pursue habitual violators? Fines may 
have been levied, but the sheer number of violations suggests 
that those penalties are failing to induce mine operators to 
faithfully comply with the law. Mine safety demands aggressive 
leadership.
    The coal miners in my State, Senator Rockefeller and I 
speak for them. The coal miners of my State have been doing 
this kind of work for a long, long time. I should say here 
again that Senator Rockefeller would have been here today at 
this hearing, but he could not be because of personal matters, 
important matters. Senator Rockefeller's heart is right at the 
center, and he works and hopes and feels like a Senator should 
whose coal miners are losing their lives. He has worked hard. 
Mine safety, as I say, demands aggressive leadership. The 
miners in my State are incredibly brave. Coal miners are a 
different breed. I know. They will go into mines after a 
disaster like this and risk their own lives to try to help 
their buddies, try to find their comrades, their companions. 
They are brave--brave, and so are their families. They know 
there's danger there. All the time, they know it. Mine safety 
demands, as I say, aggressive leadership, and these people are 
very resourceful when confronted with an emergency. MSHA and 
the Labor Department are filled with career professionals with 
hundreds of years of accumulated experience in coal safety.
    Given the resources of this Government and the statutory 
authority of The Mine Act, most mining fatalities should be 
preventable. We've come a long way, but we have more to do. The 
Congress must not tolerate complacency about the health and 
safety of the brave men and women who risk their lives to 
provide energy, power to run these lights, which make it 
possible for us here, the energy and power in time of peace and 
in time of war. Profits should never come before protection for 
the coal miners.
    Did you hear me out there? Profits should never come before 
the safety and welfare and wellbeing of the brave men and women 
who go into the bowels of the earth and scratch the coal out, 
sometimes with their hands, with water seeping from overhead 
and with danger all around. Politics must never play a role in 
the enforcement of safety and health regulations. We must not 
allow the call for action in the wake of this tragedy to recede 
without stiffening mine enforcement and emergency preparedness.
    Senator Specter and Senator Harkin, I thank you again for 
scheduling this hearing. I hope that this is the beginning of 
an effort that will yield a worthy legacy for the Sago and Alma 
miners and for other miners throughout West Virginia and 
throughout the country. Thank you very much.
    Senator Specter. Thank you very much, Senator Byrd. As 
Senator Byrd has noted, Senator Rockefeller could not be here 
today, he is out of town. Similarly, Senator Santorum could not 
be here today because he is out of town, and I know both of 
those Senators and doubtless others would have wanted to have 
been here.
    I might say, by one word of explanation, that the Senate is 
not in session today although this room has been fully occupied 
by the Alito hearing for many, many days, so the Judiciary 
Committee members are here in any event as is Senator DeWine. I 
want to yield now to my distinguished ranking member. We have a 
custom of 5-minute opening statements, but the clock got stuck 
when Senator Byrd was speaking. We'll give him a little more 
latitude.
    Senator Byrd. Thank you, Mr. Chairman.
    Senator Specter. Senator Byrd has been in the Congress of 
the United States since Harry Truman was President, elected to 
the House in 1952 and the Senate in 1958, and it is quite a 
record, so we have a little more latitude.
    Senator Byrd. You have been very kind, Mr. Chairman, very 
courteous as is characteristic of you. Thank you.
    Senator Specter. You deserve it, Senator Byrd. Senator 
Harkin.


                    STATEMENT OF SENATOR TOM HARKIN


    Senator Harkin. Thank you, Mr. Chairman, and thank you for 
hosting this hearing. I especially want to thank Senator Byrd 
for requesting this hearing. Over the years, I've listened to 
Senator Byrd speak eloquently about the mining communities of 
West Virginia. We heard it again here this morning. I know that 
there is no greater advocate for the families of miners, no 
more dedicated public servant for the people of West Virginia 
than Senator Robert C. Byrd, so I thank him for requesting this 
hearing, but also for the dedication that he has exhibited over 
all his lifetime in fighting for miners not only in West 
Virginia, but all over the country.
    I know that he will continue to hold the folks of Congress 
on this important matter, and for that we should all be 
grateful. Although I'm far from the first, I offer my sincere 
condolences to all the families of the fallen miners, the 12, 
the 4 and the 2 on Saturday in the Logan County Fire. As the 
son of a coal miner myself, your families remain in my thoughts 
and prayers.
    My father was born in 1886, started mining coal somewhere 
around 1910, in Iowa. A lot of people don't know that there 
were coal mines in Iowa. There used to be a coal mine capital. 
There's still a lot of coal there, but that's soft coal. They 
don't mine it any longer. A great coal mine leader, John L. 
Lewis, many people think he came from Pennsylvania or West 
Virginia. He actually came from Iowa.
    That is where he was born and raised, and that's where he 
mined coal, and that's where his brother mined coal all his 
life in Iowa. So we had a lot of coal mines in those days. My 
father mined coal for 23 years in a coal mine in Iowa. And a 
lot of coal mines in Iowa. It wasn't till the depression hit 
that he finally left the mines, but the mines never left him 
because later in life, he had what we all called miner's lung. 
We never called it black lung, never heard of that word before. 
Black miner's lung, every miner had it.
    Of course, it didn't help that he smoked all the time too, 
but he had miner's lung, and that is what finally got him. I 
can remember as a kid, listening to my father tell me stories 
of working in the coal mines. Now, you're talking about the 
teens, World War I and the 20s and going down in these shaft 
mines a couple of hundred feet and going out in the laterals 
and telling me about how dark it was and the cave-ins and the 
miners that would get killed. Trying to get air down, they 
would pump air down in these mines. He would describe it to me. 
They would take these rickety old elevators down first thing in 
the morning before the sun would come up. Sometimes he would 
come back out of the mines after the sun went down. He never 
saw the sun during the day. Twelve hours a day they worked down 
in those mines and never see the sun.
    Well, he was one of the lucky ones, but he told me, and I 
used to see his friends as I was a kid growing up, some of the 
old miners would come around, people from the old country, the 
Italians, the Slovenes, the Welsh, the Irish, would all come 
around, all the old miners, and they would tell their stories. 
As a kid, I used to listen to these. They were frightening 
stories. They all would tell the stories of their friends who 
had died in the mines, and to me, it just painted a very 
awesome portrait of what it meant to be a miner and how brave 
these people were in what they did. Of course, that was before 
any safety regulations or anything.
    So, I come at this with that kind of a background, and I 
wonder why it is--why it is--why it is over--after over a 
hundred and some years of mining coal in this country, why do 
we still have mining tragedies like this? Now, I know it's a 
dangerous occupation. People make mistakes. It's a very 
dangerous occupation, but still, accidents like these are 
avoidable. Why--why don't we have, after all this time, better 
coordinated risk procedures?
    Now we have regulations for rescue procedures, but it seems 
like they more often fall into the breach rather than the 
letter. Why, after 208 violations in 2005, 95 of them, 40 
percent were for serious and substantial violations at Sago, 
why wasn't something done? Technology. Why? In 1969, the Bureau 
of Mines, it's not called that anymore, promoted a handheld 
system so miners could find their exact location, and they 
could be found. That was used in Utah a couple of years ago in 
a disaster.
    I have a nephew of mine, my sister's boy, he's a miner, 
goes down everyday in the mines, not coal mines, Trona Mine. 
Soda Ash, Wyoming, deep mines, they go down a long way. He's 
been mining there for over 20 some years in that Trona Mine. 
He's on a rescue team. They have competitions to see how they 
would set up. They don't have those kind of accidents. Why, in 
a Soda Ash Mine in Wyoming, don't they have the kind of 
accidents that we're talking about here? Why is it just coal 
miners? Why is it just coal miners? Are they less valuable or 
something? Is there something less valuable about coal?
    As Senator Byrd eloquently said, that's what makes our 
lights light. It makes my timer go down and says I'm overdue. 
It seems to me that we do have other mines, and they operate 
more safely, and they have better procedures. Why haven't we 
done this? It seems to me, and I'll just end on this, Mr. 
Chairman, it just seems to me that we have ignored this and 
ignored it and ignored it and put it off and put it off. We've 
cut the budgets. We don't have the safety procedures. We don't 
have the handheld devices so that we can find a miner and his 
exact location. That is possible. It can be done. It's not that 
expensive. So why haven't we done it? These are the questions 
that MSHA needs to answer. Thank you. Thank you, Mr. Chairman.
    Senator Specter. Thank you very much, Senator Harkin. 
Senator DeWine.


                   STATEMENT OF SENATOR MIKE DE WINE


    Senator DeWine. Thank you, Mr. Chairman. Mr. Chairman, 
first, thank you very much for holding this hearing. Our 
thoughts and prayers go out to the families of the miners who 
have lost their lives, and we think about them this morning.
    Miners are a courageous group of people. Mining is horribly 
dangerous work. It takes a strong individual to endure the 
physical as well as the mental toughness needed to work in this 
industry. Everyday these miners put their lives on the line so 
that we can have electricity, so we enter a room and flip on a 
switch, the light does, in fact, come on. Coal plays such a 
significant role as an energy supply for our Nation. Coal 
generates more than half of the Nation's electricity. Eighty-
seven percent of my home State of Ohio is electric power, is 
generated by coal. With a 250 to 300 year supply, coal is the 
only fossil fuel this country has in great abundance. Coal is 
part of this country's future.
    Mr. Chairman, and members of the committee, we simply have 
to figure out how we can mine more safely than we do today. 
Obviously, progress has been made, but we have to do better, 
and these tragedies that this country has suffered, these 
families have suffered, brings that home to us.
    Mr. Chairman, I know our witnesses this morning who we're 
going to hear in just a moment will help give us insight into 
what we can do to see this vision move forward, and so I thank 
you for holding this hearing. And also as a member of the full 
Health Committee, I'll be working with Chairman Enzi when our 
committee holds its future hearings as well. So again, Mr. 
Chairman, I thank you for holding this very important hearing 
today. I thank our witnesses. And again, my thoughts and 
prayers remain with the families of the miners that have been 
lost in the recent tragedies.
    Senator Specter. Thank you very much, Senator DeWine. We 
now move to our first panel. We have with us the Acting 
Assistant Secretary of Labor for Mine Safety and Health, Mr. 
David Dye, U.S. Department of Labor, bachelor's degree from 
University of Texas at Austin and law degree from Franklin 
Pierce Law Center in New Hampshire. Mr. Dye's accompanied by 
Mr. Ray McKinney, Administrator of the Coal Mine Safety and 
Health unit, Mr. Bob Friend, Deputy Assistant Secretary for 
Mine Safety and Health, no relation to the former Pittsburgh 
pitcher, and Mr. Ed Clair, Associate Solicitor for Mine Safety 
and Health.
    We've allocated 8 minutes to Mr. Dye for a presentation, 
which will be joined in by Mr. McKinney. Our standard practice 
is to allocate 5 minutes for each witness. We have quite an 
extensive hearing today with good representation on the panel, 
especially for a day when the Senate is not in session. Thank 
you for joining us, Mr. Dye, and we look forward to your 
testimony.

STATEMENT OF HON. DAVID G. DYE, ACTING ASSISTANT 
            SECRETARY, MINE SAFETY AND HEALTH 
            ADMINISTRATION, DEPARTMENT OF LABOR

ACCOMPANIED BY:
        RAY McKINNEY, ADMINISTRATOR FOR COAL MINE SAFETY AND HEALTH, 
            MINE SAFETY AND HEALTH ADMINISTRATION, DEPARTMENT OF LABOR
        ROBERT M. FRIEND, ADMINISTRATOR FOR METAL AND NONMETAL MINE 
            SAFETY AND HEALTH, MINE SAFETY AND HEALTH ADMINISTRATION, 
            DEPARTMENT OF LABOR
        ED CLAIR, ASSOCIATE SOLICITOR, DEPARTMENT OF LABOR
    Mr. Dye. Thank you, Mr. Chairman, Senator Byrd, members of 
the committee. Thank you for this opportunity to appear before 
you to discuss the work of the Mine Safety and Health 
Administration and its performance related to the January 2 
Sago Mine Explosion in Tallmansville, West Virginia.
    My heart is especially heavy today as I appear before you 
just days after the Aracoma Mine accident in Melville, West 
Virginia, that claimed two lives. I would like to commend the 
21 rescue teams involved in the Aracoma Mine rescue efforts 
this weekend and the 13 mine rescue teams involved at Sago Mine 
who put their own lives on the line in the quest to save their 
fellow miners. Most of these brave rescuers are volunteers. All 
of them willingly face danger and give generously of their time 
and in constant training. We cannot thank them enough for their 
help and their sacrifice; they are the embodiment of American 
spirit, selflessness and courage.


                           PREPARED STATEMENT


    All of us at MSHA deeply grieve the loss of these miners' 
lives, and I want to assure the families and friends that we 
are conducting uncompromising investigations of these 
accidents. We will uncover the truth why both of those 
tragedies happened and how we can better protect miners and 
prevent tragedies such as these in the future. Mr. Chairman, my 
colleague, Ray McKinney, will now give a short presentation on 
the events following the explosion at Sago Mine.
    [The statement follows:]

                Prepared Statement of Hon. David G. Dye

    Mr. Chairman: Thank you for this opportunity to appear before the 
Subcommittee to discuss the work of the Mine Safety and Health 
Administration (MSHA) and its performance at the January 2 Sago Mine 
explosion in Tallmansville, Upsur County, West Virginia. We have only 
preliminary information on the mine accident in Logan County, West 
Virginia, but I will be happy to share what we know thus far with the 
committee as well.
    It is with the heaviest of hearts that every MSHA employee grieves 
for the miners who died at the Sago mine and the loved ones who mourn 
their passing.
    MSHA's reason for being is to ensure that miners return home safe 
and healthy to loved ones at the end of their shifts. That is our 
mission and our focus, every day. That is our duty to America's miners 
and the reason we are conducting the investigation with the greatest 
care and diligence so we can uncover the full truth of why this tragedy 
happened and how we can better protect miners in the future.
    Although MSHA has significantly stepped-up enforcement in recent 
years and dramatically reduced injury and fatality rates--mining 
fatalities dropped 33 percent and total injuries dropped 25 percent 
from Calendar Year (CY) 2000 to CY 2005--the Sago mine tragedy must 
cause us to carefully assess MSHA's performance, as the Subcommittee is 
doing today. From CY 2000 to CY 2005, the number of coal mine 
fatalities dropped 42 percent and the number of coal mine injuries 
dropped 22 percent, but that is cold comfort to the families of those 
who are killed, or to miners who suffer injuries. So we can never let 
up in improving mine safety, and we will not.
    In addition to emphasizing increased enforcement of mine safety 
laws, MSHA has promoted prevention and worked with the mining 
community--operators, miners, state and local regulators, trade and 
labor organizations, manufacturers, suppliers and others--to make 
America's mines the safest in the world. Collaboration on safety 
training, technical assistance, and education programs is essential and 
has been productive in helping miners, supervisors and employers adopt 
the safest and healthiest mining practices.
    These efforts complement our strong enforcement regime and our 
enforcement results from CY 2000 to CY 2005 are substantial:
  --Total Citations and Orders issued by MSHA at all mines increased by 
        6 percent, from 120,050 to 127,682;
  --Total Citations and Orders issued by MSHA at coal mines increased 
        by 18 percent, from 58,304 to 68,818; and
  --Total ``Significant and Substantial'' Citations and Orders issued 
        at coal mines increased by 11 percent, from 23,994 to 26,779.
    While enforcement went up, the number of coal mines fell. There 
were 2,124 coal mines in 2000 and 1,982 in 2005 (through the third 
quarter).
    MSHA enforcement personnel have also significantly increased the 
issuance of withdrawal orders that force coal mine operators who 
exhibit an unwarrantable failure to comply with the regulations to shut 
down areas or operations of the mine affected by the hazard. 
Unwarrantable failure orders are one of the most severe enforcement 
actions inspectors can take. And unwarrantable failure orders issued to 
coal mine operators are up 35 percent over the last five years compared 
to the previous five year period.
    MSHA's increased enforcement has also resulted in increased 
assessments--penalties imposed on mine operators for violating health 
and safety standards or the Mine Act. From CY2001 to 2005, the number 
of high-dollar final assessments imposed on all mine operators was 21 
percent higher than during the period of 1996-2000. The total dollar 
value was up by 16 percent during this same period of time. This 
increase in high dollar assessments occurred while the total number of 
mines decreased by 8 percent between 2000 and 2005. There is more 
enforcement, more assessments and more scrutiny of mine operators than 
ever before.
    Under the Federal Mine Safety and Health Act, whether issued in a 
citation or an order of withdrawal, each violation is assessed a 
penalty (statutorily capped at $60,000) which must be assessed in 
accordance with the law's six statutory criteria. The Administration 
believes that the statutory cap is too low to deter repeat and 
egregious violations of the Mine Safety and Health Act, and has urged 
the Congress to increase the statutory cap from $60,000 to $220,000. 
This would bring the Mine Act's civil monetary penalties in line with 
those authorized by the Occupational Safety and Health Act. The 
Administration again encourages Congress to act on this proposal to aid 
MSHA in enforcing the Mine Act.
    The statute itself provides for a graduated enforcement scheme that 
provides for increasingly severe sanctions for increasingly serious 
violations or an operator's unwarrantable failure to comply with a 
mandatory standard. A chain of increasingly severe enforcement actions 
causing all persons, except those necessary to eliminate the condition, 
to be withdrawn from that area until the condition is abated, serves as 
an incentive for operator compliance and is the statute's most powerful 
instrument for enforcing mine safety.
    In every link of the chain, a withdrawal order is limited by 
statute to the area of the mine affected by the violation or unsafe 
condition. In some instances, the area affected could be the entire 
mine, or the area affected could be so critical or integral to the 
operation of the mine that its closure effectively shuts down the mine 
(for example, when the area affected includes the hoist used to move 
personnel and material in and out of the mine, or a blocked escape 
way). However, as soon as the operator corrects the cited condition and 
an MSHA inspector terminates the order, the operator is free to re-
enter and resume mining work in that area. At the end of the graduated 
chain of enforcement, even a relatively minor violation can trigger a 
withdrawal order. A series of severe withdrawal orders affecting 
critical areas of the mine can inflict a severe economic penalty on the 
operator and could even lead to an operator ceasing operations. 
However, the statute does not allow preemptive permanent closure of a 
mine that has abated all violations.
    MSHA enforcement personnel were vigilant in dealing with Anker 
Mining, the operator of Sago mine until May of 2005 and International 
Coal Group (ICG), the successor mine operator. As incidents increased 
at the mine in 2005, so did the issuance of MSHA citations and 
increasingly severe withdrawal orders, in accordance with the Mine Act. 
The enforcement hours spent at the mine in 2005 demonstrate the 
increased attention the mining operation received: MSHA spent 744 on-
site inspection hours in 2005 at the Sago mine, up by 84 percent over 
the 405 hours spent during calendar year 2004. MSHA inspectors were at 
the Sago Mine on 93 different days in 2005, and often more than one 
inspector was at the mine. Inspectors issued 208 citations, orders, and 
safeguards in 2005, including withdrawal orders that shut down areas of 
the mine on 18 occasions. In addition, between April 27, 2005 and 
December 15, 2005, MSHA managers and supervisors met with ICG officials 
approximately 21 times.
    MSHA was ramping up enforcement activity at this mine by issuing 
these citations and insisting that the cited hazards be promptly 
corrected. Each one of those citations represented a hazard that was 
required to be eliminated within a fixed abatement time specified by 
the inspector--and the inspector followed up to ensure the hazards were 
indeed fixed. That meant extra inspector time in the mine, and 
inspector time talking with mine operators, management and supervisors 
to ensure that those hazards were indeed corrected. Sago Mine 
management also received substantial engineering consulting services 
from MSHA's Office of Technical Support on major issues at the mine 
including water removal, roof control, dust control and other problems. 
We were beginning to see positive results as the all-injury incidence 
rate at this mine dropped from 55.8 in the 2nd Quarter to 8.3 in the 
3rd Quarter.
    While there were many problems at the Sago Mine, liberation of 
large quantities of methane had not been among them. Methane is 
explosive between 5-15 percent. At one percent, MSHA requires operators 
to take action. At the Sago Mine, the highest readings never exceeded 
two-tenths of one percent, which is one-twenty-fifth of the explosive 
level. MSHA's routine monitoring of methane in the mine never warranted 
a citation for excess accumulations of methane gas.
    At this time, we have no information that would suggest that the 
explosion was related to a recurrence of any of the conditions that 
were required to be abated before the explosion. We will, of course, 
aggressively pursue these and all other potential causes in our joint 
accident investigation.
    Until the joint investigation team can safely enter the mine to 
thoroughly examine the site, we will not know the answers to these 
questions.
    It is a sense of duty, fellowship and a strong measure of heroism 
that mine rescue teams took with them into the Sago Mine in the quest 
to save their colleagues. Thirteen teams comprised of 109 dedicated 
team members participated in the rescue and recovery operation, putting 
their lives on the line for their fellow miners. Their bravery and 
dedication are emblematic of the 314 mine rescue teams around the 
country. It is noteworthy that the vast majority of rescue team members 
are volunteers--good Samaritans in the highest American tradition. It 
is extraordinarily courageous and generous to take on the 
responsibility of such perilous rescue missions, knowing that heartache 
may mark the end of the day. We cannot thank them enough for their help 
and their sacrifice.
    Some observers have raised concern that the Sago explosion occurred 
at 6:30 a.m. and the first rescue team did not enter the mine until 
5:25 p.m. The delay was out of concern for these rescuers' lives. MSHA 
and the mining community have a history of hard-learned lessons on the 
peril of rushing into mine accident scenes.
    Those lessons were at the cost of the lives of rescuers who went 
into mines too quickly and died in rescue attempts. As recently as 
2001, 12 miners in Alabama attempted to rescue a miner gravely injured 
after an explosion. Those miners were killed by a secondary explosion. 
In 1976, 26 miners lost their lives in mine explosions in Overfork, 
Kentucky. Fifteen miners were killed in the first explosion and 11 
miners died in a second explosion, including 3 Federal coal mine 
inspectors.
    At 5:25 p.m. at the Sago Mine, the carbon monoxide and methane 
readings finally started trending downward to a level where MSHA, West 
Virginia's state mine regulators and the Robinson Run No. 95 rescue 
team felt we could take a carefully managed and calculated risk 
allowing rescue teams to enter the mine.
    Mr. Chairman, we at MSHA are determined to find out how this 
tragedy occurred. While I cannot comment on substantive aspects of the 
accident investigation, I can promise you that it will be thorough and 
meticulous. The investigation will examine the circumstances of the 
accident, the circumstances of the rescue efforts, and every other 
germane piece of information. All of the data and information collected 
will be carefully scrutinized and analyzed, and a thorough report 
prepared and made available to the public. We will finish this 
investigation as soon as possible, consistent with the need for 
accuracy, and we will promptly take any action indicated by the 
investigation's results to improve safety and health in America's 
mines.
    As standard operating procedure, MSHA also conducts an internal 
review after every major accident. We will carefully examine whether 
MSHA followed its own policies and procedures with respect to the 
accident, including the enforcement activities preceding the accident. 
This report will be provided to the committee and made public on our 
web site. We view this internal review process as an opportunity to 
take a hard and honest look at how we do our job and use that 
information to improve our performance on behalf of America's miners.
    Miner health and safety is our bottom line and our only priority. 
Thank you.

    Senator Specter. Thank you, Mr. Dye. Mr. McKinney.

                       STATEMENT OF RAY MCKINNEY

    Mr. McKinney. Thank you. This is a very brief presentation, 
and if you would draw your attention to the power point 
projector on the right. This is the service area of the Sago 
Mine that depicts the drift portals and the surface structures. 
To the right of those portals, you'll see an earthen embankment 
and that's where the mine offices were located when we set up 
the command centers. The Sago Mine was about 2.5 miles deep, 
and the mining height was 5 to 6 feet.
    This is an overview of the mine map. The mine has two 
working sections where miners travel each day to produce coal. 
Those working sections are the First Left and Second Left. 
There are extensive shelter areas in the mine. Those are areas 
that active workings are barricaded and built with proof seals 
to isolate them from active workers where miners are. Those 
sealed areas are identified in red in the lower part a little 
bit farther up, and the most recently sealed area was 3 Left, 
just in by 2 Left.
    The red line indicates the route of travel for the miners 
to the two active sections each day. On January 2, 2006, the 
mine was examined by company mine examiners and deemed safe to 
operate in the interim. Miners went--began to enter the mine at 
approximately 6 a.m. At approximately 6:30 a.m., an explosion 
occurred underground. All of the miners exited the mine except 
for the 12 miners that had traveled toward the 2 Left section. 
The mine examiner, who remained underground, was traveling from 
1 Left to 2 Left. So, we had 13 miners unaccounted for.
    An initial rescue attempt by mine management failed when 
they encountered heavy smoke and excessive concentrations of 
carbon monoxide in the 2 Left section. MSHA was notified at 
8:30 a.m. in the morning by telephone, and we responded, 
mobilized our force and were on the ground at 10:30 a.m. at the 
mine. When we got there, the air exiting the mine portals 
contained in excess of 500 parts per million of carbon monoxide 
and indications of methane gas. Carbon monoxide is an indicator 
of a fire or heat. Methane gas, of course, is inherent to 
mining explosions, and is explosive when it constitutes between 
5 to 15 percent of the air in a mine. The carbon monoxide was 
trending upward, which meant we had indications of a potential 
fire underground. By noon, the level had reached 2,600 parts 
per million.
    Also at noon, we had to remove people from the mine offices 
due to the high and elevated levels of carbon monoxide (CO). It 
was not safe to enter the mine until the CO, or the carbon 
monoxide level, showed signs of trending downward and 
stabilizing. We have a rich history of second explosions after 
an initial explosion. In 2001, we had the Jim Walters event 
where 12 miners were trying to rescue a fallen miner after a 
first explosion. All were fatally injured during the second 
explosion. In 1976, at the Scotia Mine, 15 miners were killed 
in an initial explosion. During the second explosion, another 
group was going in, and 11 miners lost their life.
    It's a very serious situation when you send people 
underground after an initial mine explosion. You have to know 
you're doing the right thing. Air ventilation is generally 
destroyed after the first explosion. Methane gas has an 
opportunity to build quickly, and you have indications of fire. 
The expansion of gases creates hot spots or fires in the coal 
mine. At 5:25 p.m., the levels had dropped to the point that 
the teams were sent underground. This was not an ideal 
situation. This was a calculated risk when we sent those teams 
underground, and we made contingency plans that if the CO 
continued to rise again, we would evacuate the mine.
    The teams explored the mine for the next 24 hours, 
encountering water accumulations, destroyed ventilation 
controls and electrical power sources and circuits. The teams 
had to methodically explore the mine keeping attention on the 
air entry return side to make sure what the CO concentrations 
were leveling out to, and also looking for potential fires. It 
was important also to look on the intake air entry side in case 
there was a fallen miner who had tried to exit the mine.
    At 4:21 p.m. on January 3, the teams had advanced to the 
58th crosscut. That is the area just out by the intersection of 
2 Left, the number 4 belt entry. At that point in time, we had 
not explored the 2 Left section. We'd found the seals blown out 
in the 3 Left section, which was the one that was most recently 
sealed. We had not explored that area but we decided, given the 
time frame, that we needed to extend ourselves on into 2 Left.
    The teams began to travel toward the 2 Left faces and found 
the section mantrap at number 10 block. We also found evidence 
that indicated the miners had donned their self rescue devices. 
We asked the teams to advance on into the faces of 2 Left, 
which is a distance of approximately 2,000 feet, looking for 
the miners. We rarely ever go more than 1,000 feet with the 
mine rescue team, we move our fresh air up to those people and 
allow them to move on. If we have a problem beyond that, it's 
very difficult to get the team back to the fresh air base, but 
the teams worked with us in order to advance themselves 2,000 
feet.
    The miners were located in number 3 entry face area, and 
the first report came at 11:46 p.m. that all 12 miners were 
alive. At 12:30 a.m., about 45 minutes later, a second report 
came out. It was changed that there was 1 miner alive and 11 
miners who were deceased. We continued to work underground, and 
we brought the surviving miner out at 1 a.m. and rendered 
medical attention to him, and he was transported to the 
hospital.
    We asked the teams to work under oxygen and continue to 
pick up the other miners and bring them to the surface. It was 
a long, arduous task through the night. At 9:55 a.m. the next 
morning, all the miners were extracted from the mine, and that 
ended the recovery operation.
    There's been some question about the communication issues 
underground. As the honorable Senator said, mine rescue team 
members are to be honored and their work and their bravery. 
These teams stretched themselves and pushed the envelope by 
going in excess of the normal distance in order to try to save 
life. If you look at this particular slide, you'll see that our 
communication systems, and these were handheld radios. Normally 
your sound-powered systems from the mine rescue only go 1,000 
feet, so we had to use handheld radios. They work well in the 
line of sight. They don't work well around 90-degree turns, so 
we had one set up near the face, one at number 9 crosscut, one 
at number 58 crosscut and then one at--I'm sorry, one at number 
59 crosscut and one at number 58 crosscut. This information has 
been transferred at least five times, and at least in the first 
four, it has been done through a mine rescue apparatus face 
piece with a speaking diaphragm. So I think that it's 
understandable that those people were doing everything they 
can, and you create situations for communication when you 
stretch out that far. Thank you.
    Senator Specter. Thank you, Mr. McKinney. We'll now start 
5-minute rounds for each of the senators on the panel, and I 
would like everybody to stay within the time limits. We can go 
to a second round if necessary, and I will stay within the 
limit myself.
    Mr. McKinney starting with you, there was a report that 11 
of the miners had been reported alive, 12 miners reported 
alive, and it was especially cruel and to then find out that it 
was wrong, that they had died. Was there any way that report, 
false report, could have been avoided to save so much pain and 
suffering?
    Mr. McKinney. I think it is sad, and I think it saddened 
all of us. There was a lot of people who felt very good for a 
short period of time and then were deeply saddened by that 
loss.
    Senator Specter. Could it have been prevented?
    Mr. McKinney. As I said earlier, those communications were 
coming from the underground. They were relayed several times 
before they reached the surface. We always ask repeat that, and 
you get a response back what you have, but by the time it 
reaches the surface, you have to go back through six people 
again, and I think that's exactly what happened.
    Senator Specter. Let me move on to another question because 
of limited time. The Sago Mine had 208 citations in the year 
2005. Was any of those citations a causative factor in this 
disaster? Any of the facts uncovered in the citations a 
causative factor for this disaster?
    Mr. McKinney. I think we're still investigating the 
disaster itself, and that will be determined by the 
investigation team if there's anything there we saw through the 
course of those inspections.
    Senator Specter. Is it a possibility that one or more of 
these violations might have been the cause of the disaster?
    Mr. McKinney. All of those 208 violations had been abated 
and corrected. The only outstanding violations that we had, had 
to do with roof conditions and tunnel liners being cushioned, 
and things of that nature. The other conditions had been 
corrected.
    Senator Specter. I noted that these 208 citations resulted 
in fines of slightly in excess of $27,000. Was that 
sufficiently tough punishment for so many citations, Mr. Dye?
    Mr. Dye. Nine of the most serious violations have been 
appealed. Those that haven't been assessed yet could double or 
triple the amount already assessed.
    Senator Specter. Was the $27,000 in fines sufficient for 
those which hadn't been subject to appeal?
    Mr. Dye. Those assessments are determined through a formula 
under our guidelines.
    Senator Specter. We're going to take a close look at that, 
and I would like to have your one-by-one itemization of all of 
these violations specifying exactly what was involved and what 
penalty was imposed so this subcommittee can make an 
evaluation.
    Mr. Dye. We will supply that for the record.
    [Furnished as questions for the record.]
    Senator Specter. Pardon me for moving on, but there's very 
little time, and we want to pinpoint the questions and the 
answers. Your unit had its funding reduced by $2,800,000 by 
virtue of a 1 percent across-the-board cut. Had the allocation 
remained without the 1 percent cut that this subcommittee had 
put into effect, we would have met the budget request for the 
administration. But with the 1 percent cut, there was that 
reduction of $2,800,000, and I note that you've lost 183 
positions as a result of budget insufficiency. Was that cut 
causative of the reductions in personnel, and were the 
reductions in personnel a causative factor in the Sago Mine 
tragedy?
    Mr. Dye. Please allow me to answer the last question first. 
No, I don't think so. We have dealt with rescissions for 
several years in a row. Our strategy has been, where we have 
had to reduce personnel, to try to reduce them in 
administrative and support positions, to try to reorganize in a 
way that maximizes our ability to keep our enforcement 
personnel out there. In fact, the Congress directed us to do 
that, and we have tried to manage them that way.
    Senator Specter. We would like a specification of what you 
would have done with the additional $2,800,000 had you been 
allocated that, and we would also like a written followup as to 
what were these 183 positions which you had to cut and what the 
impact has been on those cuts in terms of what your ability is 
to improvise mine safety.
    In the 16 seconds left, I have a final question for you. I 
note that you have stated your intention to hold a hearing. Why 
the delay? These hearings are very important. That is why we 
responded to Senator Byrd, January 2 incident. We're still in 
January, and this requirement that the witnesses be interviewed 
in the presence of company coal officials, if true, why? Isn't 
there a likelihood that they would be able to be in a position 
to give more candid testimony if they weren't interviewed in 
the presence of the company officials?
    Mr. Dye. My understanding is under the Mine Act, the 
company can be present; miner representatives can be present 
also. It is important to get interviews if you can before you 
have a public hearing. You do get more candid interviews that 
way.
    Senator Specter. Senator Byrd.
    Senator Byrd. Thank you again, Mr. Chairman, and thank you, 
Senator Harkin. Why did it take 2 hours for MSHA to be notified 
by the Sago explosion?
    Mr. Dye. We don't have that answer yet. That's one of the 
things--you mean by the company, sir? That is one of the things 
we're going to be investigating.
    Senator Byrd. How long do you think it's going to take to 
find out the answer to that question?
    Mr. Dye. The investigation team is in the process of 
interviewing witnesses. That is a rather thorough process. That 
should be accomplished in a matter of a few weeks, and then 
there will be a hearing. As soon as our investigation teams can 
enter the mine, they'll need to go into the mine and do all 
their forensic work. Then after that, they'll have to write a 
report. It's a very careful, thorough process, and we will get 
that report out and make it public to this committee, to the 
public generally just as soon as possible, Senator. I promise 
you that.
    Senator Byrd. Well, Mr. Secretary, I think you should know 
the answer now. That would be the first question, why did it 
take 2 hours for MSHA to be notified, 2 hours. Then why did it 
take 6 hours for rescue teams to arrive? Why is a rapid 
notification and response system not available? Should that 
mine have been closed with all of those citations? All of the 
breaking and wearing down, and tearing down and easing the 
regulations, why?
    Mr. Dye. Please allow me to answer the last question first, 
Senator. The Mine Act has substantial penalties and more 
importantly, and people don't always recognize this, it has the 
authority to close areas of the mine that are affected by 
violations. We did that 18 times until each of those violations 
were abated. As Ray mentioned, there were a couple of 
violations that were outstanding at the time, that were in the 
process of being abated, but The Mine Act does not contain a 
provision that allows preemptive closure of a mine for 
accumulated bad acts or a permanent closure.
    Senator Byrd. Let me cut right to the chase. Do you think 
that there's too much cronyism in situations of this kind? Do 
you think that the regulations and penalties are too soft? 
Should there be--I wonder if the enforcement procedures are 
enforced without fear and without favor. Is there a cronyism? 
Is there a possibility that cronyism develops between the 
industry and the Federal agencies that has jurisdiction over 
the safety of these people?
    Mr. Dye. Well, there certainly isn't any cronyism between 
me and anyone. I'm not aware----
    Senator Byrd. Will you say that again?
    Mr. Dye. There's no cronyism between me and anyone in the 
industry. With respect to the agency itself, particularly our 
coal mine division, I would note that our inspectorate have to 
have 5 years' experience in the industry. They're all miners, 
sir. They come from mining families. They come from mining 
communities. They have a fire in their belly to protect miners, 
and they go out and do that. I encourage them to do that fully.
    Senator Byrd. I have no quarrel with that.
    Mr. Dye. Yes sir.
    Senator Byrd. We know those people are brave and 
courageous.
    Mr. Dye. Yes, they are.
    Senator Byrd. But what I'm asking is, is there too much 
opportunity here for cronyism? I wonder if the regulations are 
being enforced without fear or favor, and the fines aren't 
being levied. Are they heavy enough? They don't seem to be. Too 
often, industry just seems to pay the fines and go right ahead 
and keep doing the wrongs. What do we have to do to fix that? 
That seems, in part, to have been the case here.
    Mr. Dye. The administration has sent to the Senate and will 
to the House, when it reconvenes, a proposel to raise the 
penalties for flagrant violations to $220,000.
    Senator Byrd. Is that just a tap on the wrist?
    Mr. Dye. No, I think that's substantial. The other thing I 
would like to mention is with respect to withdrawal orders and 
closing down sections of the mine. That can be a very, very 
powerful tool. In my opinion, even more powerful than citations 
because if you close down a production area until the hazard is 
abated, that can cost a company anywhere from $50,000 to 
$150,000 in lost revenues for a single shift. So we did that 18 
times, and that is a way that you can get the attention of an 
operator very quickly.
    Senator Byrd. Mr. Chairman, my time is about up, maybe 
overdone. Thank you.
    Senator Specter. Thank you very much, Senator Byrd. Senator 
Harkin.
    Senator Harkin. Thank you, Mr. Chairman. Mr. Dye, in 
December 2002, you were the deputy at that time, I believe.
    Mr. Dye. No, I wasn't.
    Senator Harkin. You weren't then?
    Mr. Dye. No, I was a Deputy Assistant Secretary in the 
Employment and Training Administration at that time.
    Senator Harkin. When did you come to MSHA then?
    Mr. Dye. In May 2004.
    Senator Harkin. In December 2002, MSHA withdrew a proposed 
regulation meant to encourage mine companies to host mine 
safety teams on site. It was a proposed regulation, not 2 hours 
away, but on site. It was withdrawn, I think, by Mr. Lauriski, 
David Lauriski, with the statement that MSHA ``plans to 
evaluate non-regulatory alternatives''. That was 3 years ago. 
What did you come up with?
    Mr. Dye. I would defer to Mr. Clair if you don't mind 
because he was there at that time. My understanding of that 
proposal centered around the idea that you would rebate some 
fines and give that money to the companies to encourage them to 
have mine rescue teams. I think that after some thought, pretty 
much everybody thought that that wasn't a good idea because 
that would undercut the value of the fines for enforcement 
purposes.
    Mr. Clair. Senator, I would only add to what Mr. Dye has 
said that that proposal was something that the agency had under 
consideration. I do not believe it had been formally proposed. 
It was a work in progress, and the theory behind it had 
substantial problems to it. It looked to economic incentives to 
mine operators that essentially were funded out of civil 
penalties, and that is a serious legal problem. There were also 
difficulties when you streamline an existing standard like mine 
rescue standard that is quite complex and comprehensive. If you 
streamline that, there's a potential for reducing protections 
that already exist.
    Senator Harkin. There was also an advance notice of 
proposed rule making in 1997 that was issued to improve the 
oxygen devices worn by miners like the ones at Sago. The self-
contained rescue devices, as we know, provide an hour of 
oxygen, they were required in 1981, 12 years after they were 
invented, that every miner must be equipped with one. This 
regulation was dropped from the agenda in December 2001. The 
regulation would have addressed the manufactured quality of the 
devices, the training of miners on its use. There's one other 
part of that regulation, that my staff showed me that was very 
interesting.
    The regulation, the proposed regulation, would have 
addressed the need for updated technology including personal 
emergency devices, underground text messages devices. Why was 
that dropped in 2001?
    Mr. Dye. Well, getting back to 1997 is really getting 
backed----
    Senator Harkin. It was dropped in 2001.
    Mr. Dye. Yes.
    Senator Harkin. December 2001.
    Mr. Dye. Yes.
    Senator Harkin. Why was it dropped? Do you even know? If 
you don't know----
    Mr. Dye. Well, with respect to the self-contained rescuers, 
there was a question about the service life of those devices, 
and I'm told that it simply was not evidence that the 
manufacturer service life should be reduced.
    Senator Harkin. Say that again.
    Mr. Dye. It concerned the recommended service life of those 
devices. In other words, how long you should keep them or could 
keep them for--before you use them.
    Senator Harkin. The reason I brought up--brought it up, the 
other part of it, in 1998, there was a mine fire at the Willow 
Creek Mine in Helper, Utah. Personal emergency devices--which 
is in that proposed regulations, instantly and wirelessly 
received emergency message broadcast to each miner on the 
screen. Forty-five miners used that to escape the fire. Seven 
years later, only 17 mines in the United States utilized this 
technology. Why?
    Mr. Dye. Well, I talked to my tech support folks about 
that, and there are, as you said, some mines that use it. Some 
of them, candidly, are enthusiastic about it. Some of them have 
had a number of problems, including reliability issues. That 
is, I believe, a one-way communication device. We're looking to 
see if that can be made into a two-way communications device, 
which would be much more useful.
    Senator Harkin. Well I just wanted to read here, here's 12/
01. MSHA's withdrawing this entry from the agenda in light of 
resource constraints, and changing safety and health regulatory 
priorities. Resource constraints, you know what that means, we 
know what that means.
    Senator Byrd. That's money.
    Senator Harkin. I don't know what changing health and 
safety regulatory priorities mean, priorities. Well it seems to 
me again I harp on this, because there is technology out there 
that can be used. My time has run out too, but I want to get 
back to that Mr. McKinney, you talked about not getting around 
the corner. There's easy simple technology to take care of 
that, very simple easy technology. I realize what they had 
couldn't get around the corners, it's line of sight. But, that 
can be changed in a matter of minutes with technology.
    Thank you, Mr. Chairman.
    Senator Specter. Senator Harkin, did you say you wanted to 
get back to that, if you want a few more minutes?
    Senator Harkin. No, that's okay.
    Senator Specter. Senator DeWine.
    Senator DeWine. Mr. Dye, let me kind of follow-up on what 
Senator Harkin was asking you, and what some of the other of my 
colleagues have been asking you, I'm not sure I quite 
understand your procedure that you're going to follow in order 
to conduct the investigation, and then come forward with 
recommendations. Where are you going to go from here? What is 
the time line, how long is the investigation do you think it's 
going to take? So, you can tell us what happened, and then when 
are you going to come forward with recommendations, and what's 
the process that you're going through in regard to how you're 
going to come forward with those recommendations, because 
that's frankly what we're interested in.
    What we're interested in, is where do we go from here? That 
is what Senator Harkin was getting at. He's talking about the 
technology that could be utilized. We've all read different 
experts talking about the technology that could be used. We're 
kind of interested in what you have to say about it.
    Mr. Dye. Well with respect to the investigation, first of 
all----
    Senator DeWine. Well yes, and then where are you going to 
go from there? Because that is the bottom line, is making it 
safer for coal miners. Obviously, that's what we're all 
interested in. That's what you're interested in.
    Mr. Dye. Absolutely sir.
    Senator DeWine. That's what everybody's interested in.
    Mr. Dye. Yes sir. We appoint an investigative team in these 
kinds of matters that is independent. The Lead Investigator 
designs the investigation; they have to be custom designed 
because there's a lot of issues. One of the first is, they have 
to get in and look at the accident scene. They couldn't do that 
for a long period of time, because they had to clear out a lot 
of water and gas out of the mine. We still have mine rescue 
teams.
    Senator DeWine. I only have a couple of minutes. What's the 
timing on that?
    Mr. Dye. At the same time, the investigative team has to 
interview witnesses. However many they have to do, and they can 
do about three a day, because they're fairly extensive, and 
when they've interviewed all of those, and they have that 
information, they will hold a public hearing, in which they're 
in the process of planning now. Then they will write a report, 
and that will all be done as rapidly as possible. But because 
we don't know the conditions in the mine, and how much time 
they're going to have to spend in there, it's hard for me to 
give you hard time line, Senator.
    Senator DeWine. Then what's the process for your 
recommendations, the administrations recommendations about how 
we move forward?
    Mr. Dye. This is an accident investigation, so the 
recommendations will be specific to that situation. By the way, 
we also do an internal investigation to see if we followed our 
own procedures. Then from that separate inquiry, if we didn't 
follow our own procedures, or there's ways to optimize them, 
the internal review team will also write a report. We will 
adopt those recommendations fairly quickly.
    In the case of the Jim Walters accident in 2001, the Agency 
adopted an emergency regulation very quickly. I want to add 
with respect generally to mine rescue, we are in the process of 
publishing, and have put up on our website today, a request for 
information in the Federal Register.
    What we're seeking is information from individuals all over 
the country, or all over the world for that matter, on mine 
rescue technologies. These are emerging technologies, new 
things that are popping up all the time. So, we would like to 
take a snapshot of what's available now, look at it, evaluate 
it, and see what is out there.
    Senator DeWine. What is you time line on that? That's good.
    Mr. Dye. I believe that is a 60 day comment period going to 
March, and then our technical folks will have to evaluate all 
of the comments, and that depends on the volume, and other 
things.
    One thing about these technologies, you can't use off the 
shelf technologies. Any electrically powered device, including 
a battery device, has the potential to cause an explosion in a 
gassy mine. So they'll have to be virtually redesigned, so they 
won't do that. It is an expensive process, and manufactures 
have to be willing to do that. I couldn't take my cell phone, 
or a camera, or an MP3 player, or anything like that into a 
gassy mine. That would not be permissible, because those things 
have not been designed so that they would not create a spark.
    Senator DeWine. Mr. Dye, my time is about up. I appreciate 
your comments. I think what you're hearing from this panel, at 
least you're hearing from this Senator, is that my constituents 
expect out of this tragedy something positive to come. And what 
the positive is, is something specific, and specifics, and the 
specifics lead to better mine safety, and that's what we look 
to you to come forward with.
    To get back to what Senator Specter was talking about, if 
we're talking about additional money, we expect you to come 
forward and tell us that. We want to know how that lines up 
with the dollars and cents.
    Mr. Dye. Yes, I can tell you that there will be additional 
funding for safety technology--mine safety technology in the 
President's 2007 budget.
    Senator DeWine. I thank you, my time is up. Thank you.
    Senator Specter. Thank you, Senator DeWine. Thank you Mr. 
Dye, Mr. McKinney, Mr. Friend, and Mr. Clair. Mr. Dye and Mr. 
McKinney, I'm advised that you would like to leave. Your 
presence will be required here for at least 1 more hour while 
we move ahead with the next panel. Questions may arise, and we 
want you here to answer the questions; we've assembled four 
Senators who may need your responses.
    Mr. Dye. Senator, we've still got a mine fire going. We 
have a rescue team that's in the Sago Mine. We have another 
mine fire, in which no one was hurt, burning in Colorado. We 
have really urgent matters that we need to go back to and 
attend. We have been diverted by dealing with these matters. We 
were happy to prepare for the hearing, but we really need to 
get back, and attend to all of this. There are 15,000 mines in 
the United States, and we've got some really pressing matters.
    Senator Specter. Well Mr. Dye, I can understand the press 
of your other business. It may well be that some of the 
Senators here have other pressing matters too, it may well be. 
We're looking at the Alito hearings tomorrow in the Judiciary 
Committee, a matter of enormous importance. Senator DeWine and 
I have to be there. We're looking at the proceedings under the 
PATRIOT Act, which expire on February 3, which we have to come 
to a conclusion on. We're preparing for hearings on the 
National Security Administration, NSA, about potential excesses 
in power on the part of the President. We're in the final 
stages of preparing immigration legislation, which is an 11 
million person problem for this country. We're about to go to 
the floor for the Asbestos Reform Act, and I think Senator 
Byrd, and Senator Harkin, and Senator DeWine have other 
matters. So, we don't think we're imposing too much to keep you 
here for another hour.
    If you told me your hearings were going to start tomorrow, 
or there was some pressing business, we'd say you don't have to 
be here, but that's the committee's request. You're not under 
subpoena.
    We'll now move on to the next panel. Mr. Cecil Roberts, Mr. 
Ben Hatfield, Mr. Chris Hamilton, Mr. Davitt McAteer, and Mr. 
Bruce Watzman. Our first witness on this panel is Mr. Cecil 
Roberts, the International President of the United Mine Workers 
of America. Mr. Roberts is a sixth generation coal miner. Both 
of his grandfathers were killed in mines before he was born. 
He's a graduate of West Virginia Technical College, where he 
also received an Honorary Doctorate in Humanities.
    Mr. Roberts, you come to this hearing as a distinguished 
President of the United Mine Workers of America, and going back 
six generations is very impressive.
    Mr. Roberts. Thank you sir.
    Senator Specter. I have a little trouble going back two 
generations. I only knew one of my four grandparents. So, you 
have quite a lot of longevity. As announced earlier, the 
allocation of time is 5 minutes per witness, and we'll begin 
with you Mr. Roberts.

STATEMENT OF CECIL E. ROBERTS, INTERNATIONAL PRESIDENT, 
            UNITED MINE WORKERS OF AMERICA, FAIRFAX, 
            VIRGINIA
    Mr. Roberts. Well first of all Senator, thank you for 
agreeing to have this hearing, and your long years of service, 
and dedication to coal miners in this country, and there are 
two other very distinguished Senators I've had the opportunity 
to work with, and of course the person who lead the fight for 
coal mine health and safety legislation in 1969. My senior 
Senator from West Virginia, I've worked with Senator Harkin, 
and I've had the pleasure of working with Senator DeWine, too.
    First, I think it would be appropriate, and I will say it 
for all of us, that our hearts go out to the families, not only 
of Sago Mine, but most recently the terrible tragedy that 
occurred, and ended over the weekend in Logan County, West 
Virginia.
    In my prepared remarks, we never mentioned the situation in 
Logan, because we did not know at the time that you had 
requested us to get this information to you, the unfortunate 
outcome of that tragedy. Let me say that I'm probably going to 
take a different approach to this as I think that you might 
expect to represent the coal miners in this country as to what 
I think needs to be done, and what should've been done, and 
what is the possibilities for the Congress to do here.
    First of all, everyone is now for oxygen being available 
throughout the coal mine. Well, we've known for many, many 
years that this was a necessity. I'm not talking about the 
United Mine Workers when I say we. I would suggest to you that 
any expert in this industry, if you had asked them yesterday, 1 
year ago, 2 years ago, 6 years ago, they would've said to you, 
if you want to give coal miners the best opportunity for 
survival, then you provide oxygen in the working places of that 
coal mine, and you provide it throughout the coal mine.
    Now the question is not only should we do this, and require 
it, and should Congress act, and give a mandate of some type 
that that be done. I think it is appropriate for the question 
to be asked, why is it as we come here today, that that has 
never been done? The question was asked previously here, could 
these tragedies be prevented? That's one question. There is 
another question that should be asked and posed here, what is 
the best opportunity for a coal miner to survive, if and when 
tragically this occurs again? It's a relatively simple question 
when it comes to oxygen. If I told everyone we're going to cut 
off the oxygen to this room right here, and it's going to 
disappear, there will be no more oxygen, but we're going to 
take a vote here today, if we want 1 hour of oxygen available 
to us, or 2 hours, I know what the vote would be.
    Second of all, when I started working in the coal mines in 
1971, we had a telephone line running to the outside. This was 
30 some years ago, to the face. Today if you go in a coal mine, 
that is what you find in most instances, one telephone line 
from the surface to the working sections of the mine, now there 
is some things that could be done, and we've been contacted by 
the way, by the Pentagon, they don't know if they have the 
expertise, or the technology to solve this problem for us. But, 
they took the initiative to phone United Mine Workers, and 
suggest that we should be talking to them, and we would urge 
the Senate, this committee, to do that.
    There's other things we could do. We could put that 
portable hand held device that we were talking about the Mine 
Rescue Team's using. It is only good for 1,000 feet, but for a 
very minimal cost, I should report to this Committee we could 
do that. If we could communicate with miners on the wrong side 
of these disasters, they would know what their best opportunity 
for survival might happen to be. There's two things we do know, 
and I don't believe there will be any disagreement amongst 
anyone on this panel. Had the miners at Sago had more oxygen, 
and had they had the ability to communicate with the outside, 
their lives could have possibly been saved. But we do know one 
thing, it would certainly have given them the best opportunity.
    This issue of mine rescue teams, I must admit to this panel 
today, and I don't want to appear to be too frustrated, and too 
angry here, but I am. We have known for years that we need 
additional mine rescue teams in this country, and the problem 
isn't getting better, it's getting worse. Because, as people my 
age retire from this industry, who have life long experience as 
mine rescue team members, then I submit to you, we're losing 
those experienced mine rescue team members. We've known this 
for many years, but we have not corrected this problem.

                           PREPARED STATEMENT

    Let me submit one final comment, and I realize that this 
may be getting into an area that may offend some people, but in 
1969 Mr. Chairman, Congress decided this industry was incapable 
of policing itself, and established MSHA as an independent body 
to protect the coal miners of the United States of America. In 
2001, we put the coal industry in charge of this Agency, and we 
have submitted 17 rules that were withdrawn, in 2001 or later, 
that would have protected coal miners in the United States of 
America, and there are other issues that we could raise with 
you today, but I realize my time is up, but I did not want to 
leave here with that on my heart, and on my mind. Thank you.
    [The statement follows:]

                 Prepared Statement of Cecil E. Roberts

    I thank you for this opportunity to appear before Congress. I only 
regret that I have come to speak on the heels of the terrible tragedy 
that befell the Sago miners. Our hearts and our prayers go out to the 
miners' families, their loved ones and their communities. We also wish 
to extend our deep appreciation for the mine rescue teams that 
participated in the Sago Mine rescue efforts, and to the Federal and 
State inspectors and UMWA safety committee members who travel and 
inspect the mines regularly, working tirelessly to protect the nation's 
miners.
    Today I will be reviewing how current mine safety laws came into 
being; problems with the enforcement of the current laws and 
regulations and how those deficiencies have contributed to make coal 
mining one of the most dangerous industries in the nation; how we have 
the knowledge and ability to substantially improve miners' health and 
safety; and what Congress can do to help improve miners' health and 
safety.
    For 116 years, the UMWA has been unwavering in its efforts to 
enhance miners' health and safety.
    However, too often remedial activity follows only after another 
tragedy has focused our nation's attention: in 1968, 78 deaths at 
Farmington, West Virginia lead directly and quickly to passage of the 
Coal Act in 1969; it was then expanded to other mining industries and 
re-named the Mine Act in 1977. Since the Coal Act was passed, 
fatalities in coal mining have decreased dramatically: while over 300 
miners died in 1968, the year before the Coal Act was enacted, fewer 
than 100 miners have perished in any single year over the last 20 
years, since 1985. While increased mechanization has meant fewer miners 
are engaged in coal mining, the fatality rate has also dropped 
significantly. This is commendable; but we can and must do much better. 
Mining remains the second-most dangerous industry in this country.
    This nation possesses the knowledge and ability to substantially 
improve miners' health and safety, and to reduce the fatality rate. We 
can direct some of the national attention generated by the Sago tragedy 
to enhance health and safety conditions for all coal miners today, and 
the generations to follow.
    Unfortunately, what happened at Sago did not really surprise me. 
Indeed, the underground coal industry has experienced tragedies, as 
well as near tragedies, on a recurring basis. In just the last few 
years the underground coal industry experienced these large-scale, 
well-publicized, events:
    Jim Walters Resources #5 mine explosions; 13 fatalities.--There was 
a terrible series of events on the evening of September 23, 2001, two 
short weeks after 9/11. What happened there eerily echoed the Twin 
Towers' experience inasmuch as numerous rescuers also perished at this 
Alabama mine. In fact, 12 miners lost their lives in a second explosion 
while trying to rescue a miner who had been immobilized by an explosion 
that had happened nearly an hour earlier. Communication problems 
contributed to the deaths of the 12 rescuers; we believe the rescuers 
were given insufficient and faulty information about the underground 
conditions, and attempted the rescue without knowing the hazards they 
faced.
    Quecreek.--In July 2002, 9 miners were trapped by a water 
inundation in a Pennsylvania mine, after 9 others were able to escape. 
The trapped miners were rescued 4 days later; again, communication 
inadequacies frustrated an easier and quicker rescue.
    Sago Mine; 12 fatalities.--On January 2, 2006, this tragedy claimed 
the lives of 12 miners, while the full extent of injuries to the 13th 
miner, Randal McCloy, Jr., remain uncertain. Hopefully, with time we 
will learn all of the conditions that caused and contributed to these 
lost lives.
    These dramatic events represent only the headline-grabbing 
incidents. Thousands of miners are still disabled and dying from black 
lung disease, while other miners also die in mining accidents each 
year. Typically they die one or two at a time, from roof falls, 
equipment failures, and other accidents.
    There are also countless near-misses that occur on a regular basis. 
In fact, just since August 2000, the Mine Safety and Health 
Administration, known as MSHA, has records of well over 400 mine fires, 
ignitions, explosions and inundations that far too-easily could have 
developed into significant disasters and fatalities; many other 
incidents likely went unreported.
    With better regulations, more regular enforcement, and with support 
from the highest echelons of the Agency, many of these accidents could 
have been prevented. Senseless deaths and injuries must stop. Mining 
will probably always be a dangerous job. But we can do a lot more than 
we are doing today to make it safer. Miners should not have to get 
sick, or to risk their lives just by going to work.
    The most basic point I wish to make today is that as a nation, and 
as an industry, we already possess the knowledge and the ability to 
prevent most of the deaths that are still occurring in the coal mining 
industry. What is needed is a real commitment by our government--in 
this case, MSHA--to do better.
    For example, if MSHA would require additional oxygen units (``self- 
contained self-rescuers'') to be stored on the section where miners 
work, and throughout the underground mine in addition to the self-
rescuers that each miner is required to have, then miners trapped in an 
emergency would have a better chance of surviving. Self-contained self-
rescuers are what miners don to escape, when noxious air fills an 
underground mine after a fire or explosion. They typically last for 
about one hour. We understand that miners trapped at Sago wore self-
rescuers. However, there is now no MSHA requirement that coal operators 
store any additional oxygen units in their underground mines. We can 
only speculate about whether more miners would have survived at Sago if 
additional self-rescuers had been stored underground. What we do know, 
is that they would have had a better chance of surviving until help 
arrived if they had more fresh air to breathe. MSHA should implement a 
rule requiring additional units to be maintained in strategic locations 
around the mine in order to provide miners with oxygen from the deepest 
penetration of the mine out to the surface. No new technology is 
required to implement this improvement. Self-rescuers can save lives. 
Are they worth some modest additional costs? We submit the clear answer 
is ``Yes.''
    There is also technology available today that would enable trapped 
miners to maintain better communications in an emergency situation. If 
MSHA would require secondary telephone lines to be placed in a separate 
entry, that would increase the likelihood that communications could be 
maintained between miners and those on the surface, even after an 
explosion or other emergency event. Also, if MSHA would require 
operators to place walkie-talkies in underground locations, this would 
facilitate communications during an emergency. Rescue teams rely on 
walkie-talkies while they travel underground. The equipment is 
effective for a distance of about 1,000 feet. If trapped miners could 
communicate with their would-be rescuers, the trapped miners could both 
provide and receive critical information that could assist in their 
survival. By not having this ability to communicate, their own rescue 
is hampered and rescue teams may confront additional hazards that could 
be avoided if trapped miners would report on what they know. These 
proposed communications' improvements could be implemented easily; the 
UMWA urges MSHA to quickly implement these changes. If we can talk to 
people on the moon, we should be able to talk to those trapped 
underground.
    After the Jim Walters tragedy, and again after the Quecreek near-
disaster, the need for better underground communications was crystal 
clear. Yet, MSHA made no changes to implement any such improvements.
    If we could locate trapped underground miners, we could also do a 
much better job rescuing them. MSHA should draw on all the nation's 
resources to address this compelling need. We are encouraged by 
communications we have had with people at the Pentagon about equipment 
originally developed for aviation security, which might also be 
applicable to the mining industry. We do know that since the 1970's, 
there has been electromagnetic technology available that would enable 
us to locate trapped underground miners. The then-Bureau of Mines 
reported about this equipment in the 1970's; the equipment was tested 
and deemed reliable some 30 years ago. Why then aren't we using it 
today? MSHA should take a look at this electromagnetic technology and 
all other available technology to see which should be required 
throughout this industry.
    Another way to enhance the chances of survival for trapped miners 
is to strengthen and expand mine rescue teams. The Mine Act required 
MSHA to implement regulations ensuring that ``mine rescue teams shall 
be available for rescue and recovery work to each underground coal or 
other mine in the event of an emergency.'' 30 U.S.C. 825(s). MSHA's 
regulation now provides that every operator must establish at least two 
mine rescue teams, consisting of five members and one alternate, and 
that these teams be available at all times miners are underground. The 
regulations also permit an operator ``to enter into an arrangement for 
mine rescue teams'' whereby at least two mine rescue teams are 
available whenever miners are underground. Further, to be deemed 
``available'' the mine rescue teams must be capable of presenting 
themselves ``at the mine site(s) within a reasonable time after 
notification of an occurrence which might require their services.'' 30 
C.F.R. Part 49. Pursuant to MSHA policy, if a team can arrive in two 
hours, it is deemed ``available.'' MSHA includes exceptions for small 
and remote mines, but those exceptions would not have applied to the 
Sago Mine.
    At the Sago Mine on January 2, it was between three and five hours 
before the first mine rescue teams arrived. Once mine rescue teams 
arrived on property, including six teams comprised of UMWA members, the 
sole activity they engaged in revolved around safely extracting the 
trapped miners. It is extremely important for you to understand that 
procedures followed during mine rescues are designed to save the lives 
of trapped miners while also protecting the team members. Rushing into 
the mine without a complete understanding of its conditions, would be 
inviting additional problems. You need only review mining history 
briefly to see examples of rescuers rushing in only to become 
additional victims. We cannot permit that to occur. Even if underground 
conditions may not have allowed the rescuers to immediately travel 
underground at Sago, the five hours' lapse before the first teams 
arrived constituted valuable time that was simply forfeited. We do not 
know what arrangements the Sago Mine had to address its mine rescue 
team obligations, but we do know that a five hour lag before mine 
rescue teams arrive is unacceptable.
    The UMWA submits that every underground coal mine should have mine 
rescue capabilities on site. These team members should be employees at 
the facility who would be acutely familiar with the mine. These 
individuals would not only be best able to carry out many of the duties 
required in these situations, but would also be uniquely qualified to 
brief additional offsite teams that may be necessary to complete the 
rescue. For even small and remote mines, MSHA should require mine 
rescue teams to be ready when disasters strike. No trapped miners 
should ever again have to wait three to five hours for rescue efforts 
to begin.
    While we all share the hope that all mining accidents will cease, 
all of our hopes probably won't prevent more accidents from happening. 
What we can do, and what we should do, is to give miners their best 
chance at surviving a mine emergency. I have just described a number of 
improvements that would be easy to implement; they would also go a long 
way to furthering miners' ability to survive disasters. However, unless 
MSHA requires that these improvements be implemented throughout the 
industry, miners will continue to face emergencies from a position of 
compromise.
    If the Sago miners had been able to communicate with persons on the 
surface and the mine rescue teams underground, those miners might have 
been successfully lead to fresh air. If the location of those trapped 
miners could have been more-quickly determined, we could have enhanced 
the possibility of their rescue. These ``only-ifs'' are too late to 
change what happened at Sago, but it would be inexcusable if our 
inaction now would contribute to any more deaths.
    There was a national outcry when 78 miners died in West Virginia in 
1968. Unfortunately, after the Jim Walters' disaster and the 
recommendations made from the lessons learned there, no lasting 
improvements to miners' health and safety came after those 13 miners 
perished. We cannot allow that to happen again. We need to make real 
changes so that another tragedy like what happened at Jim Walters and 
what happened at Sago will not happen again.
    MSHA knows how to do better. The Agency itself has performed 
countless internal reviews and self-analyses; the Federal Government's 
watchdog agency, the GAO, has given it direction, and the UMWA has 
communicated both formally and informally about how MSHA can and must 
do better.
    GAO recently focused on shortcomings in MSHA's performance with 
regard to the underground coal industry. The underground coal industry 
is the same part of MSHA's jurisdiction that was at issue in all three 
incidents I highlight in this testimony: Jim Walters, Quecreek, and 
Sago.
    GAO issued its report in September 2003, two years after the Jim 
Walters tragedy. In its report, which I attach to my testimony, GAO 
noted that MSHA headquarters was not performing adequately in several 
key areas. Specifically, the GAO found MSHA failed to ensure violations 
cited to mine operators were corrected in a timely fashion. In fact, 
GAO found that of all the citations issued by the Agency, including 
those written as ``significant and substantial,'' despite inspector-
imposed deadlines by which problems were to be abated, 48 percent of 
the time the Agency failed to follow-up in a timely fashion to see if 
the operator fixed the hazards.
    GAO also found that MSHA collected information about accidents and 
investigations, but then failed to use the information effectively to 
prevent future accidents. GAO noted that MSHA does not even collect 
information about how many hours contractors work at mine sites, making 
it impossible to compute the fatality and injury rates for particular 
mines. It further found that MSHA failed to ensure that the ventilation 
and roof control plans are reviewed every six months, even though the 
Mine Act and applicable regulations, as well as MSHA's long- standing 
policies, require that these reviews occur on a semi-annual basis.
    After MSHA completed its investigation into the Jim Walters 
disaster, the Agency also performed an Internal Review of MSHA's 
actions before the explosions to ``improve our inspection process to 
better protect our nation's miners.'' The review compared what MSHA 
actually did with what the Mine Act requires it to do. A number of 
problems were identified as deficiencies ``at both the district and 
headquarters level'', deficiencies ``relevant to inspection procedures, 
level of enforcement, plan reviews, the [Alternative Case Resolution 
Initiative] and accountability programs, supervision and management, 
and headquarters oversight.''
    Unfortunately the Agency's top managers have done little to move 
any of the necessary improvements from recommendation to reality. We 
hope that by having Congress add its voice now, along with the public's 
demand for its better performance on the heels of Sago, MSHA will 
finally re-focus its attention.
    MSHA should promulgate regulations that meaningfully improve 
miners' health and safety, and MSHA must consistently enforce the 
regulations it has. The circumstances that played out on January 2, 
2006 have become all too familiar. The events that unfold at these mine 
disasters change very little, they focus attention for a brief moment 
but, as the spotlight fades, MSHA is content to ignore demands for 
change. We cannot and will not permit that to happen here.
    The Mine Act is an important piece of legislation. I am not here to 
advocate an opening of any part of that historic and effective law. The 
UMWA is here to simply demand that it be enforced by MSHA as Congress 
intended. The real problems that exists with the current application of 
the Act lie in an administrative malaise. Instead of enforcing it 
aggressively, the Department of Labor has been whittling away at the 
Mine Act. Too often MSHA relies on ``policies,'' which are developed 
internally and without public comment, to circumvent the Mine Act. This 
reduction in MSHA's effectiveness didn't happen overnight; it has been 
a problem for much too long. We have been critical of MSHA under both 
Democratic and Republican administrations. But now we call upon 
Congress to put a stop to MSHA's inadequacies, to turn the Agency 
around so it can enforce the Act the way Congress intended in 1969, and 
again in 1977.
    To make the needed changes will require MSHA to take a new 
direction, beginning at the top. MSHA needs to have a larger budget for 
coal enforcement. The Agency spends too much effort at ``compliance 
assistance,'' and too little on enforcement. It needs to bolster its 
expertise. MSHA has some excellent mine health and safety experts 
working in field offices throughout the Country. Yet, they have not 
been receiving support from those above them. Too often, an inspector 
will write citations and orders upon finding violations of regulations, 
only to have them compromised away through conferences and settlements. 
While MSHA employs many experienced and dedicated inspectors, there is 
a failure on the part of individuals sitting at the higher levels of 
power to support the efforts of the MSHA field staff to enforce the 
Mine Act.
    MSHA's failure to aggressively enforce the Mine Act is the result 
of many factors. MSHA is full of former mine management executives who 
spend too much time trying to appease their friends, and too little 
time looking out for miners' interests. How can we expect a regulatory 
Agency to effectively and fairly carry out its duties and 
responsibilities when it is run by some of the very coal operators it 
was designed to regulate? Many of MSHA's top-level administrators spent 
years opposing any regulatory efforts attempted by the Agency. They 
continue to be influenced by other coal operators, effectively muting 
the voices of miners who need a healthy and vibrant enforcement Agency.
    For years, the fox has been inside the henhouse at MSHA. During his 
tenure as Assistant Secretary for MSHA, David Lauriski, scrapped 18 
proposed rules on topics MSHA had already identified as needing 
attention. I attach to this testimony a list of those proposed rules 
that former-Administrator Lauriski withdrew, and will discuss the 
importance of a few of them. However, withdrawing proposed rules was 
not the extent of his actions at the Agency. He also pushed to 
completion several regulations that weakened health and safety, because 
they benefitted mine operators. For instance, under his watch MSHA 
implemented a regulation that allows mine operators to ventilate areas 
of the mine where miners work with air that has already passed through 
the conveyor belt entry, a practice Congress specifically prohibited in 
the Mine Act (30 U.S.C. Sec. 863 (y)); and under his leadership MSHA 
implemented a regulation that permits the use of diesel-powered 
generators in confined areas of underground mines. While these 
practices may increase production for mine operators, they also pose 
new and significant risks to miners.
    Unless something changes soon, history will repeat itself at the 
Agency. Later this month, on January 31, 2006, a hearing is scheduled 
to consider the Administration's nominee for Assistant Secretary of 
Labor for MSHA. This second nominee by this Administration has been a 
coal mine supervisor for most of his life. His lone excursion into 
health and safety was marked by repeated attempts to limit regulations 
and reduce the health and safety protections of miners in the 
Commonwealth of Pennsylvania. MSHA--and our nations' miners--cannot 
afford having another coal operator sitting as the Assistant Secretary. 
Miners deserve an advocate and an ally, not another coal boss. There 
should be no doubt that when top-level appointees are too cozy with the 
industry, miners ultimately pay the price.
    Among the regulations MSHA stopped developing, and withdrew from 
its agenda were those addressing coal dust exposure; self-rescuers; 
mine rescue teams; accident investigations; and training/re-training.
    The proposed rule on Occupational Exposure to Coal Mine Dust was 
withdrawn on September 4, 2002. That proposed rule was drafted to 
comply with the Secretary of Labor's Advisory Committee on the 
Elimination of Pneumoconiosis Among Coal Mine Workers, and was intended 
to decrease the level of respirable coal dust miners could be exposed 
to during a working shift. By cutting the permissible exposure level in 
half, miners would be less likely to contract debilitating black lung 
disease. Application of such a standard would also have significantly 
reduced the amounts of highly-explosive float coal dust released into 
the mine atmosphere. Such a regulation would have significant health 
and safety benefits for underground miners. Unfortunately, the only 
efforts regarding coal dust that MSHA made under former Assistant-
Secretary Lauriski was a proposal that would have allowed respirable 
dust levels to increase by four fold. To put it mildly, his proposal 
was not well-received and he ultimately withdrew it.
    MSHA's proposed rule on Underground Coal Mining; Self-Contained 
Self-Rescuer was withdrawn on September 24, 2001. Self-contained self-
rescuers have not been updated to keep up with technology. They were 
first required in 1981 and little has changed since then. Some of these 
devices were found to be inoperable for a variety of reasons including 
deteriorating hoses, contaminated chemical beds, and unrealistically 
long shelf lives being approved by MSHA. The industry has also been 
plagued with the fact that miners sometimes cannot properly don the 
units in emergency situations. Moreover, with MSHA's continued 
acceptance of the status quo, technological advances of these breathing 
devices is being stymied. In the legislative history of the Mine Act, 
Congress indicated that mining regulations should be technology-
driving, to maximize miners' protections. We had hoped that with the 
promulgation of a new rule addressing self-rescuers, the existing 
problems would be addressed, and technological advances encouraged. The 
UMWA is convinced that such a rule would have been the catalyst for a 
new generation of self-rescuer devices. While operators are willing to 
invest in new technology when it increases production, it appears that 
they are not so willing to invest when in miners' health and safety.
    The proposed rule on Mine Rescue Teams was withdrawn on September 
4, 2002. The basis for moving this rule forward is quite simple: the 
UMWA and many industry officials recognize that, with mining operations 
contracting in the late 1980's through the 1990's, the number of mine 
rescue teams was disproportionally reduced. This left large coverage 
gaps. The industry is also facing an overall aging of the workforce, 
and this also adversely impacts participation in those rescue teams 
that remain active. In May of 2000, when it published this pre-rule, 
MSHA stated: ``We are assessing the current regulations to identify 
problem areas where we might increase flexibility and increase safety 
for miners.'' However, instead of promulgating a rule that would 
improve rescue teams' availability and capabilities, MSHA eliminated 
further work on rescue teams regulations. Meanwhile, it permits 
operators to expand on the ill-advised practice of contracting out such 
work. Withdrawing the proposed rule effectively eliminated any 
meaningful improvement in comprehensive mine rescue activity, but it 
also afforded some mine operators the opportunity to disband teams so 
they could increase their profits.
    On August 16, 2001, the Agency withdrew its proposed rule on 
Accident Investigation Hearing Procedures. MSHA has no formal rules for 
conducting investigations. While it has established policies, the 
investigation process is subject to change on the whim of the 
individuals running the Agency. This is exactly what developed in the 
early stages of the investigation of the Sago Mine disaster. Not only 
do questions arise about who should participate in various parts of the 
investigation, but for the Jim Walters' investigation, for example, 
MSHA did not conduct a single public hearing to ensure that all 
relevant information was presented.
    On September 24, 2001, MSHA withdrew its proposed rule on Training 
and Retaining of Miners. This critical proposal would have increased 
the number of hours operators are required to set aside annually for 
health and safety training of miners. This training includes first aid, 
donning and using self-contained self-rescue devices, fire drills, and 
exiting the mine in the event of an emergency. Miners' lives may turn 
on the quality of their training. The training must be updated and 
improved. There should be no doubt that the better trained miners are, 
the better equipped they will be to escape a mine emergency.
    These are only five examples of the18 regulations the Agency, under 
this Administration, determined to be insignificant, too burdensome, or 
too costly to promulgate. Several of them may have had significance in 
what developed at the Sago Mine. We may never learn that for a 
certainty, but now is the time to require better rules, offering better 
protections so that miners have a better chance of surviving mine 
emergencies.
    Congress can, and should demand that MSHA do in 2006 all that 
Congress demanded in 1969 and again in 1977. Regulations that were in 
the pipeline in 2001 and 2002 should be reactivated and finalized in a 
timely fashion; regulations that are already in place must be enforced 
regularly and aggressively. Now that the spotlight is on the issue of 
miners' health and safety, we have a unique opportunity to make 
improvements.
    The status quo is inadequate. The government failed the Sago 
miners, and when it failed them it failed all miners. In enacting the 
Mine Act, Congress plainly stated: ``Congress declares that (a) the 
first priority of all in the coal or other mining industry must be the 
health and safety of its most precious resource the miner.'' (30 U.S.C. 
Sec. 801.) We take that admonition seriously; everyone else associated 
with the mining industry must re-establish miners' health and safety as 
their top priority, too.
    I thank you for your interest in miners' safety and would be happy 
to answer your questions.

    Senator Specter. Thank you, very much Mr. Roberts. I might 
say at this point, if there are other key factors you want to 
call to the attention of the subcommittee which will not be 
covered in the questions session, we would welcome your 
supplemental responses if they're not covered in any written 
testimony which has been submitted. Attachments to your 
testimony will be in the committee files.
    We now turn to Mr. Ben Hatfield, President and CEO of 
International Coal Group. Prior to joining the company, Mr. 
Hatfield held leadership positions with the Eastern Operations 
of Arch Coal, El Paso's Energies Coastal Coal Company, and 
Massey Energy Company, Bachelor of Science, and Mining 
Engineering from Virginia Tech. Thank you for coming in today 
Mr. Hatfield, and the floor is yours.

STATEMENT OF BENNETT K. HATFIELD, PRESIDENT AND CEO, 
            INTERNATIONAL COAL GROUP, ASHLAND, KENTUCKY
    Mr. Hatfield. Thank you Mr. Chairman and members of the 
subcommittee. I appreciate the opportunity, and invitation to 
testify here today. First, I want to extend my deepest 
sympathies to the families whose loved ones were lost at the 
Sago Mine. Our communities continue to mourn the deaths of our 
friends and coworkers. We also pray for the full and speedy 
recovery of Randy McCloy, our lone survivor.
    Second, I commend the heroic efforts made by many mine 
rescue teams from other companies, and volunteers who came 
forward during our community's time of need. The outpouring of 
support from the Buckhannon area communities has overwhelmingly 
demonstrated how West Virginians come together in times of 
crisis. We also appreciate the support provided by the Federal 
Mine Safety and Health Administration, and the State of West 
Virginia.
    Our company is working closely with MSHA and the State in 
the formal investigation to determine the cause of this tragic 
accident. I personally promise to use all available resources 
to get the answers that the friends, families, and coworkers 
need and deserve.
    My written statement details the events of January 2 
through 4, beginning with an explosion at about 6:30 a.m. in 
the midst of a violent lightening storm. This explosion in an 
instant separated brother from brother, uncle from nephew, and 
friends from friends. The supervisors onsite immediately 
reacted with a rescue effort in an attempt to reach their 
fellow miners. Sadly, their efforts were to no avail. It would 
be many more hours before conditions improved to allow 
additional rescue teams to enter the mine, and throughout we 
waited, waited, and then waited more.
    During those dark hours we prayed for 13 miracles, and in 
the end, were blessed to receive one. We deeply regret the pain 
caused by that early and erroneous communication to the 
families. There was never any intent to misinform, mislead, or 
raise false hopes. We, like you, and like the families of the 
miners, and like Americans throughout our country road that 
emotional roller coaster that has sadly lead us to where we are 
today.
    Within the next few days, investigators should be able to 
safely enter the mine to begin determining the cause of the 
accident. After a devastating accident of this nature, it is 
understandable that questions are asked about the Sago Mine's 
safety record. My written statement details some of the safety 
improvements made by our company beginning on June 1, 2005, 
when we assumed management oversight. What is important to 
emphasize is that since June 1, our company has worked closely 
with Federal and State Regulators to make this mine as safe as 
possible. For example, we were the first coal company to 
voluntarily invite MSHA's technical support group on incident 
reduction, to help implement a new program to continually 
improve mine safety. Most importantly, that effort helped us to 
dramatically reduce the injury rate at the Sago Mine by nearly 
60 percent from the first half of 2005 to the second.

                           PREPARED STATEMENT

    I believe that if there is any good that can come of this 
horrible event, it will be in inspiring greater innovations in 
mine safety. Although it's too early to determine the cause of 
the tragedy, we intend to be a leader in efforts to prevent 
future tragedies. As an example, our company is already moving 
forward with evaluation of improved wireless communication 
technology, and supplemental oxygen supplies on demand. We will 
work on our own, and with others in the mining community to 
improve technology, and we will continue to base our business 
decisions on worker safety as the first, and most critical 
consideration.
    We must learn lessons from this explosion that will better 
protect coal miners. That is our company's commitment to the 
families of the 12 miners who perished. Thank you.
    [The statement follows:]

               Prepared Statement of Bennett K. Hatfield

    Mr. Chairman and members of the Subcommittee, I am Ben Hatfield, 
President and Chief Executive Officer of International Coal Group, Inc. 
(ICG). By way of brief background, International Coal Group, Inc. is a 
leading producer of coal with operations in West Virginia, Kentucky, 
Maryland, and Illinois. We employ approximately 2,100 people throughout 
our operating area. During 2005, we sold about 19 million tons of coal 
to utility, industrial, and metallurgical customers located throughout 
the Eastern U.S. I appreciate the invitation to testify today.
    First, I want to extend my deepest sympathies to the families whose 
loved ones were lost at the Sago Mine. Our community will continue to 
mourn the deaths of our friends and coworkers. We also pray for the 
full and speedy recovery of Randal McCloy, Jr.
    Second, I commend the heroic efforts made by many mine rescue teams 
from other companies and volunteers who came forward during our 
communities' time of need. The outpouring of support from the 
Buckhannon area communities, churches, local businesses, civic 
organizations, and emergency personnel has overwhelmingly demonstrated 
how West Virginians come together in times of crisis. We also 
appreciate the support provided by the Federal Mine Safety and Health 
Administration (MSHA) and the State of West Virginia.
    Our Company is working closely with MSHA and the State of West 
Virginia in the formal investigation to determine the cause of this 
tragic accident. We vow to use all available resources to get the 
answers that the families, friends, and coworkers need and deserve.

                        BRIEF SUMMARY OF EVENTS

    I would like to briefly summarize what we know about the events of 
January 2, 3 and 4, 2006. At approximately 6:00 A.M. on Monday morning, 
January 2, after their travel route and worksites had been reported 
safe by certified safety examiners, two production crews and mine 
support staff totaling 27 miners entered the Mine. Each production crew 
needed to travel about two miles to reach the two working sections 
(First Left and Second Left) of the Mine where they were scheduled to 
work. The Second Left crew entered the Mine on a rail manbus that 
departed roughly 10 minutes ahead of a similar manbus carrying the 
First Left crew. One of the certified safety examiners had remained 
underground and traveled to his normal workstation. Therefore, a total 
of 28 miners were underground.
    At 6:31 A.M., Sago mine management heard the audible alarm of the 
mine monitoring system indicating the presence of carbon monoxide 
underground. At about the same time, the electrical power supply to the 
Sago Mine was disrupted. All of this occurred in the midst of a violent 
storm with unusually strong lightning strikes. Shortly thereafter, the 
supervisor of the First Left production crew telephoned the dispatcher 
on the surface to report that his crew had just experienced a very 
strong rush of air with substantial smoke and dust emanating from 
deeper in the Mine. Mine management directed the First Left supervisor 
to bring his crew out of the Mine through one of the two primary 
escapeways. No communication was received from the Second Left crew of 
miners. Repeated efforts by mine management and the First Left crew to 
contact the Second Left crew, via mine phone and underground walkie-
talkie, were unsuccessful. So, mine management immediately became 
concerned that they were in danger.
    At 6:41 A.M., Mine Superintendent Jeff Toler, whose uncle was one 
of the missing miners, and three other mine supervisors headed 
underground to investigate. After traveling about one and a half miles 
by rail manbus, they encountered the First Left crew coming out of the 
Mine on foot. The rail manbus was given to the First Left crew to 
expedite their safe exit from the Mine. The supervisor for the First 
Left crew, whose brother was one of the missing miners, joined the mine 
management team in an attempt to reach the Second Left crew located 
roughly 2,000 feet deeper in the Mine. They quickly gathered tools and 
ventilating materials and then proceeded toward the Second Left 
section. This initial rescue effort by the five-man management team 
continued for over two hours as the group encountered thick, black 
smoke and attempted to redirect ventilating air to open a route of 
access to the missing crew. Repeated calls to the Second Left crew via 
mine phone received no response. The rescuers became increasingly 
concerned that a possible explosion could be ignited as they directed 
fresh air toward the Second Left section. Consequently, the mine 
management group exited the Mine at 9:45 A.M.
    Meanwhile on the mine surface, at about 7:00 A.M., company safety 
managers not already on site were called and briefed on events at the 
mine. Following various communications between those safety managers 
and mine management on site regarding immediate emergency procedures 
required, we began calling MSHA and State safety officials to report 
the accident at about 7:40 A.M. Both MSHA and State safety officials 
were reached between 7:56 and 8:28 A.M., and began arriving on site 
soon thereafter. At 8:32 A.M., MSHA inspector Jim Satterfield orally 
implemented an emergency mine closure order (a ``103k order'') 
prohibiting further entry to the Mine. At about the same time, State 
mine inspectors began monitoring the air quality at the Mine portal. 
High concentrations of carbon monoxide were found, indicating 
significant risk of an active underground mine fire that could ignite 
an explosion, so state and Federal mine regulators on site determined 
it was not safe for mine rescue teams to enter. This agonizing process 
of monitoring the carbon monoxide and methane levels in the Mine air 
and having to wait and wait for confirmation that it was safe to enter 
would continue throughout the day.
    In anticipation that mine rescue teams were going to be allowed to 
enter the Mine soon, the first call to the Barbour County Mine Rescue 
Team was made at 8:04 A.M. That Team arrived on site at approximately 
10:40 A.M., and waited for State and Federal authorities to approve 
their entry into the Mine. Other mine rescue teams were also contacted, 
and continued to arrive at the Mine through the course of the day, with 
eventual deployment of 13 to 15 mine rescue teams by late afternoon on 
January 2.
    After experts from MSHA, the State of West Virginia, and the 
Company agreed that an underground mine fire was no longer likely based 
on the air monitoring results, the first mine rescue team entered the 
Mine (carrying special breathing apparatus) at 5:51 P.M. The search and 
rescue efforts continued throughout January 2 and January 3. Progress 
had to be careful and deliberate to protect the safety of rescuers, 
given that many rescuers had fallen victim to secondary explosions in 
coal mine disasters of years past. During the evening of January 3, a 
rescue team found one miner's body. Just before midnight, our remaining 
12 missing miners were found. As a result of the extreme difficulties 
in communication hundreds of feet below the surface while wearing 
special breathing apparatus, the now well-known miscommunication about 
the number of survivors occurred. We, too, rode that same emotional 
rollercoaster and suffered the inevitable pain when the truth was 
learned.
    We expect that investigators will be able to safely get back into 
the Mine soon to determine the cause of the accident.

                        SAFETY AT THE SAGO MINE

    Even before we completed the acquisition of the Sago Mine, on 
November 18, 2005, we assumed management oversight through a consulting 
agreement that allowed us to begin making safety improvements as of 
June 1, 2005. Since that time, our Company has worked closely with 
Federal and State regulators in an effort to make this Mine as safe as 
possible. Specifically ICG has voluntarily:
  --Rehabilitated two miles of primary intake escapeway and more than 
        doubled the amount of fresh air reaching the working sections. 
        This is the escapeway used by the surviving crew.
  --Upgraded the rail system used to move miners and supplies into and 
        out of the Mine.
  --Invited MSHA's Technical Support Group on Incident Reduction to 
        help implement a new program to continually improve mine 
        safety. We were told that ICG was the first coal company to 
        voluntarily work with MSHA under the Agency's Incident 
        Reduction Program.
  --Required Sago Mine hourly employees to receive eight hours of 
        supplemental safety training during September 2005, in addition 
        to the extensive training already required under the Mine Act. 
        Then, during October-December, we required our Northern West 
        Virginia Region supervisors to receive two days of supplemental 
        training at the MSHA Training Academy at Beckley, West 
        Virginia.
  --Established a Performance Group Initiative that gives every 
        employee a forum for addressing any safety or production 
        concerns or suggestions, anonymously, if they so chose, and 
        addressed any points raised in monthly meetings.
    These voluntary initiatives helped us to dramatically reduce the 
lost time injury rate at the Sago Mine by nearly 60 percent from the 
first half of 2005 to the second.
    Sago's employees are well-trained, skilled coal miners who 
understand safety. Each employee is aware that if an unsafe condition 
is identified, they are authorized to withdraw immediately from the 
hazardous area and notify their supervisor of the danger.
    MSHA data shows that:
  --Mining operations at the Sago Mine more than doubled between 2004 
        and 2005, prompting MSHA to dramatically increase--by 84 
        percent--its on-site inspection and enforcement presence.
  --Of the 208 citations, orders, and safeguards issued in 2005, none 
        involved an immediate risk of injury and all but three had been 
        fully corrected by January 2. The three remaining issues, which 
        relate to roof control, are being addressed by Sago in 
        compliance with the Mine Act.
  --Only when MSHA completes its investigation will we know the cause 
        of the accident, but we do know that none of the health and 
        safety violations cited by MSHA at Sago Mine last year involved 
        immediate risk of injury and that the Mine has worked to 
        correct all health and safety problems in accordance with the 
        requirements of the Mine Act.
    The Mine Act also authorized MSHA to shut down an operation that is 
unsafe, and MSHA's trained mine safety professionals, who were at the 
Mine nearly every working day in the several months before the 
accident, would certainly not have allowed the continued operation of 
the Sago Mine if they believed it to be unsafe. In addition, as 
required by law, our certified mine examiners inspect the Mine before 
and during every shift. They, too, are fully authorized to shut down 
any part of the Mine they consider unsafe. While the tragic events of 
January 2 confirm that we must be ever vigilant on mine safety, the 
safety record at the Sago Mine demonstrates that our management team 
aggressively focused on mine safety and protecting our people.

                       THE FUTURE OF MINE SAFETY

    Although it's far too early to determine the cause of the tragedy 
and the extent to which it may have been preventable, we intend to be a 
leader in the effort to identify and develop safety technologies that 
will help to prevent future tragedies. We will work on our own, and 
with others in the mining community, to improve technology, and we will 
continue to base our business decisions on worker safety as the first 
and most crucial consideration.
    We expect that this terrible series of events will further motivate 
the entire mining community to identify and implement significant 
improvements in mine safety through cooperation, information sharing, 
and improved technologies. For example, working with MSHA, we should 
vigorously seek to advance the development of permissible wireless 
communications and breathing apparatus technologies that could further 
improve the coal industry's underground mine rescue capabilities. Also, 
this experience highlighted critical weaknesses in the design of MSHA's 
V-2 robot that could likely be remedied with the technology now used by 
NASA in space exploration. Once the actual cause of the Sago explosion 
is known, there may be more specific measures that could help prevent a 
recurrence.
    We must learn lessons from this explosion that will better protect 
coal miners. That is our Company's commitment to the families of the 12 
miners who perished.

    Senator Specter. Thank you very much Mr. Hatfield. Our next 
witness is Mr. Chris Hamilton, Senior Vice President to the 
West Virginia Coal Association, an organization which he has 
been affiliated with for more than 20 years, has a Mine Foreman 
Certifications from West Virginia to Ohio, Undergraduate and 
Graduate Degrees from West Virginia University. I thank you for 
joining us Mr. Hamilton, and we look forward to your testimony.

STATEMENT OF CHRIS R. HAMILTON, SENIOR VICE PRESIDENT, 
            WEST VIRGINIA COAL ASSOCIATION, CHARLESTON, 
            WEST VIRGINIA
    Mr. Hamilton. Mr. Chairman, Senator Byrd, other 
distinguished members of the committee, thank you for the 
invitation to address this committee, and for placing this 
important topic on your agenda for review and discussion. In 
West Virginia, we have approximately 40,000 individuals, men 
and women who work directly in, or around a coal mining 
facility, and without exception all miners, managers, 
engineers, support staff, along with our entire State, have 
been deeply saddened by the Sago and Alma tragedies, and the 
mourning will continue for years to follow.
    Our hearts and prayers are with the families and loved ones 
of the miners who perished at Sago and Alma, and we continue to 
pray for Randal McCloy's full and speedy recovery. We now 
extend those prayers to the families of Don Bragg and Ellery 
Hatfield of the Alma Mine tragedy as well.
    West Virginians share a special bond with their families, 
church, and communities. They have an unparalleled inner 
strength, inner faith, and nowhere is that bond more prominent 
than in West Virginia's coal industry. For the record, the West 
Virginia Coal Association wholeheartedly embraces Governor 
Manchin in his plain spoken sentiments, that no miner should 
ever be fatally injured in a West Virginia coal mine. We also 
fully support the Governor's commitment to operate the safest 
mines in the world. We will commit the necessary resources over 
the months to come, and we'll do everything humanly possible to 
achieve that shared goal.
    First and foremost, that is our commitment which we believe 
is realistic and achievable. We also maintain that the primary 
responsibility for achieving that goal rests firmly with those 
who own, operate, and manage coal mining operations. A 
responsibility we not only acknowledge, but aim to fulfill. 
These tragic events have caught the eye of practically all of 
America in the past three weeks, and the media has done a good 
job, an accurate portrayal of the courage and overall character 
of the men and women who have selected mining as a profession. 
They have a passion for their work, and they do it with great 
pride, in an exceptional level of professionalism.
    Unfortunately, the events of January 2 and those of last 
week, have not accurately portrayed how technologically 
advanced mining has become, and all the progress, and safety 
achievement that's been made over the past several decades. 
But, one mining death is too many, and despite all this 
progress recorded in recent years, we now realize that much 
work remains. Particular focus is required in the post accident 
phase, so that the effect of accidents can be truly minimized 
or mitigated.
    By its nature, mining is unique unlike any other business 
or industry, and that it is dependent on natural conditions, 
and natural geology. Through their skills, training, 
dedication, and hard work, miners attempt to control and manage 
the challenges of their environment, and they're good at it. It 
requires a supreme vigilance every minute of every shift. New 
mining technologies such as long wall mining, remote controlled 
equipment design, and mine wide atmospheric monitoring systems, 
combined with the extraordinary skill and experience level of 
today's workforce has truly lead to safer conditions, and fewer 
accidents.
    As a relevant part of my testimony on record today, I 
incorporate a copy of the most recent directory of mines, which 
is published annually by the West Virginia Office of Miners 
Health Safety and Training. It contains useful statistical 
information, and also charged the Mine Safety Performance of 
the industry over the years. This directory reflects a dramatic 
reduction in mine related deaths since the passage of the 1969 
Act, from 162 fatal accidents reported, to just three for all 
of 2005. Again despite all this progress, we realized that much 
work remains, and that vigilance will be continued day in, and 
day out.
    This report also depicts a significant reduction of mine 
accidents, and lost time injuries over the same period. The 
State's annual report also reveals that the State of West 
Virginia has one of the more comprehensive mine safety programs 
found anywhere in the country, with a full compliment of mine 
safety inspectors, safety officials, and extremely aggressive 
legislative and regulatory program. I'm pleased to report that 
the State of West Virginia is taking a lead in dealing with 
this terrible tragedy that we've all experienced. I believe we 
will learn later this afternoon that three important pieces of 
legislation will be introduced into the West Virginia 
legislature for immediate action today. I believe that 
legislation will address the points raised by Senator Byrd in 
his opening remarks. I think we'll see legislation calling for 
a central rapid response program, and modeled after our State 
Office of Emergency Services. I think we'll see requirements 
calling for additional self contained breathing apparatuses 
stored at strategic locations underground, and I think we'll 
see legislation that preempts, and that drives technological 
advances to have improved communication systems within the mine 
itself, and from above the surface to the underground mine, and 
connected with all miners. I'm convinced we'll see that. I'm 
confident we'll see that legislation introduced today, and it's 
my understanding that by mid afternoon, that legislative action 
ought to be well on its way.
    We have joined with the Governor, we want to be part of the 
solution, we recognize despite all of the improvement, all the 
technological advancements, that much work remains. We are 
truly here, part of this proceeding to offer our help and our 
assistance to my colleagues sitting at this table, to the State 
of West Virginia, and to this particular Subcommittee. We have 
a wealth of technical capability.
    Senator Specter. Mr. Hamilton, you're substantially over. 
Could you summarize at this point?

                           PREPARED STATEMENT

    Mr. Hamilton. Yes. Again, I can't express enough the sorrow 
that is shared by all West Virginians. We compliment you Mr. 
Chairman, other senators for placing this on your agenda, and 
we look forward to working with you in the months to come, to 
address these concerns. Thank you.
    [The statement follows:]

                Prepared Statement of Chris R. Hamilton

    Mr. Chairman, Members of the Committee: Thank you for the 
invitation to address this Committee and for placing this important 
topic ``Coal Mine Health & Safety'' on your agenda for review and 
discussion.

                              INTRODUCTION

    My role and contribution to today's hearing will be defined by the 
following four (4) key points: First, to express our heartfelt prayers 
for the families who suffered great personal loss at the Sago Mine. Our 
prayers continue for Randal McCloy's full recovery and for his wife and 
family. We now extend those prayers and our state's unique circle of 
support to the families of Don Bragg and Ellery Hatfield of the Aracoma 
Mine tragedy. The deceased miners will forever be with us as we 
implement the necessary steps to improve coal mine safety and prevent 
recurrences. We also thank the mine rescue team members, the State, 
Federal, and company officials who directed and guided their heroic and 
brave efforts at Sago, and whose performance in those dark and anxious 
hours will be analyzed for years to come. It is our hope that their 
performance will be constructively reviewed with an eye towards 
improving future rescue efforts; Thirdly, we are here as one of the 
nation's largest trade associations to offer our pledge to work with 
you in whatever capacity you deem appropriate in the discharge of your 
important work and to direct our Association's collective attention 
towards the identification and implementation of appropriate remedial 
measures; and Fourth, and subordinate to the preceding points, is the 
perceived need to preserve the integrity and future of the coal 
industry--to implement the necessary changes from the lessons learned 
from the horrific accident that brings us here today and to elevate the 
understanding and appreciation of our industry which means so much to 
West Virginia and to our nation!
    My personal background: I have nearly thirty-five years of 
experience in the coal mining industry beginning in 1971 during the 
immediate implementation of the 1969 Federal Mine Health and Safety Act 
and over thirty years of experience in mine health and safety.
    I worked as an underground miner and for underground and surface 
mining companies. I have also worked for the Federal and State mine 
safety agencies as a mine safety professional and safety instructor--
certified to train and certify miners in all aspects of mining and mine 
safety including mine emergency preparedness and mine rescue 
operations.
    As Training Director for the West Virginia Department of Mines (for 
then Governor Jay Rockefeller), I was responsible for approving mine 
training facilities, mine training plans and individual mine training 
instructors.
    I possess underground Mine Foreman--Fire Boss certifications from 
WV and the state of Ohio where I worked for several years in the 
industry. I received my undergraduate and graduate degree from West 
Virginia University and have also completed many college level courses 
in mine safety, mining technology and mine industrial engineering.
    I presently serve under gubernatorial appointment on the West 
Virginia Coal Mine Health and Safety Board; the West Virginia Mine 
Safety and Technical Review Committee; The West Virginia Board of Miner 
Training Education & Certification; and, the West Virginia Diesel 
Equipment Commission.
    During my tenure on as a mine safety official, I have been involved 
in the review/investigation of serious mining accidents and practically 
every single mining death in West Virginia for the past twenty-five 
years.
    As a member of the West Virginia Board of Coal Mine Health & Safety 
(the only independent entity in West Virginia with a statutory charge 
to investigate and respond to mine accidents), I will be part of the 
state's investigation and regulatory response to the Sago and Aracoma 
accidents!
    West Virginia's coal industry is comprised of approximately 40,000 
individuals who work directly in, or around a coal mining facility and 
without exception, miners, managers, engineers and support staff along 
with our entire state have been deeply saddened by the ``Sago and 
Aracoma tragedies'' and will continue to mourn for years to follow.
    Our hearts and prayers are with the families and loved ones of the 
miners who perished in the Sago incident and we continue to pray for 
Randal McCloy's full recovery. We now extend those prayers and our 
state's unique circle of support to the families of Don Bragg and 
Ellery Hatfield of the Aracoma mine tragedy. I would observe that next 
to the immediate families of the deceased miners, nobody is saddened 
more than mine management officials over this tremendous loss. West 
Virginians share a special bond with their families, church and 
communities.
    They have an unparalleled inner strength and inner faith and no 
where is that bond more prominent than in the coal industry.
    For the record, The West Virginia Coal Association wholeheartedly 
embraces Governor Manchin's sentiments ``that no miner should ever be 
fatally injured in a West Virginia coal mine''. We also fully support 
the Governor's commitment to operate the safest mines in the world! We 
will commit the necessary resources over the months to come and will do 
everything humanly possible to achieve that shared goal!
    First and foremost, that is our commitment which we believe is 
realistic and achievable!
    We also maintain that the primary responsibility for achieving that 
goal rests firmly with those who own, operate and manage coal mining 
operations. A responsibility we not only acknowledge but aim to 
fulfill!
    These tragic events have caught the eye of practically all of 
America in the past three weeks and the media has presented an accurate 
portrayal of the courage and overall character of the men and woman who 
have selected mining as a profession. They have a passion for their 
work and they do it with great pride and an exceptional level of 
professionalism!
    Unfortunately, the events of January 2nd and those of last week 
have not accurately portrayed how technologically advanced mining has 
become and all of the progress and safety achievement that's been made 
over the past several decades. But one mining death is one too many and 
despite all the progress recorded in recent years, we now realize that 
much work remains! Particular focus is required in the post accident 
phase so that the effect of an accident can be minimized or mitigated!
    By its very nature, mining is unique (unlike any other business or 
industry) in that it is dependent on natural conditions and geology. 
Through their skills, training and hard work, miners attempt to control 
and manage the challenges of their environment--and they are good at 
it! It requires a supreme vigilance every minute of every shift.
    Undoubtedly coal mining is a dangerous occupation with unique 
hazards inherent to the workplace but I would maintain that mining is 
much safer today than what was realistically believed possible a few 
short years ago.
    New mining technologies such as longwall mining systems, remote-
controlled equipment design and mine wide atmospheric monitoring 
systems combined with the extraordinary skill & experience level of 
today's workforce has led to safer conditions and fewer accidents.
    As a relevant part of my testimony and record today, I incorporate 
a copy of the most recent ``Directory of Mines'' which is published 
annually by the West Virginia Office of Miners' Health, Safety & 
Training. It contains useful statistical information and charts the 
mine safety performance of the industry over the years.
    The ``Directory'' reflects a dramatic reduction in mining related 
deaths since passage of the 1969 Mine Safety Act when 162 fatal 
accidents were recorded to 3 for all of 2005. It also depicts a 
significant reduction in mine accidents and lost time injuries over 
this same period. 




    The state's annual report also reveals that the state of West 
Virginia has one of the more comprehensive mine safety programs found 
anywhere in the country with a full complement of mine safety 
inspectors, safety officials and an extremely aggressive legislative 
and regulatory program. It is also noteworthy to point out that no 
provision exists under Federal law for States to acquire ``primacy'' 
over the administration of mine safety laws.
    Consequently, all West Virginia mines are examined by State and 
Federal inspectors throughout each and every quarter.
    The significance of the industry and the important role coal plays 
in our everyday lives, which ranges from our basic quality of life to 
national defense and national security and should also serve as a 
tribute to the men and families of Sago!
    Over the past several weeks we have heard local, regional, national 
and international media sources all ask a similar question: Why do we 
continue to mine coal?
    Coal, and in particular, West Virginia coal, is crucial to our 
advanced society and extraordinarily quality of life. Coal continues to 
account for over fifty percent of the America's electricity. In West 
Virginia that figure is closer to 99 percent.
    Over the past several decades our state's coal industry has a 
remarkably record of safety achievement, reclamation accomplishments 
and environmental stewardship. We are coordinating proposed mine sites 
with local and state planning agencies to ensure meaningful and more 
productive development occurs.
    West Virginia is a shining example of where you can have a robust 
coal industry along with a thriving tourism industry--you can truly 
have both and I submit to you that nobody is doing it better!
    Today's industry represents a technologically advanced enterprise 
with a highly skilled and efficient workforce and has established a 
healthy presence in an international marketplace.
    West Virginia produces approximately 160 millions tons of coal 
annually. Of that total, over 105 million tons or 65 percent percent 
comes from underground mines and approximately 55 million tons of coal 
is produced from surface mines.
    West Virginia continues to lead the nation in underground coal 
production and is second only to the state of Wyoming in overall coal 
production. West Virginia is the world's leader in Longwall mining and 
is the leading coal export state.
    All told, West Virginia coal is shipped to 23 foreign countries and 
accounts for approximately one half of the United States total export 
product leaving domestic boundaries contributing immensely to the 
United States balance of trade.
    We also have more processing plants than any other state, more 
transportation outlets and one of the more elaborate transportation 
systems and infrastructures you find anywhere in the world. It is 
comprised of rails, trucks and barges and we have the best quality and 
variety of coals found anywhere in the world.
    Due to its clean and high quality, West Virginia coal is shipped 
throughout the eastern half of the United States to thirty-three states 
to generate electricity for industrial and household energy and for 
coking and steel production. West Virginia has the highest quality of 
coal found anywhere in the world and we have plenty of it (Reports of 
our diminishing reserve base has been wrongly placed)! We have over 52 
billion tons of demonstrated mineable reserves or 350 years of 
production remaining at today's production levels.
    The coal industry remains vitally important to our state and its 
economy. Together, with the states electric power industry, it accounts 
for nearly 60 percent of the total state business tax collections. 
These tax dollars translate directly into important education, 
government and community services and provide a reliable revenue stream 
for many other county, local and municipal programs.
    No other state business or industry affects so many people in so 
many different ways! It's overall impact is staggering in terms of 
employment, wages, taxes and overall economic activity.
    The state's industry is postured with an abundance of opportunity 
as the worlds thirst for low-cost, reliable energy grows on an 
incremental basis of nearly 2 percent annually. Thus, coal generally 
and West Virginia's coal particularly will continue to be a major 
player in the world wide energy mix on a going forward basis.
    A strong energy market and high demand has created an uplifting and 
positive energy around the state that most of us in the business have 
not witnessed since the 70s--and with that optimism comes the 
realization that we can do so much more if we are able to capitalize on 
todays opportunities. West Virginia Coal will be relied upon more than 
ever for industrial and household energy; domestic energy independence; 
national strategic defense; homeland security, and today's ever popular 
``coal-to-liquids'' and ``coal conversion technologies.''
    And lastly, just as all miners and mine managers have come together 
to grieve over the tragic events of the last three weeks, they all need 
to be part of the solution so we may effectively prevent a similar 
event in the future. They all have unique experiences and 
qualifications to contribute!
    Today, more than ever before miners, mine managers, engineers, 
research institutions and government officials need to become engaged 
to develop safer mining plans, better designed equipment and more 
effective ways to control our environment. Matters of safety, security 
and stability are shared responsibilities.
    And as the industry prepares to retrain its existing workforce 
along with the next generation of skilled miners, the ``Sago'' miners 
will be forever remembered and serve as a daily reminder of the supreme 
vigilance required in the workplace!
    Our membership has an abundance of safety, technical and 
operational expertise which has been called upon to respond to the 
challenges before us. We hereby extend those resources for your use and 
dedicate the same towards making the West Virginia Coal industry the 
safest in the world!
    I'll close by reciting the inscription on the ``the West Virginia 
Coal Miner'' statue located on the grounds of our state capital which 
captures the essence and summarizes best the importance of the coal 
miner and coal mining to West Virginia and to the Nation . . .

    ``In honor and in recognition of the men and woman who have devoted 
a career, some a lifetime, towards providing the state, nation and 
world with low-cost, reliable household and industrial energy . . . Let 
it be said that `Coal' is the fuel that helped build the greatest 
country on earth, has protected and preserved our freedom and has 
enhanced our quality of life. God bless the West Virginia Coal Miner.''

    Thank You.

    Senator Specter. Thank you very much, Mr. Hamilton. Our 
next witness is Mr. Davitt McAteer, Vice President of Sponsored 
Programs of Wheeling Jesuit University, former Assistant 
Secretary of Labor, Mine Safety and Health, worked as a 
consultant to the former West Virginia Governor, Bob Wise, a 
Bachelor's Degree from Wheeling Jesuit University, and a Law 
Degree from West Virginia University. Thank you for coming in 
today, Mr. McAteer, and we look forward to your testimony.

STATEMENT OF DAVITT McATEER, VICE PRESIDENT OF 
            SPONSORED PROGRAMS, WHEELING JESUIT 
            UNIVERSITY, SHEPHERDSTOWN, WEST VIRGINIA
    Mr. McAteer. Thank you, Chairman Specter, Senator Harkin, 
Senator Byrd, Senator DeWine. Thank you for inviting me to 
participate in this hearing on the Sago Mine disaster and on 
mine safety generally. It is tragic that we now must expand the 
scope of this inquiry to include the fatal mine fire at Massey 
Energies Alma Mine in Logan County, West Virginia. As you 
mentioned, I am presently Vice President of Wheeling Jesuit 
University, but previously served as the Assistant Secretary of 
Labor from 1993 to 2000. I have been involved in mine safety in 
a number of capacities since 1968.
    On Friday, shortly after the Sago disaster, Governor Joe 
Manchin called, and asked me to conduct an independent 
investigation to determine the cause of that disaster, to 
assess the rescue and recovery operation, and to make 
recommendations to improve mine safety in West Virginia. In 
accepting that assignment, I pledged that we would take every 
step to identify the problems that caused the disaster that 
contributed to the difficulties of the mine rescue operation, 
as well as to the missed communication to the Command Center, 
and the families, and that we would fix these problems.
    Today, I want to make that same pledge to the families of 
Don Bragg and Elvis Hatfield, the miners who lost their lives 
at Alma. We will hold public hearings, we will hold them on or 
about the 1st of March. We will have a report to the Governor 
on or about July 1. I understand, as Mr. Hamilton does as well, 
that Governor Manchin is this very morning, introducing three 
pieces of legislation in an effort to try and remedy problems 
that we have discovered and know exist as Cecil Roberts pointed 
out, know exist in the mine rescue area. Those are the rapid 
response system, more SCSR's, so you can get more time to get 
out, and new technologies to try and deal with them. Let me say 
here and now, that technologies do exist, and that is present, 
and available, that is off the shelf, and has been approved by 
MSHA.
    This is the tracking system that the PED, the Personal 
Emergency Device that Senator Harkin mentioned, this is 
currently available. It is in roughly a dozen mines in this 
country. It can be put into all the mines in this country, and 
one of the proposals Governor Manchin is making is to do just 
that, to put this device in the mines. Had we had this device, 
a signal could've been sent from the surface to the miners 
immediately upon the learning of the danger, and to tell them 
to evacuate.
    Second, this is the tracker system. This is a tracker that 
can be carried by an individual miner, and he can be located. 
These devices have been approved by the Mine Safety and Health 
Administration. I must disagree with Acting Assistant Secretary 
Dye. These devices have proved to be reliable. They have proved 
to be effective, both in this country and abroad. These devices 
are in mines in China. These devices are in mines in Australia. 
They were developed after disasters in Australia by government 
funded research that suggested that we need to be able to 
locate them. They are available to the industry, and it is my 
hope that the Senate and that the Administration would work to 
see that these devices get put into the mines immediately.
    Senator Harkin. How much do one of those cost?
    Mr. McAteer. This is the system that you put a loop--an 
antenna loop, a high end cost would be $100,000 for a mine the 
size of Sago. These devices are much less expensive, quite 
inexpensive; $15 to $20 is my understanding. I have literature, 
which I will submit for the record from the mine company.
    Obviously, we're not here to endorse this one company--the 
technology from this one company. We are hear to say, products 
exists that are out there, and are not being put into the 
mines. Unfortunately, this mining industry, and I must say as 
Assistant Secretary, we did make progress. We have brought the 
number of deaths down, but we have not brought the number of 
deaths down to where they need to be. But, the industry has 
focused its attention on productivity, and that is their want, 
and that is how they make a profit, and we don't take that away 
from them.
    But we do have to say, that the failure to focus attention 
on safety devices, and improvements, and put them in the mines 
is most unfortunate for those miners that we are all here and 
concerned about, and that needs to be changed. We need a safety 
system that will tell us at all times where miners are, and how 
we can get them out.
    Now, let me turn to the mine rescue system. In 1995, we 
held a conference, a summit of all parties in the industry, 
labor, management, State agencies, Federal agencies, and I 
chaired that summit, and asked the question, are we prepared 
now to deal with mine emergencies? The answer is we are not 
adequately prepared. This is in part due to the fact, that we 
are a victim of our success. We do not have the horrible number 
of disasters that we had when Senator Harkin's father was in 
the mines, but we have them often enough that we need to be 
prepared. We identified declining numbers of people 
volunteering. We identified declining equipment, we identified 
declining interests on the part of the companies in this area.

                           PREPARED STATEMENT

    Senator Specter. Mr. McAteer, your substantially over time. 
Could you summarize at this point, please?
    Mr. McAteer. I will sir, and I apologize.
    Senator Specter. It's good testimony.
    Mr. McAteer. We have the bravest mine rescue teams in the 
world. We have not provided them with good equipment, and a 
good system, and that needs to be fixed.
    Thank you, Senators.
    [The statement follows:]

                  Prepared Statement of Davitt McAteer

    Chairman Specter, Senator Harkin, and Members of the Subcommittee, 
thank you for inviting me to participate in this hearing on the Sago 
Mine Disaster and Mine Safety.
    I am presently the Vice President for Sponsored Programs of 
Wheeling Jesuit University. Previously I served as Assistant Secretary 
of Labor for Mine Safety and Health from 1993 to 2000, and I have been 
involved in the mine safety field in various capacities since 1968. I 
live in Shepherdstown, West Virginia.
    As you know, the Sago Mine Disaster was really two disasters--first 
the crisis of the trapped miners, and then the appalling 
miscommunication to the waiting families, leading them to believe for 
nearly three hours that all but one of the miners had been saved when 
in fact all but one had perished. Both parts of this dual disaster must 
be investigated and their causes identified. And we owe it to the lost 
miners, and to their families, to take all necessary steps to ensure 
that there will never again be such a dreadful double tragedy.
    Governor Joe Manchin of West Virginia called me soon after the 
disaster and asked if I would serve as his personal advisor and conduct 
an independent investigation of the cause or causes of the disaster, 
assess the rescue and recovery operation, and make recommendations to 
improve mine safety in West Virginia and across the nation.
    In accepting that assignment, which Governor Manchin announced on 
January 9, I pledged that we would pursue every lead, follow every 
avenue of inquiry, and take every step necessary to identify the 
problems that caused the disaster and contributed to the difficulties 
of the rescue operation and the miscommunication to the command center 
and the families, and that we would fix those problems.
    That is the charge of the Special Commission on the Sago Mine 
Disaster, and we are in the process of carrying it out. I am grateful 
to Governor Manchin for entrusting me with this responsibility, and I 
have promised to deliver a report to him by July 1.
    Subsequently the leaders of the West Virginia Legislature, House 
Speaker Bob Kiss and Senate President Earl Ray Tomblin, appointed six 
legislators to work with me to help determine what went wrong at the 
Sago Mine and, in the Speaker's words, to decide ``what we as a 
lawmaking body should do to minimize the chances of such a tragedy 
happening again.'' I appreciate the opportunity to work with these 
bipartisan colleagues, many of whom have been closely involved with 
coal mining.
    Before going further, I want to stress that our investigation is 
not intended to duplicate or in any way impede the investigations under 
way by the Federal Mine Safety and Health Administration (MSHA), the 
West Virginia Office of Miners' Health, Safety and Training, and the 
West Virginia Board of Coal Mine Safety and Health, or by any other 
regulatory agency or lawmaking body with statutory responsibilities in 
this area. But I must also stress that our investigation, unlike the 
others, has a very specific constituency to whom we are accountable.
    We are, of course, accountable to Governor Manchin, and through him 
to the people of West Virginia and the nation. But ultimately we must 
answer to twelve good men--Tom Anderson, Alva Bennett, Jim Bennett, 
Jerry Groves, George Hamner, Jr., Terry Helms, Jesse Jones, David 
Lewis, Martin Toler, Jr., Fred Ware, Jack Weaver, and Marshall Winans--
and to the families to whom they will return now only in memory. And we 
are accountable as well to the sole survivor of the disaster--Randal 
McCloy, Jr.--and to his wife and children and all those who continue to 
pray for his recovery. We must answer to them.
    Our responsibility, therefore, is to ask and explore all of the 
questions to which the miners' families must have answers, including 
these:
  --Was the Sago Mine being operated unsafely?
  --Were the responsible Federal and State agencies doing an adequate 
        job of enforcing safety and health laws and regulations at the 
        mine?
  --What triggered the explosion on the morning of January 2?
  --Could the explosion have been contained?
  --Why did it take so long to get the rescue operation launched?
  --Were the mine rescue teams adequately equipped and supported?
  --Was the command center properly organized and directed?
  --Were all communications between the command center and the rescue 
        teams properly verified and logged?
  --What caused the breakdown in communications at the mine?
  --What caused the miscommunication to the miners' families?
  --What were the reasons for the long delay in getting the facts to 
        them?
  --What steps must be taken to ensure that such a dual catastrophe 
        will never occur again?
  --Is the mine rescue system as effective as it should be?
  --As we absorb the lessons of the Sago Mine Disaster, what additional 
        steps must be taken to better protect miners nationwide?
    To help us address these questions, I am gratified that we are 
receiving support from miners, mine rescue teams, miners' families, 
mining companies, the United Mine Workers of America, Federal and State 
agencies concerned with occupational safety and health, and the entire 
West Virginia Congressional delegation.
    And, because this disaster has touched everyone in West Virginia--
and indeed has affected so many people in Pennsylvania as well, and 
throughout the coalfields of the United States--it is also gratifying 
that we are being assisted by student volunteers at Wheeling Jesuit 
University and West Virginia Wesleyan College and by engineering and 
law students at West Virginia University. They are the future of our 
state, and it is heartening that they have volunteered to help us build 
a safer future for those who dig the coal that creates the energy to 
drive our nation's economy.
    Most of our investigative work is ahead of us, and I do not want to 
pre-judge any part of that work or speculate about our findings and 
recommendations. But at the outset I do want to emphasize four points.
    First, we have nothing but praise for the mine rescue teams. I have 
been in charge of such operations, and I know the demands that a mine 
rescue makes on those who willingly risk their own lives to try to save 
the lives of others. I'm not sure that anyone but a miner can truly 
understand what it's like to go underground in the full awareness of 
hazards seen and unseen that can instantly end your own life, and to go 
anyway because there are miners trapped in there who are counting on 
you to come after them. That is a good working definition of heroism, 
and we will not second-guess anyone who went to the aid of the Sago 
miners.
    Second, we are going to get to the bottom of the Sago Mine 
Disaster, but we are not going to scapegoat anyone who was on the 
scene, including those who responded on behalf of MSHA and the West 
Virginia Office of Miners' Health, Safety and Training. It has been my 
privilege to know and work with many of those involved, and I want to 
say here that my strong belief, going in, is that the failures that 
occurred at the Sago Mine were primarily failures of the mine safety 
system, probably compounded by human fatigue and communications 
confusion, rather than failures that are somehow attributable to the 
personalities of those involved. These are good and competent people. I 
will be enormously surprised, and deeply disappointed, if our inquiry 
leads us to conclude otherwise about any of them. And we should all 
respect the fact that they are grieving the loss of their brother 
miners just as the miners' families are.
    Third, every aspect of our investigation will be open to the 
miners' families and the public. We will hold a public hearing in 
Buckhannon, West Virginia, on or around March 1, and we may schedule 
additional public hearings if needed. I can see no valid reason for 
conducting any part of our investigation behind closed doors. We need 
abundant sunshine to help determine what went wrong at the Sago Mine 
and how to strengthen mine safety protection and enforcement.
    Fourth, it is very clear that the nation has not invested as much 
money and energy in the cause of coal mine safety as in the pursuit of 
coal production and profits. That imbalance must be corrected, I 
believe, if we are to recover from the Sago Mine Disaster by making the 
mines of the United States as safe as they possibly can be made.
    I do not subscribe to the fatalistic view that coal mining is so 
inevitably dangerous that we must be willing to accept the loss of life 
as part of the price of coal. Let me point out that there are 
productive coal mines in West Virginia and elsewhere that have operated 
for twenty, thirty, forty years without a disaster. Tough Federal 
legislation, enacted in 1969 after the Farmington No. 9 Mine Disaster 
and subsequently strengthened, has helped to drive down the frequency 
of disasters and fatal injuries in the mines. And the best mine 
managers today know full well that safety and productivity go hand in 
hand.
    But legislation must be diligently and aggressively enforced to 
have meaning, and adequate funds must be committed to the protection of 
miners--both in day-to-day enforcement and in the development and 
implementation of technologies as advanced as those that we rely on in 
other fields, such as reducing the hazards of air travel, pinpointing 
the location of vehicles, or facilitating instantaneous, error-free 
communication around the globe and in outer space.
    In some industries, investing in state-of-the-art technologies is 
considered to be just part of the cost of doing business, and it must 
at long last be seen that way in mining as well. The energy industry 
can well afford to partner with the government in ensuring that new 
health and safety technologies are rapidly developed and put to work in 
the nation's mines.
    In 1995, we convened a conference on mine emergency preparedness at 
MSHA's National Mine Health and Safety Academy in Beckley, West 
Virginia. Out of that conference came many useful recommendations to 
improve mine rescue technology and communications, and a number of 
initiatives, including proposed regulatory changes, were subsequently 
advanced. But I share Senator Byrd's concern, which he expressed last 
week, that MSHA today may be ``under-staffed, under-funded, and under-
equipped.'' If that is true, it will be incumbent upon Congress to help 
fix the problem, so that MSHA can guide the mining industry in creating 
a truly modern model of mine safety.
    I want to mention one initiative that deserves our immediate 
attention and support regardless of the findings and recommendations 
that may emerge from the various investigations.
    It appears that precious time was lost on the morning of January 2 
while company officials attempted to contact MSHA, and perhaps while 
MSHA organized its response. That remains to be determined, but it is 
clear that MSHA's emergency response system is outmoded. Mine emergency 
communications need to be brought into this century, in part by 
establishing, perhaps at MSHA's National Mine Health and Safety Academy 
facilities, the equivalent of the nationwide 9-1-1 emergency system 
that is used to rapidly dispatch emergency personnel and other first 
responders. We should begin immediately to develop plans for such a 
nationwide system and to set target dates for implementing it.
    Governor Manchin is today introducing legislation to develop such a 
system for West Virginia. I want to commend him for his leadership in 
launching this initiative, and I hope that we will soon be able to 
expand the 9-1-1 system to protect miners nationwide.
    Again, I do not want to anticipate our investigation's findings and 
recommendations, but I think it is already clear that certain other 
safety improvements deserve priority consideration. These include:
  --Technological advances in mine communications.--If the technology 
        can be developed to facilitate space exploration and to equip 
        consumers with the ability to instantaneously transmit and 
        receive information worldwide from their laptops, cellphones, 
        and BlackBerries, there is no question that mine communications 
        technologies can be brought out of the dark ages. The only 
        question is whether we have the will to make the necessary 
        commitment to research and development. It is imperative that 
        we make that commitment and find the funds to fulfill it.
  --Improved mine safety equipment.--The Self-Contained Self-Rescuer 
        (SCSR), which provides an oxygen supply for at least one hour, 
        is a technology developed more than 50 years ago. The fight to 
        make the government require SCSRs took more than twelve years, 
        ending with the adoption of SCSR regulations in 1981, and the 
        sad fact is that there has been little industry support since 
        then for efforts to improve the SCSR so that it can sustain 
        life for longer periods of time. What is true of the SCSR is 
        true of mine safety technology in general. In the wake of the 
        Sago Mine Disaster an industry spokesman was quoted as saying 
        that it was not the industry's responsibility to develop new 
        safety equipment. ``We're not in the self-rescuer manufacturing 
        business,'' he reportedly said. That attitude harks back to the 
        19th century and has no place in the 21st. The industry may not 
        manufacture safety equipment, but going forward it has an 
        obligation to support accelerated research and development so 
        that technologies available elsewhere can be transferred to the 
        protection of miners with all deliberate speed.
  --Improved mine emergency training.--There are fewer mine disasters 
        today than in the past. That is the good news. The bad news is 
        that in such a situation it is difficult for all involved in 
        mine safety to maintain a constant state of high readiness. 
        Mine safety officials and mine rescue teams are generally well 
        trained, but there are few opportunities for them to test their 
        training and proficiency in real-world situations. Among other 
        things, we need to look into the feasibility of augmenting 
        today's training systems and improving response times by 
        conducting emergency rescue drills at operating mines on a 
        surprise, unannounced basis. Doing so would, of course, involve 
        considerable expense and inconvenience in temporarily halting 
        production, simulating emergency conditions, and deploying mine 
        rescue teams and support personnel. But we need to consider the 
        trade-offs in improving emergency readiness. The paradox is 
        that the safer we can make mining, the more we need to be fully 
        prepared for something to go terribly wrong anyway. If, for 
        example, the explosion at the Sago Mine turns out to have been 
        triggered by an abnormally severe lightning strike, it will 
        underscore the probability that we can anticipate but not 
        entirely eliminate all hazards and must be ready--totally 
        ready--when they suddenly put miners' lives in peril. Given the 
        importance of those lives and our reliance on the coal they 
        mine for our use, anything less than state-of-the-art training 
        should not be acceptable.
    I would add that there is also a need to address the issue of 
information control at the mine site. The Sago Mine Disaster was the 
first such emergency to occur in the cellphone era. While cellphones 
may or may not prove to be implicated in the miscommunication that 
occurred between the command center and the families waiting at the 
Sago Baptist Church, the presence of numerous cellphones at the site 
and the difficulty of controlling their use underscores the need to 
ensure that rescue and recovery information is strictly confined to the 
command center until it has been verified. This is not to suggest that 
there should be an information blackout during future rescue and 
recovery operations, but that only authenticated information about the 
progress of such operations should be released, by authority of the 
responsible official in charge of the command center.
    Governor Manchin has repeatedly pledged that the miners who died in 
the Sago Mine will not have died in vain. We can fulfill that pledge 
only by diligently working to identify what went wrong and then by 
taking every necessary step to fix what went wrong by improving the 
safety and health protection of miners. Your decision to convene this 
hearing reflects your commitment to this goal and should be applauded. 
I look forward to sharing our investigative findings and to working 
together in the months ahead to make America's mines safe.
    I would be glad to answer any questions and to provide any 
additional information that may be helpful to you. Thank you.

    Senator Specter. Thank you very much, Mr. McAteer. Our 
final witness on this panel is Mr. Bruce Watzman, Vice 
President for Safety, Health, and Human Resources for the 
National Mining Association, appointed recently to the Mine 
Safety and Health Research Advisory Committee, has 
Undergraduate Degrees in Economics and Psychology, and a 
Postgraduate Degree in Environmental Health Management. Thank 
you for joining us today Mr. Watzman, and the floor is yours.

STATEMENT OF BRUCE WATZMAN, VICE PRESIDENT, SAFETY AND 
            HEALTH, NATIONAL MINING ASSOCIATION, 
            WASHINGTON, DC
    Mr. Watzman. Thank you, Mr. Chairman. On behalf of the 
National Mining Association, its member companies in coal and 
mineral production, I too, once again, extend our heartfelt 
sympathies to the families in West Virginia. These tragedies 
leave no one in the industry untouched. The events have shaken 
the mining community, and all committed to mine safety. We join 
with others here today, to ensure that out of these tragedies, 
will emerge a stronger resolve, and greater cooperation in the 
pursuit of mine safety. Our expectation is that from this, and 
similar hearings, and from the exhaustive investigations now 
underway, we can do better, what we've tried to hard to do 
well. It's in this spirit that I appear before this 
subcommittee today, to offer information on where we have been, 
and recommendations for what we can do to advance mine safety.
    As this committee considers the recommendations as a result 
of this hearing, I urge that we not create an atmosphere in 
which the parties feel the need to retreat to their respective 
corners of the ring, and defend themselves. Rather, let us 
create an atmosphere where we come to the center of the ring, 
stand together, and fight against the common enemy, workplace 
accidents.
    The coal mining industry takes seriously its commitment to 
protect its workforce. Since the first oil embargo in 1970, the 
coal industry has been called upon to provide more coal to meet 
our energy's needs. We've answered that call while providing a 
safer working environment for our workforce. Since 1970, coal 
production has increased 83 percent, and mine fatalities have 
decreased 92 percent. As we look at the first 5 years of the 
21st century, we see a continuation of that trend that began in 
early 1970's, safer coal mining. This demonstrates that safety 
and productivity are not competing goals, but rather 
complimentary goals. Working in what are inherently hostile 
environments, today's mining companies have proven that a well 
trained, experienced workforce, using state of the art 
equipment, can accomplish the dual goals of working safely and 
being productive, and yet, these accomplishments are diminished 
by what remains to be done, and what NMA, and its members are 
committed to working towards.
    Following a tragedy such as these, these statistics 
understandably pale before the names of those lost. The events 
of recent weeks have strengthened our resolve to work harder, 
and work smarter on mine safety. This effort must begin with a 
close and comprehensive examination of current safety 
practices, and procedures. Our ability to further advance coal 
mine safety and health, will require an examination of the 
structural and technologic hurdles that must be overcome. It 
will require a commitment to identify, and foster the 
development of 21st century technology that will perform 
effectively and reliably in the mining environment.
    Advances in technology have been integral to the safety 
improvements thus far, and will contribute to further advances. 
In pursuit of this goal, and to ensure a focused, and 
transparent effort, this week, the National Mining Association 
is announcing the formation of a National Safety Technology and 
Training Commission. The Commission will be drawn from safety 
experts for the purpose of examining technologies, emergency 
response, rescue procedures, and rescue conditions in our 
Nation's mines.
    The Commission will be chaired by a recognized expert in 
mine safety, Dr. Larry Grayson, Chairman and Professor of 
Mining and Nuclear Engineering at the University of Missouri-
Rolla. The Commission will report its preliminary findings by 
July 1, with a final report by the end of the year. We 
anticipate it will examine among other things, current and new 
promising technologies for communication, tracking miner 
location, rescue technology, and the methods to more reliably 
detect potential hazards.
    This subcommittee is very aware of the need to maintain a 
vibrant mining research program within the National Institute 
for Occupational Safety and Health. The tragic events we've 
witnessed underscore this need. The Federal Government has an 
important role to play in technology development, and we urge 
your continued support to strengthen this vital government 
function. It's especially important for us to continue to work 
together as partners, because as Mr. Roberts said, coal is an 
industry with a changing face. Many of the people who joined in 
the 1970's, and who have built a career producing America's 
energy, are now retiring. We all must work together to develop 
programs to train and educate a new generation of employees, so 
that they can have a safe and productive career in an industry 
vital to this country's energy markets, and national interests.

                           PREPARED STATEMENT

    Before closing Mr. Chairman, I would be remiss if I did not 
recognize the efforts of those who participated in the rescue 
activities. The efforts of these brave and often unheralded 
rescue team members cannot be minimized. We as an industry are 
fortunate to have these people as part of our mining family. We 
thank them for their service. Mr. Chairman, we stand in the 
ring with you and the members of this subcommittee ready to 
lead in the effort to advance coal mine safety. Thank you.
    [The statement follows:]

                  Prepared Statement of Bruce Watzman

    Good morning. My name is Bruce Watzman. I'm the Vice President for 
Safety and Health for the National Mining Association.
    On behalf of NMA and its member companies in coal and minerals 
production, I extend, once again, our heartfelt sympathies to the 
families who lost loved ones at the Sago mine.
    This tragedy leaves no one involved with the industry untouched. 
Anyone who has worked in a mine . . . or knows someone who does . . . 
feels deep sorrow.
    In addition, it compels all of us in the mining community to work 
harder towards the one goal we all share--zero fatalities.
    We join with others here today to ensure that out of this tragedy 
will emerge a stronger resolve and greater cooperation in pursuit of 
safer mines. Our expectation is that from this and similar hearings . . 
. and from the exhaustive official investigation now underway . . . we 
can do better what we've tried hard to do well.
    It is in this spirit that I appear before this subcommittee today--
to offer information on where we have been and recommendations for what 
we can do to advance mine safety.
    As this committee considers the recommendations as a result of this 
hearing, I urge that we not create an unproductive atmosphere in which 
parties feel the need to retreat to their respective corners of the 
ring and defend themselves. Rather, let us create an atmosphere where 
we come to the center of the ring stand together and fight against a 
common enemy--workplace accidents.

                          INDUSTRY PERFORMANCE

    The coal mining industry takes seriously its commitment to protect 
its workforce. Since the first oil embargo in the early 1970s, the coal 
industry has been called upon to provide more coal to meet our Nation's 
energy requirements. We have answered that call while providing a safer 
working environment for our workforce. Since 1970, coal production has 
increased by 83 percent and coal mine fatalities have decreased by 92 
percent.



    As we look at the first five years of the 21st century we see a 
continuation of the trend that began in the early 1970's--safer coal 
mining.




    This demonstrates that safety and productivity are not competing 
goals, but rather complimentary goals. Working in what are inherently 
hostile environments, today's mining companies have proven that a well-
trained, experienced workforce, using state-of-the-art equipment, can 
accomplish the dual goals of working safely and being productive. And 
yet these accomplishments are diminished by what remains to be done and 
what NMA and its members are committed to working towards. Following a 
tragedy such as this, these statistics understandably pale before the 
names of those lost.

                       MINE SAFETY AND TECHNOLOGY

    The events at the Sago Mine have strengthened our resolve to work 
harder and work smarter at mine safety. This effort must begin with a 
close and comprehensive examination of current safety and rescue 
procedures. Concurrent with a well-trained workforce and state-of-the-
art equipment, the coal industry has incorporated safety management 
into its business and moral ethic. Safety management comprises four 
functions: prevention, detection, first response and sustained 
response. These are not new to the coal industry; we've incorporated 
them for decades as sound safety practice dictates.
    Our ability to further advance coal mine safety and health will 
require an examination of the structural and technologic hurdles that 
must be overcome. It will require a commitment to identify and foster 
the development of 21st century technology that will perform 
effectively and reliably in the mining environment. Technologies such 
as the introduction of remote control miners, integrated methane 
monitors on mining equipment, atmospheric monitoring systems, longwall 
mining systems and canopies on equipment are a few of the advances that 
have contributed to the industry's improved safety record. Advances in 
technology have been integral to our safety improvements thus far and 
will, we believe, contribute to further improve mine safety in the 
future.

                     MINE SAFETY COMMISSION FORMED

    In pursuit of this goal and to ensure a focused and transparent 
effort, this week the National Mining Association is announcing the 
formation of a Mine Safety Technology and Training Commission. The 
commission will be drawn from safety experts in academia, labor, 
industry, and public and state agencies for the purpose of examining 
safety technologies, emergency response and rescues procedures and 
training regimes that could significantly enhance safety and rescue 
conditions in our nation's underground coal mines. The commission will 
be chaired by a recognized expert in mine safety, Dr. R. Larry Grayson, 
chairman and professor of mining and nuclear engineering at the 
University of Missouri-Rolla. Dr. Grayson will report the commission's 
preliminary findings to the public and mine safety authorities by July 
1, with a final report by the end of this year. We anticipate the 
commission will examine, among other items, the current and new 
promising technologies for mine communication, tracking miners' 
locations, rescue technology and methods to more readily and reliably 
detect potential safety hazards.
    This subcommittee is very aware of the need to maintain a vibrant 
mining research program within the National Institute for Occupational 
Safety and Health (NIOSH). The tragic events at the Sago Mine 
underscore this need. The Federal Government has an important role in 
technology development--in order to bring safer, new devices to a 
relatively small market for safety equipment. We urge your support to 
strengthen this vital government function.
    In addition to government participation, our industry will continue 
to examine how new technology and training can be adapted to further 
improve mine safety performance. We must continue to use the labor- 
business-government model that has served us well in the past on coal 
mine safety.
    It is especially important for us to continue to work together as 
partners because coal is an industry with a changing face. Many of the 
people who joined this industry in the 1970s, and who have built a 
career producing America's energy, are now retiring. We--government, 
industry and workers--all must work together to develop programs to 
train and educate a new generation of employees so that they can have a 
safe and productive career in an industry vital to this country's 
energy markets and national interests.
    In conclusion, if we work together, as partners, and if we focus on 
improvement we will continue to advance both the cause of mine safety 
and the cause of energy security.
    Before closing, Mr. Chairman, I would remiss if I did not recognize 
the efforts of those who participated in the rescue activities at the 
Sago Mine. The efforts of these brave and often unheralded rescue team 
members cannot be minimized. We as an industry are fortunate to have 
these brave individuals as a part of our mining family. We thank them 
for their service.
    We stand in the ring with you ready to lead in this effort to 
advance coal mine safety.
    Thank you, Mr. Chairman.

    Senator Specter. Thank you very much, Mr. Watzman. Before 
proceeding to the 5 minute rounds for questions, I'd like to 
place in the record the identities of the miners who've lost 
their lives. Mr. Tom Anderson, Mr. Alva Bennett, Mr. James 
Bennett, Mr. Jerry Groves, Mr. George Hamner, Jr., Mr. Terry 
Helms, Mr. Jesse Jones, Mr. Dave Lewis, Mr. Marin Toler, Jr., 
Mr. Fred Ware, Mr. Jackie Weaver, Mr. Marshall Winans from the 
Sago Mine, and also Mr. Randal McCloy, who is still 
hospitalized in a light coma, and the fatalities from the Alma 
Mine are Mr. Don Bragg and Mr. Ellery Hatfield. And on behalf 
of this subcommittee, and the full committee, and the Senate, 
and the whole Congress we express our sympathy to the families 
of those individuals.
    Let us proceed now, to the 5 minute rounds for members, and 
let me begin with you Mr. Hatfield. What steps has the company 
taken to compensate the families of the deceased miners?
    Mr. Hatfield. That's an excellent question Senator. Let me 
explain what we've taken to this point--the steps we have 
taken, first, we immediately sent a letter to each of the 
family members--surviving family members, confirming that the 
funeral costs would be covered, and also confirming to the 
families, the amount of the insurance coverage they would 
receive, by virtue of our companies benefits program, and the 
life insurance.
    Senator Specter. How much is that insurance coverage?
    Mr. Hatfield. It varies depending on the individual, from 
$164,000 to nearly $300,000 per miner.
    Senator Specter. What beyond that, will the company do for 
the families?
    Mr. Hatfield. Beyond that, recognizing that sometimes these 
issues take awhile to work through the system, and the 
insurance companies often need information that takes awhile to 
provide, we immediately----
    Senator Specter. I'm talking about the company's efforts 
beyond the insurance companies.
    Mr. Hatfield. We extended their salaries, they continued to 
get paid today just as much as they earned when their husband 
or brother was alive.
    Senator Specter. How long will that last?
    Mr. Hatfield. We have extended that for 6 months, to be 
sure that it safely bridged the period of time when the life 
insurance proceeds will be delivered, and when West Virginia 
Worker's Compensation awards are available to the families.
    Senator Specter. Anything beyond that?
    Mr. Hatfield. We've also extended medical care, full 
benefits, medical, vision, and dental through the entire COBRA 
period of 18 months at no cost.
    Senator Specter. Mr. Hatfield, let the subcommittee know 
what, if any, additional benefits the company intends to 
provide here.
    Mr. Hatfield. We will certainly do so Senator.
    Senator Specter. The issue Mr. Hamilton as you articulated, 
is to commit whatever resources are needed, that's a pretty 
broad statement, considering the necessary resources. We've 
seen Mr. McAteer testify about certain devices which could be 
very, very helpful. The thought crosses my mind, that the coal 
companies have been very profitable lately. Energy is in big 
demand. We know what has happened. Coal's an enormous resource. 
What about a potential for a user fee? When we wrestle with the 
budgetary problems, the allocations to the Department of Labor 
are at risk with other very, very high priorities, Pell Grants, 
Community Health Services, OSHA. What do you think Mr. Roberts, 
about the possibility of imposing a user fee to deal with your 
testimony about additional mine rescue teams, or what Mr. 
McAteer testifies about additional equipment?
    The real responsibility lies with the industry, as opposed 
to the taxpayers generally, and if you have a user fee, it's 
not subject to a one percent cut, it's not subject to 
bargaining with Pell Grants, or scholarships, or health for 
children. What do you think Mr. Roberts?
    Mr. Roberts. Let me make a couple of comments if I might, 
Senator, about that. One, is some of the suggestion is that 
have been made here today, particularly with respect to oxygen 
on the section. I believe that the cost of a canister is 
somewhere around $3,500 to $5,000 it's in that range, depending 
upon what type of oxygen you're talking about providing. The 
cost of running an extra telephone line into the mines is 
probably 10 cents, or 15 cents a foot, so those couple of 
things are minimal cost.
    I think the larger cost items going to get into--and let me 
just make a comment if I might, about mine rescue teams, and 
what the law says, and somehow we've strayed away from that, 
and I do want to say, and he may be surprised by this, but Mr. 
Hatfield commented in a press conference that he intended to 
place onsite a mine rescue team, consisting of miners who 
worked at that mine, and I think that is something that the law 
says should be done. But over the years, we've gotten away from 
that, and I think he's on the right track with respect to that. 
The law says that there will be two teams at every mine, and 
within 2 hours of the mine, but we don't require that to 
happen.
    Obviously since the--when the downturn in the economy hit 
in the 1980's, we got away from some of this. But we, I think, 
need to go back to a strict interpretation of the law. But to 
answer your question, our union would support any actions to 
fund protections for coal miners, because we should remember 
one thing here today, I believe Senator, when the act was 
passed in 1969, it said the coal miner was the most precious 
resource in the mining industry. So, whatever steps we have to 
take to protect the coal miners in this country, we should 
take.
    Senator Specter. Thank you, Mr. Roberts. The red light went 
on in the middle of your testimony, so I would yield at this 
time to Senator Byrd.
    Senator Byrd. Let me thank Governor Joe Manchin, his 
efforts have been recognized here appropriately. He has left no 
stone unturned in his desire to help the families of these 
stricken miners, and to help the industry, and the other 
elements that are working together to protect the coal miner. 
He is doing everything he possibly can.
    Let me also thank my colleagues, Senator Rockefeller, and 
Congressman Ray Hall--Nick Ray Hall, they were at the scene, 
they stayed at the scene. I also want to thank the other 
members of the delegation for the support that they give to 
coal mining, coal miners, and safety of coal miners, and I'm 
talking about Mr. Mollohan, and I'm talking about Ms. Capito. 
They are supporters of every feature of the law, and 
appropriations that come before the Congress.
    But let me say this too, I also want to thank the witnesses 
here today who've come. You've given splendid testimony, and 
you can give more in whatever you wish to put in writing for 
the record of this committee. You can be sure that it will be 
read, not just by the staff's of the Senator's, but by 
Senator's themselves. I particularly call to the attention of 
the committee that the members of the delegation would all like 
to have been here, but for one reason or another, some of them 
could not come.
    My question would be to Mr. McAteer, what needs to be done 
to get to the problem of habitual violators? Are the penalties 
sufficient, or not? Do the fines need to be increased? What 
suggestions do you have? What needs to be done to get to the 
problem of habitual violators, they just pay the fine? That's 
not enough; they go on with the violations. That happens, and 
is happening. How do we get at this problem?
    Mr. McAteer. Senator, thank you. Let me try to give two 
parts of an answer. First, as has been mentioned here by 
Senator Specter--Chairman Specter, that the price of coal has 
increased dramatically, and the price of the penalties have not 
increased dramatically, and as with all of us, unless it makes 
a difference in our monetary budgets then penalties are below 
what they need to be, don't have the impact that you want them 
to have, and I'm afraid that situation is where we have it 
today.
    Second, with regard to the habitual violator, there is a 
scheme, a pattern of violations scheme within the act, which 
was enacted in 1969 and amended in 1977, which is an effort to 
get at the--at that type of violator. That has had a limited 
use in the past 5 years. The second part of it, it can be used 
in several ways. It can be used one, as a notice to the 
operator that that change needs to be made, and that by sending 
the letter, and saying we're looking at you to put you on a 
pattern, is one way to do it.
    Another way to do it is to send an inspector with some 
degree of regularity to the mine, and then the mine company 
pays attention to the violations, because you have an inspector 
there. The ultimate responsibility, as everyone has testified 
here today though, rests with the mine operator. When you see 
the violations increasing, when you see the citations 
increasing, when you see the accidents rates increasing at the 
mines, then you as the mine operator must take action, and we 
the Federal Government and State government, have the 
responsibility to urge, and encourage, and use penalties to 
force that. But, it has to be on the part of the mine operator.
    So, I would say yes to the increase in fines, and yes to 
the increase in use of pattern of violations. That needs to be 
done, and it needs to be done immediately.
    Senator Byrd. Mr. Roberts, would you like to comment?
    Mr. Roberts. I'd like to just add Senator, thank you for 
the question, one thing to this issue of fines, its one thing 
to issue fines and penalties, and quite frankly, we have a 
number of Federal Inspectors out there who do a good job of 
that. But if you do some investigating here, you're going to 
find something. That on appeal, those fines are reduced to 
almost nothing, and so there are two issues here.
    One, what level should the fines be? But at the end of the 
day, when coal companies do spend a lot of money on attorneys, 
at the end of the day, if those fines get reduced to zero, the 
issue that I would ask you to look at, as Jim Walters resources 
in 2001 where 13 miners lost their lives, and the Government 
assessed huge fines there, but they were dropped to nothing. 
More or less, I think it was $3,000 dollars over a period of 
time, through legal work, and appeals. There's something wrong 
with the process when that can be allowed to happen.
    Senator Byrd. A miner knows his work best. The Mine Act 
guarantees every miner, union or non-union, the right to leave 
a mine whenever he feels that his safety is threatened. How can 
we ensure that miners feel that they're free to exercise that 
right without fear of retribution? Mr. Hamilton.
    Mr. Hamilton. Thank you, Senator Byrd. My response to that 
question is you're absolutely correct; there are more than 
adequate protections under the law. The law is crystal clear, 
that every miner in every mine has the ability to personally 
withdraw himself if he thinks that his personal safety is 
jeopardized, or the safety of others is compromised.
    I've been a coal miner, I've been around coal miners all my 
life. They're a very strong, courageous people. I have seen 
those rights exercised, and there are protections in the event 
that someone wants to discriminate against a miner, or any 
other person in our workforce for exercising those rights. 
There are ample adequate protections there as well. Those 
rights are exercised; those protections are clearly understood 
by the men and women who mine coal. Again quite honestly, 
there's an extraordinary skill, and qualification and maturity 
level amongst the miners in West Virginia, and these miners 
take great pride in their work, and if in fact there are unsafe 
conditions, they report those unsafe conditions. If in fact 
they cannot, or don't have the qualification to correct those 
conditions initially on their own.
    Senator Byrd. Mr. Chairman, thank you. You've been very 
liberal with your extension of time to me. I also again, want 
to thank the witnesses. I think the testimony has been helpful, 
and will be helpful. Thank you very much.
    Senator Specter. Thank you, Senator Byrd. Senator Harkin.
    Senator Harkin. Thank you, Mr. Chairman. Mr. Roberts, in 
your testimony you said something--you didn't say it, but it's 
in your testimony. I read it, it said just since August 2000, 
MSHA has records of well over 400 mine fires, ignitions, 
explosions, and inundations that far too easily could've 
developed into significant disasters and fatalities. Many other 
instances likely went unreported. Four hundred. Think about 
that, 400. Then you went on to say miners should not have to 
get sick, or to risk their lives just by going to work. Now, 
that's kind of a straight forward statement. People shouldn't 
have to risk their lives, or get sick just to go to work 
everyday, and this is why I'm trying to get to Mr. McAteer.
    The one basic point to make today for the reading, Mr. 
Roberts, from your testimony, is that as a Nation, and as an 
industry, we already posses the knowledge and the ability to 
prevent most of the deaths that are still occurring in the coal 
mining industry. What is needed is a real commitment by our 
Government, in this case MSHA, to do better.
    Senator Byrd. Amen.
    Senator Harkin. You know, that really sums it up. Now, I go 
to Mr. McAteer. Now you were there when I pointed out, that 
this one regulation on the SCSR's--self contained devices--
rescue devices, you promulgated that, and I think what--1998 
maybe, 1999--I'm sorry, I don't remember when. But, that was 
the one I asked Mr. Dye about that was done away with in 
December 2001, do you recall that?
    Mr. McAteer. Yes, Senator.
    Senator Harkin. Do you recall at that time, when it was 
withdrawn? This proposed rule came under your jurisdiction, it 
sat there, then it was withdrawn in December 2001. Do you 
recollect at that time, what you might've been thinking, or did 
you make any kind of statements at that time?
    Mr. McAteer. Senator, we didn't withdraw. It was done after 
I left.
    Senator Harkin. That's what I mean, it was done in December 
2001. You were gone at that time. I'm asking you, what did you 
think about it at that time; you were gone at that time?
    Mr. McAteer. If I could answer it in a little broader way, 
I think it would be helpful to the senators. To place the self 
contained, self rescuer devices in the mines in the 1980's, 
took a lawsuit on behalf of the individual miners. The 
companies who were making them said they had prototypes, but 
not production models. The industry said we didn't have orders, 
so we didn't order it, and the agency wouldn't promulgate the 
rule.
    We had the good fortune of going before Judge Higgenbotham, 
who was up in Philadelphia in the Third Circuit, and Judge 
Higgenbotham had been an engineer, and he said this is 
nonsense, put the devices in the mines. But, it took a lawsuit 
to get there. To try to enhance that--to try to enhance that 
device and others, it takes a tremendous amount of work. 
Unfortunately historically, our industry had not been the most 
receptive to new devices, and now when it was withdrawn, I was 
saddened. I was saddened because we don't have the process.
    I must tell you, that NIOSH about 9 months ago, asked an 
institution at the university that I work with, the National 
Technology Transfer Center, that Senator Byrd has supported, 
and asked them to help us look for improvements in the SCSR. 
And in addition to that, NASA has agreed for us to look for new 
technologies at that center.
    Senator Harkin. Let me ask you about this tracking device, 
where was it made--do you know where it's made? I think it's 
made in Australia; I'll just answer my own darn question. You 
say it's about $20?
    Mr. McAteer. Something to that effect.
    Senator Harkin. With this $20, you could locate any miner?
    Mr. McAteer. There's a transponder that's stationed 
throughout the mines, and then you can locate any individual, 
at any time.
    Senator Harkin. You know exactly where they are.
    Mr. McAteer. That's correct.
    Senator Harkin. $20. How many miners wear these, and take 
these underground with them?
    Mr. McAteer. Currently?
    Senator Harkin. Yeah.
    Mr. McAteer. I don't know. It's very, very small. Perhaps 
none in this country.
    Senator Harkin. Do you know Cecil?
    Mr. Roberts. If I had to give you an answer, I'd say none.
    Senator Harkin. $20, and in case a mine has two of the most 
important things, communication, location, and where you are, 
and how can we communicate? This tells you where you are. Now 
you've got this device here. Now this was the text messaging, 
is this the same device they used out in Utah, or do you know?
    Mr. McAteer. That's correct sir.
    Senator Harkin. This is what they used in Utah.
    Mr. McAteer. Exactly that device. Yes sir.
    Senator Harkin. It's got a battery, the message comes 
through right there, so you can talk to one another, and 
obviously I didn't know this, it also has a light too. 
Obviously, you have to have a light to be able to----
    Mr. McAteer. This is attached to the miners lamp, every 
miner carries one of those.
    Senator Harkin. The device is attached to the battery.
    Mr. McAteer. The advance is that part on top, which gives 
you ability to text message to the surface to the miner.
    Senator Harkin. Right here?
    Mr. McAteer. That's correct sir.
    Senator Harkin. So you can tell them what to do. So now 
you--with this, you know where they are. With this, you can 
tell them what to do?
    Mr. McAteer. That's right.
    Senator Harkin. Those are available. Now how many miners 
have this available to them?
    Mr. McAteer. We believe--we've checked with the 
manufacturer, and 14 mines in this country are using those.
    Senator Harkin. 14 mines are using these.
    Senator Byrd. Out of how many?
    Senator Harkin. Out of how many mines?
    Mr. McAteer. 15,000.
    Senator Harkin. 14 out of 15,000?
    Mr. McAteer. That's with metal and nonmetal, if you can't 
metal, nonmetal at the locations, as well as coal locations.
    Senator Harkin. You said it would cost about $100,000 a 
mine to hook that up, that sounded high to me, because I know 
about the antenna loops that go with it, and that seemed to be 
kind of high, just putting in an antenna.
    Mr. McAteer. That's the manufacturer--gave us that amount, 
and said that is the high end.
    Senator Harkin. That sounds like a high end to me.
    Mr. McAteer. It's not that much, but I'm suggesting to you 
that that would be the size of the Sago Mine, $100,000.
    Senator Harkin. That would be the loop, and buying these 
devices for every miner?
    Mr. McAteer. Yes sir, I'm sorry. That's the entire 
installment. That's what they told us.
    Senator Harkin. It seems like a small price to pay per 
mine.
    Mr. McAteer. Yes it is.
    Senator Harkin. How many miners at Sago? How many miners?
    Mr. McAteer. 150.
    Mr. Hatfield. Approximately 150.
    Senator Harkin. So, 150. You're talking about less than a 
$1,000, $750 per miner to be able to have communications, $20 
to tell us where they are. It seems to me, this is the 
investment we ought to be making. I can't believe in this day 
and age with the technology that we have, even with the oxygen 
breathing devices, we know we can develop oxygen breathing 
devices that are longer than an hour. Much longer than an hour. 
Why aren't we doing it?
    Mr. Roberts said a miner shouldn't have to get sick, or 
fear death just to go to work everyday. This is just--I don't 
know Mr. Chairman, it just seems to me that you hate to 
regulate everything, but doggone it, if they won't do it, we 
got to tell them to do it. It seems to me, this is something 
that every miner ought to have available, period. Thank you, 
Mr. Chairman.
    Senator Specter. Thank you, Senator Harkin. Senator DeWine.
    Senator DeWine. Thank you, Mr. Chairman. Mr. McAteer, you 
did describe in your written testimony how the technology for 
this self contained, self rescuers, which provides oxygen 
supply for an hour, was developed more than 50 years ago.
    Mr. McAteer. That's correct.
    Senator DeWine. Could you tell us a little bit, where is 
that technology now, and where should it be going, and what's 
the potential here, and maybe Mr. Roberts, you could comment as 
well. You talked a little bit about where it ought to be, and 
I'd like for you to comment--one of you, then the other, then I 
have a couple of other questions if you can do it real quick.
    Mr. McAteer. Yes sir. There are two types of technology, 
bottle oxygen and potassium super oxide--a device that you 
breathe into, and it gives you back oxygen. Those two 
technologies are advanced. We have them, we are examining under 
this NIOSH grant, the industry where those are, and what 
advances have been made. We think there has not been a large 
expenditure of funds on new technologies. We think that there 
are some advances that we can bring in, and we can add numbers 
of ours. I can't tell you what the outside limit of that would 
be, but we believe that we could give you longer periods of 
time, for less weight, which is one of the problems. The 
advances that we are putting in place in West Virginia today 
are to add devices, so that you can carry two or three, and get 
from one location to another.
    Senator DeWine. Mr. Roberts, do you want to comment about 
that, and also while I've got you, let me just say you're 
testimony was very concise, very good, very helpful. If you had 
to prioritize, what would you prioritize? Give me one, two, 
three, four.
    Mr. Roberts. The most immediate thing we could do, I 
believe, to give the miners the best opportunity to survive, 
and I think that is where we have to start, is to immediately 
put additional oxygen devices underground. That is point one.
    Point number two, and I don't want to speculate on what 
could've happened, but I think I heard Mr. Hatfield say 
something similar to this, I think experts looking at this will 
tell you if there had been an ability to talk to these miners, 
I probably would've told them, you're best opportunity for 
survival is to leave, and we could've even sent mine rescue 
teams toward them, as they walked outside. Perhaps, we would've 
had more miracles here today.
    So, I would start with the oxygen, I would go to the 
communications, and we have to Senator, begin right this minute 
to deal with this mine rescue team, because Mr. McAteer talked 
about 1995 when he brought all of the people together in this 
industry, I would submit to him, and to you, and to everyone 
else the problem is worse right now, than it was in 1995 when 
he knew then that we had a problem. The problem is much, much 
worse. We have difficulty finding enough mine rescue teams in 
these disasters to deal with these problems.
    Senator DeWine. Mr. Hatfield, let me ask you, did you have 
any difficulty in contacting the Federal Government for rescue 
assistance on the day of the accident, and how would you 
preliminarily assess their involvement in the rescue and 
recovery efforts?
    Mr. Hatfield. Senator, we did have some initial difficulty, 
because of the nature of the holiday season, in that our first 
calls to the various offices were going into a voicemail and 
the like, but there were messages on those voicemails that gave 
us alternate numbers to call, and so the process took a little 
longer than it might otherwise have. But, we were on the phone 
with the State regulators by I think around 7:50, and we had 
contacted MSHA by about 8:28.
    Senator DeWine. You lost a little time.
    Mr. Hatfield. We lost a little time, but they were 
responsive through cell phones and home numbers, because these 
are mining professionals, they want to be available. They 
always give contact numbers; it just took us a little while 
longer to get there.
    Senator DeWine. I wanted to make sure we understood what 
the situation was there. Mr. McAteer, where's the 
responsibility now? Where do we go in regard to the different 
technology that you've talked about? How does this get done? 
We've talked about several different technologies, how do we 
move forward now?
    Mr. McAteer. Senator DeWine, that's the problem. You put 
your finger on a problem. We have not spent the money, neither 
the industry, nor the Federal Government has spent the dollars 
necessary to advance the technology, and we need to do that. I 
think both the industry and the government, at a Federal and 
State level need to be doing that to see that that happens.
    This Australian device I mentioned grew out of research 
that was done after mine disasters in Australia, and that was 
paid for by the Australian Government. But I think we have this 
problem here, where we have a gap, and we don't have the kind 
of enhancements in technology that we should be doing, and it's 
not on a continuing basis. We might do it for one thing like 
the SCSR, but then there's never--we don't advance it, and we 
don't bring in technologies from NASA, the Department of 
Defense into the mining community.
    Robots for example, that can go on the moon, we've tried to 
introduce into mine rescue systems, because it's better to send 
a robot down, then your not so much concerned about the fires 
and the explosions. But we have only used that on one or two 
occasions, and we haven't made the enhancements, and it's 
expensive, but we haven't made the advancements that we need to 
be making in that area.
    Senator DeWine. I thank you. Thank you, Mr. Chairman.
    Senator Specter. Well, thank you very much gentlemen.
    Mr. Hatfield. Mr. Chairman, Ben Hatfield. Could I ask the 
indulgence of the committee to make a brief comment on the 
technology, because we certainly share that interest with you.
    Senator Specter. Go ahead Mr. Hatfield.
    Mr. Hatfield. The PED system Mr. McAteer has referred to, 
has certainly holds much promise. Indeed, we have that system 
installed at our Illinois Mine, the Viper Mine, and have seen 
some success with it. The couple of limitations that certainly 
we want to focus on in moving forward with this effort is that 
it's a one way communication. There are reliability issues. But 
first and foremost, with respect to the Sago incident, we have 
not found the technology yet, that will allow us to communicate 
through 300 feet of rock. When you don't have a line on sight, 
unless there's a cable or an antenna that goes underground, and 
there's the challenge. Because in an explosion--the nature of 
the Sago explosion, it obliterates everything in its path. 
Antennas and cables were immediately gone in an instant, and 
that is the challenge for the technology advancement efforts 
here to find ways, do not rely on things that are going to be 
damaged in an explosion, and that's something we certainly want 
you to understand. But our Viper Mine in Illinois is making use 
of this system, and I think it does hold significant promise.
    I would also like to clarify, and apologize for my earlier 
answer to you, not being as thorough as it should have been. 
One other step that International Coal Group has taken with 
respect to helping the Sago families is the establishment of 
the Sago Mine Fund. We made our own initial contribution of $2 
million, and that fund, through the generosity of many others 
across the country, has already $20 million, and that is 
exclusively dedicated for the benefit of the Sago victims, and 
we expect that fund to continue growing. Thank you.
    A letter from Senator Durbin asking about the safety of the 
Viper Coal mine, with my response follows:

                                      United States Senate,
                                   Washington, DC, January 8, 2006.

Bennett K. Hatfield, President and CEO, International Coal Group, 2000 
    Ashland Drive Ashland, KY.

    Dear Mr. Hatfield: I am writing to you to determine the safety 
condition of the Viper Coal Mine in Williamsville, Illinois that is 
owned by your International Coal Group (ICG). It has come to my 
attention that the Viper Mine is one of 12 coal mines that the ICG owns 
and operates, including the Sago Coal Mine in West Virginia where a 
mine accident recently killed 12 employees.
    I am deeply saddened by the tragic loss of life in West Virginia. 
Like many others, I also am disturbed by the egregious number of 
citations issued to the Sago Coal Mine in the year prior to the mine 
accident for ``serious and significant'' violations of Federal safety 
laws. Coal miners accept that their occupation is risky. But they have 
a right to expect that their employer will put safety ahead of other 
considerations, and the ICG has a moral obligation to ensure that 
unsafe practices are scrupulously identified, promptly corrected, and 
proactively avoided. The ICUs safety record at the Sago Mine raises 
questions about the company's commitment to safety. Moreover, Federal 
regulators appear to be unable to prevent companies from amassing large 
quantities of safety violations.
    According to Mine Safety and Health Administration records, the 
Viper Mine in IIlinois was cited for 42 ``serious and significant'' 
violations out of 124 total citations in 2005, which was the f rst full 
calendar year in which the ICG owned the mine. It is my hope that the 
Sago disaster was an isolated incident and does not reflect a pattern 
of mismanagement by the ICG and unsafe conditions at the Viper Mine.
    At a time like this, the burden is on the ICG and our nation's 
regulatory authorities to ensure that the ICG is properly and safely 
managing its mine in Illinois, Please provide to me and the people 
whose lives are affected by operations at the Viper Mine a safety 
status report on the Viper Mine. In addition, please include the safety 
record of the mine since the ICG purchased the operation and the extent 
of any steps the ICG has taken to address past safety violations and 
what the ICG plans to do to prevent them in the future.
    Thank you for your time and attention to this matter. I look 
forward to hearing from you within the next month.
            Sincerely,
                                         Richard J. Durbin,
                                                      U.S. Senator.

                            International Coal Group, Inc.,
                                     Ashland, KY, January 16, 2006.

Hon. Senator Richard J. Durbin,
U.S. Senate, 332 Dirksen Senate Office Building, Washington, D.C. 
        20510-1304.

Re: Your letter of January 8, 2006 regarding ICG Illinois, LLC's Viper 
Mine

    Dear Senator Durbin: Thank you for your concern for the safety of 
the employees at the ICG Illinois, LLC (``ICG Illinois'') Viper Mine. 
Although your interest in mine safety at Viper seems to have been 
prompted by the Sago mine disaster at an affiliated subsidiary of ICG 
Illinois' parent company, International Coal Group Inc. (ICG), I am 
pleased to have an opportunity to introduce you to ICG's and ICG 
Illinois' commitment to workplace safety.
    Misapprehension of the facts about ICG subsidiary safety records is 
understandable given the recent media-induced rush to judgment about 
the safety record at the Sago Mine, with no meaningful review of that 
record, or discussion of safety at ICG subsidiaries as a whole. The ICG 
family continues to grieve concerning the devastating Sago tragedy. Our 
thoughts and prayers remain with the families of the miners who lost 
their lives and to the single survivor, Randal McCloy, Jr., and his 
family.
    To give you a more balanced appreciation of ICG's overall safety 
record and commitment, I am attaching a one page summary of the 
numerous safety nominations and awards ICG subsidiaries received in 
2004 and 2005. (See Attachment A). I would especially like to bring 
your attention to the Sentinels of Safety Award won by ICG Knott 
County, LLC (ICG Knott County) in 2004. This is the premier mine safety 
award in the nation. The Sentinels of Safety Award is an incredibly 
prestigious safety award with an 85 year lineage, originating under the 
presidency of Herbert Hoover. Simply put, ICG Knott County's Clean 
Energy Underground Mine was the safest underground mine in the large 
mine category in the entire U.S. during 2004. The Clean Energy Mine had 
128,525 manhours of operation without a single loss time accident in 
2004. Needless to say, ICG is extremely proud of Clean Energy's safety 
accomplishments, and its safety performance is a model ICG strives to 
emulate at all of its subsidiary operations.
    ICG Illinois purchased the Viper Mine from the bankrupt and now 
defunct Horizon Natural Resources Company on September 30, 2004. Over 
the past fifteen months ICG Illinois has invested over eleven million 
dollars ($11,000,000) of capital in the Viper Mine, including upgrades 
to the mine's carbon monoxide (CO) monitoring system, the underground 
communications system, a new rock-dust system, addition of a second 
mine master supply system for both production and fire fighting, and 
upgrades to other underground equipment.
    The Viper Mine is well known for its immediate response to both 
State and Federal regulatory directives during inspections. In 2005 
MSHA conducted 257 inspection shifts at the Viper Mine and the State of 
Illinois over 40 inspection shifts. The MSHA district inspectors and 
State inspectors are highly regarded for their expertise in the mining 
industry. The Viper mine enjoys a positive, constructive relationship 
with both agencies.
    Viper's employees are well trained, well educated coal miners who 
understand safety. Each employee is aware that, if necessary, they are 
authorized to shut down all or any part of the mine if an unsafe 
condition is identified. Employees may also address concerns 
anonymously. An employee can also contact the Illinois Department of 
Natural Resources (IDNR) directly or call MSHA's hotline. International 
Coal Group also has provided every employee a card with a hotline to 
call to report safety concerns, environmental issues, harassment, 
discrimination or other concerns.
    In fact, Viper is proud to host a fully certified ``Mine Rescue 
Team''. As you may know, the Viper Mine Rescue Team was one of the 
teams that was dispatched to the Sago site and, while putting their own 
lives at risk, proudly took part in the attempt to rescue the trapped 
miners.
    Regarding your concerns about safety citations, please refer to the 
tabulated data below. This list consists of safety data for nine of the 
older, larger Illinois underground coal mines for 2005. As you can 
clearly see, the Viper mine ranks at the top of its peer group.

------------------------------------------------------------------------
                                                              Number of
                                                              citations
               Mine                 Employees      Total     significant
                                                 citations        and
                                                             substantial
------------------------------------------------------------------------
Viper............................          243          124           42
Mine A...........................          341          138           52
Mine B...........................          229          167           81
Mine C...........................          151          184           50
Mine D...........................          277          173           42
Mine E...........................          221          227          105
Mine F...........................          383          397          112
Mine G...........................          253          777          277
Mine H...........................          867        1,456          389
------------------------------------------------------------------------

    A serious infraction that would jeopardize miner's lives or could 
cause a disaster results in a closure order by State or Federal 
authorities which remains in place until the infraction is corrected. 
The larger a mine becomes, the more area there is to inspect for safe 
working conditions, with size accordingly increasing the number of 
inspection shifts per year. For example the Viper Mine has four 
operating sections and over ten miles of conveyor belts transporting 
the coal out of the mine, thus resulting in a rather high number of 
inspection shifts.
    In December of last year, I presented my vision of ICG to our 
operating subsidiary management. A critical component to this vision 
includes being the employer of choice by providing employees a safe, 
secure, competitive paying job where they are treated with respect.
    I extend an open invitation to you to visit the ICG Illinois Viper 
Mine to see first hand ICG's and ICG Illinois' commitment to the safety 
of its employees. I am confident that your doing so will convince you, 
even more than what I have shared with you in this letter, that the 
attacks upon ICG's safety commitment are unwarranted.
    ICG Illinois, like all other ICG subsidiaries, will continue to 
operate the Viper Mine with a meaningful ``Safety First'' philosophy 
and practice. It is our principal priority for protecting our most 
valuable resource--our employees.
    Thank you for the opportunity to share with you ICG's safety 
accomplishments and its safety commitment. Should you have any further 
questions, please do not hesitate to contact me.
            Very truly yours,
                                       Bennett K. Hatfield,
                                                 President and CEO.

                                                  SAFETY AWARDS
----------------------------------------------------------------------------------------------------------------
                 Year                       Subsidiary Unit                     Award Description
----------------------------------------------------------------------------------------------------------------
2004.................................  ICG Addcar Systems, LLC.  Bridger Highwall Miner Operation--Perfect
                                                                  Safety Record for 59,759 Man-hours with no
                                                                  LTA's in 2004, presented by the State of
                                                                  Wyoming.
2004.................................  ICG Addcar Systems, LLC.  Bridger Highwall Miner Operation--Nominated for
                                                                  a Sentinels of Safety Award for having no
                                                                  LTA's in 2004.
2004.................................  ICG Eastern, LLC........  Birch River Preparation Plant--Safety
                                                                  Achievement for 2,257 Days Without an LTA,
                                                                  presented by the State of West Virginia.
2004.................................  ICG Hazard, LLC.........  County Line Mine--Nominated for a Sentinels of
                                                                  Safety Award for having no LTA's in 2004.
2004.................................  ICG Hazard, LLC.........  Rowdy Gap Mine--Nominated for a Sentinels of
                                                                  Safety Award for having no LTA's in 2004.
2004.................................  ICG Hazard, LLC.........  Thunder Ridge Surface Mine--Surface Safety
                                                                  Award, presented by the Kentucky Coal
                                                                  Association.
2004.................................  ICG Hazard, LLC.........  Tip Top Mine--Nominated for a Sentinels of
                                                                  Safety Award for having no LTA's in 2004.
2004.................................  ICG Knott County, LLC...  Clean Energy Mine received the Kentucky Coal
                                                                  Association's Surface Safety Award for the
                                                                  best underground mine in Hazard District.
2004.................................  ICG Knott County, LLC...  Clean Energy Underground Mine--The 2004
                                                                  Sentinels of Safety Award for the Safest
                                                                  Underground Mine in the United States,
                                                                  presented by the National Mining Association
                                                                  and MSHA.
2004.................................  ICG Knott County, LLC...  Classic Underground Mine--Best Safety Record in
                                                                  Martin County, presented by the Kentucky
                                                                  Office of Mine Safety.
2004.................................  Sentinel................  Sentinel Preparation Plant--Best Safety Record
                                                                  in 2004, presented by West Virginia Holmes
                                                                  Safety Association.
2005.................................  Vindex Energy...........  Vindex Douglas Surface Operations--Safest Mine
                                                                  in the State of Maryland for 2005, presented
                                                                  by the Maryland Coal Association.
----------------------------------------------------------------------------------------------------------------

    Senator Specter. Well thank you, Mr. Hatfield. I'm a little 
disappointed and a little surprised to note that Mr. Dye and 
Mr. McKinney have departed the premises, so we're going to ask 
Mr. Friend and Mr. Clair to supplement for the record on a 
number of questions. We note, that consistent with what Mr. 
Roberts testified to, when there was a fine imposed on the 
Alabama incident where 13 miners were killed in December of 
2001, a fine was imposed of $435,000, and that an 
administrative law judge reduced it to $3,000, and a number of 
us wrote a letter to the Secretary of Labor, dated November 22, 
2005, protesting that, and asking that an appeal be taken, and 
without objection, that will be made a part of the record.
    [The letter follows:]

                              United States Senate,
       Committee on Health, Education, Labor, and Pensions,
                                 Washington, DC, November 22, 2005.

Hon. Elaine Chao, Secretary, U.S. Department of Labor, Washington, DC.

    Dear Secretary Chao: We are writing to express our deep concern 
about the recent decision of the Mine Safety and Health Review 
Commission to drastically reduce the fines sought by the department 
against Jim Walter Resources, Inc., in the case involving the deaths of 
thirteen miners at Blue Creek No. 5 Mine in Brookwood, Alabama.
    We urge you to appeal that decision. The men and women working in 
America's mines do so in an extremely hazardous environment and they 
are entitled to vigorous enforcement of the Federal workplace safety 
laws. In the tragic explosion at Blue Creek, our Nation's deadliest 
mine disaster in over 20 years, the Mine Safety and Health 
Administration sought fines of $435,000. Fines of this level were 
clearly justified due to Walter Resources' long history of safety 
violations which led to perilous, and ultimately fatal, workplace 
conditions.
    Incredibly, however, on November 1, an Administrative Law Judge 
reduced these fines from $435,000 to a mere $3,000--a decision that 
harms workers and erodes MSHA's authority to protect the health and 
safety of America's miners.
    President Bush has clearly recognized the importance of fines in 
deterring safety violations and enforcing worker safety laws with his 
budget request for fiscal year 2005 to increase the monetary penalties 
imposed by MSHA. We urge you to appeal the unfortunate decision in the 
Blue Creek Mine case, and we look forward to hearing from you about 
your action.
            Sincerely,
                                   Arlen Specter,
                                   John D. Rockefeller IV,
                                   Edward M. Kennedy, and
                                   Robert C. Byrd.

    Senator Specter. The subcommittee would like to know what 
has happened on that. Other questions, which we had intended to 
press on Mr. Dye, where reportedly 16 failures over the last 2 
years at Sago, were seriousness enough to be designated as 
``unwarrantable failures'' which has been defined as 
``aggravated conduct constituting more than ordinary 
negligence'', and we would like to have a specification for 
what they were, and whether there was any cause of the effect. 
We've also noted the reduction in the number of criminal 
prosecutions recommended from 38 in the year 2000, to 12 in the 
year 2004. We would like to have some specification as to what 
the recommendations there were on criminal prosecutions.
    One other question that we have for Mr. Dye, we note that 
he's the Acting Assistant Secretary of Labor, and of course the 
Assistant Secretary is a confirmable position, confirmable by 
the Senate, and I'd be interested to know if Mr. Dye has any 
aspirations to move from Acting Secretary to Secretary of the 
Department. We will be pursuing these issues further. There 
have been a number of suggestions made as to possible 
legislative action. The act, we're informed, requires the 
company to be present at the time that witnesses are 
interviewed. My own instinct is that they ought to be 
interviewed privately, to have the maximum freedom in answering 
questions.
    May the record show a couple of nods from the panel, Mr. 
McAteer and Mr. Roberts. We're going to explore the issue of 
increased penalties.
    Senator Byrd. Were the nods, Mr. Chairman, affirmative 
nods, or negative?
    Senator Specter. Affirmative nods. Thank you, Senator Byrd. 
Do you think I would've commented if they'd been negative nods?
    You didn't nod Mr. Hatfield, what do you think about having 
a requirement that the company representative be present at the 
time witnesses are questioned? Isn't that a little 
intimidating?
    Mr. Hatfield. Well I don't think so, because again, we've 
been very open with our people about what we know about the 
investigation, and have encouraged them to speak out.
    Senator Specter. There's nothing to stop the company from 
interviewing them. There's nothing to stop anybody from 
interviewing anybody. Nobody has to answer a question, unless 
they're under a subpoena. You don't even have to answer a 
question if the subcommittee asks you to stay, you could just 
walk right out.
    Anybody can ask anybody about everything, but nobody has to 
answer, unless there is a subpoena, because with subpoena you 
don't have to answer either if you take the privilege against 
self incrimination, plus your granted immunity. But why should 
an employee have to have a company official present? The 
possibility of retaliation, I know your company wouldn't, but 
some company might. What do you think?
    Mr. Hatfield. I can't speak with any knowledge as to what 
brought the rules to the position they're in, but I can speak 
on behalf of my company, that we certainly have no fear of 
leaving the interviews directly between the government 
representatives, and our people. We've encouraged them to be 
very open and honest with everything we know, and we certainly 
willing to absent ourselves from all the interviews, so long as 
the people interviewing our people, are truly the Mine Safety 
and Health Administration group, and the West Virginia 
Department of Miner Health and Safety.
    Senator Specter. A second item, we're going to look at 
increased penalties. I'll give you a chance to reply to that 
Mr. Hatfield if you want to. Do you think it would be time to 
increase the penalties, perhaps commensurate with the increase 
coal prices?
    Mr. Hatfield. I think it's certainly a reasonable area of 
review. We have no objection whatsoever to review of the 
penalties, because I suspect they've not been visited in some 
time. But I would echo a comment that made by Mr. Dye earlier, 
that real penalty is the threat of shutting down a production 
section because that gets far beyond the fine level.
    Senator Specter. I was just about to come to that. Do you 
endorse that?
    Mr. Hatfield. I believe that is an appropriate tool for 
enforcers and regulators to use, and we certainly saw it used 
at the Sago Mine, and other mines. Yes sir.
    Senator Specter. How about the idea of user fees on the 
mines to pay for all of these investigations, and rescue teams, 
and all of the other safety procedures which may be an order? 
They really go to the benefit of the mine companies that earn 
the profit. What do you think about a user fee?
    Mr. Hatfield. I think it's certainly an area that is worth 
looking into. I don't know many of the facts, or the premises, 
but certainly it's an area that's fair for discussion, and 
advancement.
    Senator Specter. How about Senator Harkin's photo op with 
this orange mechanism, do you think they ought to be required?
    Mr. Hatfield. I think that's remarkable opportunity.
    Senator Specter. I take that to be a yes. But one other 
thing we're going to do, is ask the GAO to undertake a follow-
up study. At the request of the subcommittee, and others, we 
made that request of GAO in November 2002, and they filed a 
report in September 2003, and we're going to request a follow-
up study of the September 2003 mine safety report focusing on 
underground coal mines. Anybody on the panel, do the Senators 
have any further questions they'd like to ask at this time?
    Senator Byrd. Yes, Mr. Chairman, if you will. Mine safety 
is a moral imperative. These miners are not to be considered 
expendable. What do we need to do right now to ensure a tough 
enforcement? Now there's been some talk about penalties, what 
do you think of assessing penalties as a percentage of the 
profits? What do you think of companies paying fines up front, 
so that penalties cannot be reduced by negotiation? Does 
anybody want to volunteer to respond? Mr. Roberts.
    Mr. Roberts. I think we have a long history of support for 
the toughest enforcement possible. It's our view that complying 
with the law does not interfere with profits whatsoever. If 
you're running a safe mine, you have nothing to fear. From 
increased fines, we also would encourage that these fines could 
not be dramatically reduced on appeal, as has happened time, 
and time, and time again. I think that point that was made by 
Senator Specter, with respect to shutting down mines, I think 
that's the ultimate weapon in the hands of MSHA Inspectors, and 
I think that Congress should mandate that the law that was 
passed 1969, and was amended in 1977, be complied with.
    Senator Byrd. Anybody else want to comment?
    Mr. McAteer. If I might Senator. We have a good law on the 
books. It is a law that has an enforcement aspect of it that is 
quite strong. In recent times, that enforcement aspect has not 
been emphasized as much as has cooperation. Now there's nothing 
wrong with cooperation, and I'm in favor of it, and I think it 
should be, but to do that and to diminish enforcement while 
cooperation takes precedence, I think is an unfortunate turn of 
events, and is not what this Congress, when it passed the law 
in 1969, and amended in 1977, had in mind.
    Senator Byrd. Mr. Chairman, I simply want to say again, I 
think this has been helpful. Where is the point that we should 
attack? Should it be the Federal agency that oversees safety? 
Is that where the problem is? I think there ought to be 
stricter enforcement. There ought to be assurance that the 
agency will do its best to do its duty.
    It seems to me that there is a kind of a mental attitude 
here, and mind set that yes, it is dangerous, mining is 
dangerous, but these violations will pay the fines, but maybe 
they can be negotiated downward, we'll just keep on doing it. I 
think there ought to be tough enforcement with this idea in the 
background that we're going to close you down. Why shouldn't 
this mine have been closed down? Does anyone want to respond?
    Mr. Roberts. It is the question, should the Sago Mine--I 
think in trying to be as fair as I can be, and as objective as 
I can be, I think it would be until a complete investigation is 
completed, perhaps we don't know the answer to that, but I do 
support Senator, as I have just previously said, that if there 
is a pattern here, which is part of the law that was alluded to 
previously, I think that is the ultimate penalty, and it should 
be carried forward.
    Senator Byrd. Mr. McAteer, did you want to respond to Mr. 
Hatfield's remarks regarding the reliability of safety devices?
    Mr. McAteer. Yes, I do Senator, and I have to take 
exception to his remarks. These devices have proved to be 
reliable. They are not perfect. It would be, in a perfect 
world, better to have a device that could communicate both 
ways. This device communicates one way, it is a step. The way 
we improve technology in this country, is not--the perfect is 
the enemy of the good. We have to put in what we have, and get 
them there, and then make that very next step, and 
unfortunately Mr. Hatfield's comments are suggestive of what 
we've worked with in the past. Some in the industry have had an 
attitude of suggesting that they won't take that step. They'll 
wait until the perfect is in place, and that day never comes. 
That's the unfortunate part of our situation at the present 
time.
    Senator Byrd. Mr. Hatfield, did you want to take 30 seconds 
for any response to what has been said?
    Mr. Hatfield. I certainly agree with Mr. McAteer's point 
that we can't wait until the solution is perfect to implement 
it. I don't think any of us around this table believe we can 
wait for perfection. We've got to make our coal miners safer 
today, and that's got to be the utmost priority. But, I believe 
we continue to have to deal with the challenge of how do you 
keep an explosion from blasting away the antennas, and the 
cables, and everything else that we use to support these 
systems, and still maintain that line of communication through 
300 feet of rock. I'm only pointing out there are technical 
issues that we need to move forward with, and that's the first 
one to tackle, and we certainly want to tackle it.
    Senator Byrd. Thank you.
    Mr. McAteer. If I could on that question, there is no 
question that part of the mine is destroyed in an explosion, 
and at the time of an explosion, part of the mine is not 
destroyed at the time of the explosion. The concept behind the 
internet is that we have multiple nodes, so if part of it goes 
down, all of it doesn't go down. That's the concept that we 
need to adopt in the mines, multiple redundant systems that can 
be overlaid. In the case of the transponder, and this device 
here, the mine company--the Australian mine company that puts 
this together has suggested, that if you lose communication 
they can take a loop antenna, and put it on the surface, and 
then you can reignite the ability to communicate underground 
from the surface, and they've done that in other countries. 
Redundancy.
    Senator Specter. Thank you very much, Senator Byrd. Senator 
Harkin, did you have anything else?
    Senator Harkin. No. Just when you were speaking Mr. 
Hatfield, I thought about redundancy. That is what you do, you 
build in redundancy, and there may be some technologies I'm not 
aware of. I'm not aware of this surface device you're talking 
about, but I'm sure there are new technologies out there of 
which I'm totally unaware of that help.
    Mr. Hatfield. I thoroughly agree. Redundancy can be part of 
the solution.
    Senator Specter. Redundancy is something Senator's know a 
lot about.
    Well, thank you very much. Mr. Hatfield, Mr. McAteer, Mr. 
Hamilton, Mr. Roberts, Mr. Watzman, thank you very much for 
coming in today, and we will be pursuing the matter, I know all 
of you will be too.

                     ADDITIONAL PREPARED STATEMENT

    The subcommittee has have received a prepared statement 
from the National Stone Sand & Gravel Association that will be 
included in the record at this point.
    [The statement follows:]

   Prepared Statement of Charles E. Hawkins, III, CAE Executive Vice 
      President and COO, National Stone Sand & Gravel Association

    Mr. Chairman, the National Stone Sand & Gravel Association (NSSGA) 
appreciates the opportunity to submit a statement for the record of 
this oversight hearing on the Mine Safety and Health Administration 
(MSHA) and its regulatory and enforcement programs.
    Based near the nation's capital, NSSGA is the world's largest 
mining association by product volume. Its member companies represent 
more than 90 percent of the crushed stone and 70 percent of the sand 
and gravel produced annually in the United States and approximately 
115,000 working men and women in the aggregates industry. Sale of 
natural aggregates (crushed stone, sand and gravel) generates nearly 
$38 billion annually for the U.S. economy. The estimated output of 
aggregates produced in the first half of 2005 was 1.3 billion metric 
tons, a four percent increase over the same period in 2004 (2.85 b MT). 
According to the U.S. Geological Survey, the significant increases in 
aggregates production were due to the increase in construction 
activity, which has risen every year for the past decade. Construction 
spending amounted to $617.9 billion during the first half of 2005, a 
nine percent increase over the same period in 2004.
    Aggregates are used in nearly all residential, commercial and 
industrial building construction and in most public works projects, 
such as roads, highways, bridges, railroad beds, dams, airports, water 
and sewage treatment plants and tunnels. While the American public pays 
little attention to these raw natural materials, they go into the 
manufacture of asphalt, concrete, glass, paper, paint, pharmaceuticals, 
cosmetics, chewing gum, household cleansers, and many other consumer 
goods.
    The events in the Sago Mine and Aracoma Coal Alma No. 1 Mine 
disasters are tragic and the loss of even one life, let alone 14 lives, 
is devastating. Nevertheless, the safety record of the mining industry, 
and the aggregates industry in particular, has improved due to a 
heightened level of effort invested by the industry to sustain an 
improved performance. The improvement in the aggregates industry safety 
record is attributable to a combination of more effective safety and 
health programs developed and implemented by the industry over the past 
decade in response to increased Mine Safety and Health Administration 
regulation of the mining industry and heightened enforcement.
    The first priority for the aggregates industry is and will continue 
to be the safety and health of its miners. The industry recognizes that 
its employees are its most valuable asset, an asset that must be 
protected for the well being of the industry now and in the future. As 
the workforce ages, it has become increasingly difficult to recruit new 
miners to the industry. Maintaining an excellent safety record through 
the implementation of effective safety and health programs is 
considered a critical element for attracting and keeping a highly 
skilled workforce. Members of the National Stone, Sand & Gravel 
Association have developed and agreed to a set of guiding principles to 
assist member companies in their efforts to understand the importance 
of safety to their individual organizations as well as to the industry 
as a whole. In addition, a safety pledge was developed in 2002 
incorporating the safety guiding principles. More than 90 percent of 
the operations of NSSGA member companies are now covered by this 
pledge, signifying the importance of safety and a commitment toward 
ensuring the safety and health of all their employees.
    Recent news articles have ascribed some of the responsibility for 
the Sago incident to the cooperative alliances MSHA has signed with the 
industries it regulates, implying an inappropriately close 
relationship. We would argue the opposite. The NSSGA and MSHA 
formalized the first such alliance in 2002, setting forth a cooperative 
agreement to develop programs and tools for the improvement of safety 
and health in the aggregates industry. The resulting reduced incidence 
rates speak for themselves.
    It should also be noted that MSHA has similar alliances with labor 
organizations, including the International Association of Bridge, 
Structural, Ornamental and Reinforcing Iron Workers and the 
International Union of Operating Engineers. Important alliances also 
exist with the National Safety Council and the American Society of 
Safety Engineers. Through these alliances, MSHA has been able to 
enhance its mission of protecting worker safety and health.
    Another collaborative effort resulted in the MSHA Part 46 
``Training and Retraining of Miners'' regulation in 2000. This 
excellent regulation ensures every miner knows and understands how to 
perform their job safely by covering the important safety and health 
information prior to starting work and annually thereafter. This 
regulation was developed collaboratively, with input from both labor 
and industry groups, guaranteeing support of the rule by all involved 
stakeholders and assuring their commitment to the ultimate goal of 
injury reduction. The Coalition for Effective Miner Training included 
many industry groups working in a joint industry/labor arrangement in 
conjunction with MSHA to develop an effective standard for the 
aggregates industry. The Part 46 regulation resulted from this effort.
    In another example, the NSSGA and MSHA developed a cooperative 
workplace-based sampling training program of noise and dust monitoring 
workshops. A partnership agreement was signed and the training workshop 
program launched on December 1, 1997. These workshops have been given 
to industry representatives using training specialists from the Mine 
Safety Academy every year since 1997. These workshops have won two 
awards from Innovations in American Government for this joint venture 
aimed ay reducing hearing loss and silicosis through a program of 
recognition, evaluation and control of workplace hazards.
    The NSSGA/MSHA alliance was further enhanced by an ad hoc coalition 
consisting of the U.S. aggregates industry (NSSGA and MSHA) and the 
quarrying industry (Health & Safety Executive and the Quarry Products 
Association) in the United Kingdom (England). This informal alliance 
was developed to share best practices between the countries in a 
similar industry.
    Based on the sharing of information about successful programs in 
the UK, the NSSGA/MSHA Alliance has moved forward with joint efforts to 
implement programs that will further improve the safety and health of 
U.S. aggregates miners. The alliance first assembled a Data Mining Task 
Force to review the incident data (not fatalities) with the hope of 
elucidating specific areas where efforts could be targeted to reduce 
injuries. It is this focus on incidents, rather than the focus on 
fatalities, that offers the best chance of improving the safety 
performance and at the same time reducing fatalities.
    Simultaneously, the alliance began working on a model safety and 
health program to take the best of industry and develop a model that 
could be used by both small and large aggregate producers to develop a 
safety management system. This resulted in the publication in December 
2005 of the ``Core Principles of a Safety Program'' by the Alliance. It 
is available free on the MSHA and NSSGA websites.
    At present, the Alliance is working on promoting safety and health 
through the publication of ``rip & share'' safety tools in the 
bimonthly association magazine and articles on timely safety topics for 
the industry to use in improving their safety programs. MSHA and NSSGA 
member company representatives jointly develop these tools. The 
cooperative relationship has made great strides toward improving the 
safety of the aggregates industry.
    You can see this clearly using the data required to be submitted by 
mine operators on injuries/illnesses and manhours. The attached chart 
``Comparison of Aggregate Industry Workhours vs. Incident Rates'' shows 
that even with an increasing number of hours worked at aggregates 
producers' sites there has been a significant reduction in the total 
incidence rate in the industry. The second chart ``Aggregate Industry 
Incident Rates 1989-2004'' shows this data broken down by aggregates 
industry sector. More progress has been made since 2002 through the 
cooperative efforts of the NSSGA/MSHA Alliance.
    In no way does the NSSGA/MSHA Alliance interfere with the 
compliance program of the agency. MSHA has an important role in 
ensuring that safety at aggregates mines and quarries maintain 
standards that protect employees. The MSHA enforcement program operates 
independently of any of the cooperative industry alliances. The Mine 
Safety Act, unlike any other safety agency, requires complete 
inspections of every mine property 2 or 4 times per year depending on 
whether it is surface or underground, respectively.
    The mining industry is more heavily regulated and inspected than 
general industry covered under the Occupational Safety and Health 
Administration regulations. It is important that caution be exercised 
before rushing to impose more regulations on the mining sector. Careful 
study of the programs in place must be made and effective enforcement 
ensured.
    NSSGA believes that the cooperative relationship the aggregates 
industry has developed with MSHA has led to increased safety for 
aggregate industry employees. We believe that these relationships 
rather than being discouraged should be encouraged. They are especially 
helpful to the small- and medium-sized companies that are unable to 
afford a staff safety professional by providing the mechanisms 
necessary for continuous improvement to the safety and health of 
aggregate workers.
    NSSGA appreciates the opportunity to provide comments to this very 
important hearing.

                     ADDITIONAL COMMITTEE QUESTIONS

    Senator Specter. There will be some additional questions 
which will be submitted for your response in the record.
    [The following questions were not asked at the hearing, but 
were submitted to the witnesses for response subsequent to the 
hearing:]

                  Questions Submitted to David G. Dye
              Questions Submitted by Senator Arlen Specter

    Question. Has the Department made a decision whether to appeal 
citations invalidated and penalties reduced in the Jim Walters 
Resources Case?
    Answer. On January 17, 2006, Solicitor of Labor Howard Radzely 
responded to your letter of November 22, 2005, suggesting that the 
Department should appeal a decision of a Federal Mine Safety and Health 
Review Commission (FMSHRC) Administrative Law Judge dismissing some of 
the citations against Jim Walters Resources in connection with the 
September 2001 explosion at the JWR#5 Mine. The Solicitor's letter 
explains that on December 1, 2005, MSHA appealed the decision. A copy 
of that letter, which explains the specific issues appealed, is 
attached.
    Question. Please provide me documentation of the 16 unwarrantable 
failure orders the agency issued to Sago Mine in 2005, including 
specification of what conditions caused MSHA inspectors to issue these 
orders, and whether any of these conditions that led to the issuance of 
these orders led to the accident on January 2, 2006.
    Answer. The table below offers a description of each of the 16 
unwarrantable failure orders issued to Sago Mine in 2005. It is 
important to remember that the investigation of the Sago Mine accident 
is ongoing, and the cause of the accident is at this time unknown. In 
addition, all of these conditions were fully corrected before normal 
mining activity was allowed to resume in the affected area.

----------------------------------------------------------------------------------------------------------------
                                        Regulation and type
      Violation number and Date               action                               Condition
----------------------------------------------------------------------------------------------------------------
7097825--5/20/05....................  75.400--104 D-1         Loose coal permitted to accumulate at the No. 3
                                       Citation.               Belt Drive and Take-up. Condition was obvious and
                                                               extensive.
7097827--5/20/05....................  75.360(a)(1)--104 D-1   Inadequate examination of the area cited in the D-
                                       Order.                  1 citation. The mine examiner failed to identify
                                                               and record this hazard.
7097835--6/02/05....................  75.340a (1)(i)--104 D-  Battery charging station was located in the
                                       1 Order.                primary escapeway. The air current ventilating
                                                               the charger was coursed directly to the two
                                                               working sections instead of to the return entry.
                                                               Repetitive citation and the conditions were
                                                               obvious.
7097941--7/12/05....................  75.203(b)--104 D-2      The mining methods used exposed the miners to
                                       Order.                  hazards related to reduced pillar size. The No. 5
                                                               entry was driven off sites to the point that in
                                                               60 feet of face advancement the pillar width was
                                                               reduced to 10 feet. No. 9 entry off sites and cut
                                                               into the No. 8 entry in only 56 feet of
                                                               development. Management failed to provide a sight
                                                               line and exposed the worker to the hazard of
                                                               overriding forces.
7098153--8/16/05....................  75.202(a)--104 D-2      Mine roof and ribs were not controlled in the
                                       Order.                  sections where the miners work or travel. There
                                                               were two prior serious accidents. The conditions
                                                               were obvious and extensive.
7098154--8/16/05....................  75.380(d)(1)--104 D-2   Primary intake escapeway from the pitmouth to the
                                       Order.                  No. 3 Belt Head had loose hazardous roof for 50
                                                               feet and loose hanging roof through out the area.
                                                               Also, the travelways were obstructed. This was
                                                               obvious and extensive and prevented safe passage.
7098158--8/16/05....................  75.364(b)(1)--104 D-2   Inadequate examination of the primary escapeway
                                       Order.                  for the above order. Management failed to
                                                               identify and record obvious and extensive
                                                               hazardous conditions in the weekly examination
                                                               book.
7098159--8/16/05....................  75.340(a)(1)(i)--104 D- Battery charging station was located in the
                                       2 Order.                primary escapeway. The air current ventilating
                                                               the charger was coursed directly to the two
                                                               working sections instead of to the return entry.
                                                               Repetitive citation and the conditions were
                                                               obvious in an area that required a pre-shift
                                                               examination to be conducted.
4890534--9/12/05....................  75.400--104 D-2 Order.  Accumulations of loose dry coal were permitted to
                                                               accumulate around and under the No.6 Belt Drive
                                                               and Take-up. Extensive and obvious.
4890535--9/12/05....................  75.400--104 D-2 Order.  Accumulations of loose dry coal were permitted to
                                                               accumulate around and under the No. 5 Belt Drive
                                                               and Take-up. Extensive obvious, and repetitive.
4890536--9/12/05....................  74.400--104 D-2 Order.  Accumulations of loose dry coal were permitted to
                                                               accumulate around and under the No. 4 Belt Drive
                                                               and Take-up. Extensive, obvious, and repetitive.
4890537--9/12/05....................  75.1403--104 D-2 Order  Walkway along the Nos. 6, 5, and 4 Belt Conveyors
                                                               Belt Drives and Takes-ups were obstructed with
                                                               deep mud and water. This hazard was obvious,
                                                               extensive, and repetitive.
4890539--9/12/05....................  75.360--104 D-2 Order.  An inadequate pre-shift examination was conducted
                                                               as identified in the previous 4 orders. Examiner
                                                               failed to identify the obvious and extensive
                                                               hazards and record them in the examination book.
7098364--9/18/05....................  75.330(b)(1)--104 D-2   Line curtains used to ventilate the Number 7 and 8
                                       Order.                  entries were rolled, thereby interrupting the
                                                               face ventilation. No movement of air could be
                                                               detected with a smoke tube and 0.5 percent
                                                               methane had accumulated. This condition was
                                                               obvious. Management was present and allowed this
                                                               condition to exist.
7149865--11/8/05....................  75.360(f)--104 D-2      The MSHA inspector issued 6 S&S citations and none
                                       Order.                  of the hazards were identified and recorded in
                                                               the pre-shift examination book. All of these
                                                               citations were in the area that was examined by
                                                               mine management.
7098644--12/14/05...................  75.400--104 D-2 Order.  Obvious and extensive accumulations of loose coal,
                                                               coal fines, and float dust were permitted to
                                                               accumulate in the 006 MMU. This was a repetitive
                                                               condition that was obvious and extensive.
----------------------------------------------------------------------------------------------------------------

    Question. I note that recommendations by MSHA for criminal 
prosecutions had declined from 38 in 2000 to 12 in 2004; we would like 
some specification as to what recommendations there [are] on criminal 
prosecutions.
    Answer. MSHA is unaware of the source of information indicating 
that the agency made 38 criminal referrals to the Department of Justice 
in 2000, or that the agency made 12 such referrals in 2004. According 
to the agency's information, the following criminal referrals have been 
made since 1998, along with the most recent information regarding the 
disposition of those referrals by the Department of Justice. Please 
note that the Office of the United States Attorney to whom MSHA makes 
these referrals may take several years to decide how to proceed on a 
particular referral.
    Please also note that the number of referrals in any given year 
will vary due to factors beyond the agency's control.

--------------------------------------------------------------------------------------------------------------------------------------------------------
                                                                                                                                 Percent
                          Year                             Referrals   Declined       Criminal         Open       Percent       criminal        Percent
                                                                                  dispositions \1\     cases     declined   dispositions \1\     open
--------------------------------------------------------------------------------------------------------------------------------------------------------
1998....................................................         8           4              4       ..........        50.0           50.0     ..........
1999....................................................        18          13              5       ..........        72.2           27.8     ..........
2000....................................................        18           8             10       ..........        44.4           55.6     ..........
Avg 1998-2000...........................................        14.7         8.3            6.3     ..........        55.6           44.4     ..........
2001....................................................         9           4              5       ..........        44.4           55.6     ..........
2002....................................................         8           2              4              2          25.0           50.0           25.0
2003....................................................         3           1              1              1          33.3           33.3           33.3
2004....................................................         3           1              1              1          33.3           33.3           33.3
2005....................................................         4    ..........            1              3    ..........           25.0           75.0
Avg 2001-2005...........................................         5.4         1.6            2.4            1.8        27.2           39.4           33.3 
--------------------------------------------------------------------------------------------------------------------------------------------------------
\1\ Includes pleas, convictions, acquittals, and pre-trial diversions (diverted out of the criminal system).

    Question. Is the Acting Assistant Secretary, David Dye, seeking 
appointment as permanent administrator of MSHA?
    Answer. On September 15, 2005, the President nominated Richard E. 
Stickler of West Virginia to be Assistant Secretary of Labor for Mine 
Safety and Health.
                                 ______
                                 
            Questions Submitted by Senator Richard C. Shelby

    Question. As we all know the work of a coal miner is desperately 
important to the life of our economy. But the job of the coal miner is 
also extremely dangerous and requires that measures are taken and 
standards are in place and that those safety standards are enforced. In 
light of the Sago mine disaster, is MSHA properly staffed and funded to 
carry out its mission to protect the safety of America's miners?
    Answer. Yes. Congress has appropriated sufficient funding for MSHA 
to carry out its mission. The number of Federal mine inspectors has 
remained relatively constant over the last ten years, from a low of 902 
in 1998 to a high of 986 in 2003.

                       MSHA ENFORCEMENT PERSONNEL
------------------------------------------------------------------------
                                                 Metal and
               Year                    Coal       nonmetal      Total
------------------------------------------------------------------------
1996.............................          690          289          979
1997.............................          634          272          906
1998.............................          615          287          902
1999.............................          631          318          949
2000.............................          660          305          965
2001.............................          653          326          979
2002.............................          605          339          944
2003.............................          621          365          986
2004.............................          579          371          950
2005.............................          584          357          941
------------------------------------------------------------------------

    More importantly, the inspection workload has remained steady at 15 
mines per inspector in each of the last ten years. The number of coal 
mines decreased 26 percent over the last ten years but the number of 
coal mine inspectors declined only 15 percent during that time. 
Consequently, the coal mine inspector's workload has declined. In the 
late nineties there were 3.8 coal mines for each inspector. Since 2000, 
that workload has been reduced to 3.3 coal mines for each inspector.
    Although the workload has remained relatively constant, inspection 
activity, as measured by the average number of hours Federal inspectors 
spend at each mine, has increased 14 percent in the coal sector to 219 
hours per mine over the last five years (2001-2005) compared to 192 
hours per mine over the previous five years (1996-2000). MSHA has been 
able to retain its enforcement capacity through management efficiencies 
such as optimizing office locations and reducing the paperwork burden 
on inspectors.
    In 2005, MSHA issued the highest number of citations and orders 
since 1994. In recent years, MSHA increased its use of ``withdrawal 
orders'' to gain compliance with the standards. The use of this 
powerful enforcement tool--withdrawal orders require miners to be 
removed from the area affected by the order, often disrupting 
production--increased 17 percent over the last five years when compared 
to the previous five years. MSHA issued more ``withdrawal orders'' in 
both 2004 and 2005 than in any year since 1994.
    These increased levels of enforcement have helped reduce mining 
injury and fatality rates to historic lows. In 1978, the first year the 
MSHA operated under the new Mine Safety and Health Act of 1977, 242 
miners died in mining accidents. Last year, there were 57 mining 
fatalities, 22 at coal mines and 35 at metal and nonmetal mines. From 
2000 to 2005, the mining industry experienced a 33 percent decrease in 
fatal accidents nationwide--with coal mines seeing a 42 percent 
decline. The coal mine lost-time injury rate declined one-third over 
the last five years.
    Question. There has been a lot of talk about the communications 
devices and safety equipment available to mine workers and mine 
operators. What are the practical applications of these devices and 
more to the point, do they work? Where are they being used, with 
success, in deep coal mines?
    Answer. Currently, the only MSHA-approved communication systems 
that are available to mine operators, for use in gassy mines, are mine 
page phones, leaky feeder communication systems (which use hand-held 
radios) and the Australian-manufactured PED and Tracker IV TAG Systems. 
Mine page phones and leaky feeder systems require underground power and 
interconnecting wires. Communications provided by these systems could 
be compromised in the event of an emergency, fire or explosion.
    MSHA has reviewed several installations of the Mine Site 
Technologies PED Cap Lamp/Paging system to determine if it is a viable 
solution to the mining industries' communications concern. There are a 
limited number of PED installations currently in the United States, and 
there have been installation and reliability problems with the system 
reported in the past by mine operators. MSHA investigators have 
recently visited PED installations at Peabody's Air Quality and 
Twentymile Mines and Consol Energy's Blacksville and Robinson Run 
Mines. MSHA has plans to evaluate the PED installation at BHP's San 
Juan Mine, the only U.S. mine with a surface-installed antenna loop.
    MSHA specialists are evaluating the Australian-made Mine Site 
Technologies Tracker IV TAG system, which is used to identify the 
general location (within zones) of miners underground. This system is 
not currently in use in the United States, but is installed in one 
underground coal mine in Australia. Due to the need for underground 
power for these two systems, MSHA will evaluate their ability to remain 
functional after power is interrupted as is often the case after an 
explosion or mine fire.
    One of the significant obstacles that we encounter in dealing with 
the currently available technology is that most systems are not 
designed for hazardous location use. Systems intended for use in 
underground gassy mines, if not properly designed, could themselves 
present an explosion hazard. The current availability of communications 
equipment that is designed for hazardous location use is very limited.
    In response to MSHA's recent Request for Information (RFI), we have 
received over 50 proposals from manufacturers for evaluation of 
emergency communication and tracking systems. Additional proposals are 
still being received on a daily basis. MSHA is currently reviewing 
these proposals, arranging meetings with the manufacturers and field 
testing of the systems. MSHA has posted an internet request for 
manufacturers of advanced communication and tracking systems to contact 
MSHA to partner in the development of mine-worthy systems. We have 
already met with manufacturers of several very promising systems. We 
are focusing our review of emergency systems on those systems that do 
not rely on a wire back-bone and that would remain functional in a fire 
or explosion. Once systems have proven to be mine-worthy and can 
provide the desired emergency operations, MSHA will assist the 
manufacturers in obtaining MSHA approval.
    Our most current findings on these and other communications and 
tracking technologies, including the PED and Tracker IV Tag systems, 
can be found at: http://www.msha.gov/Techsupp/PEDLocatingDevices.asp
    MSHA is also working with NIOSH and equipment manufacturers to test 
and develop thru-the-earth, electromagnetic, locator beacons that may 
be able to locate miners who have become trapped in the mine. This 
system may also be able to work with existing proximity systems that 
MSHA helped develop for underground mines, to prevent miners from being 
injured by mining equipment. This technology is promising, but needs 
further development.
    MSHA strongly believes that this approach will lead to much 
improved technology being made available to mine operators to respond 
to any future emergencies.
    Question. What are the protocols or requirements for life support 
and rescue equipment in a mine?
    Answer. At the time of the Sago accident, MSHA regulations required 
the operator to provide one approved self-contained self-rescuer (SCSR) 
for each miner who goes underground as well as for each visitor who go 
underground. An SCSR must have the capacity to protect the individual 
one hour or longer. On March 9, 2006, MSHA published an emergency 
temporary standard (ETS) requiring operators to immediately provide an 
additional SCSR for each miner and to provide caches of SCSRs for each 
miner at strategic locations in the mine if it takes longer than one 
hour to exit the mine. Training in evacuation procedures and use of 
multiple SCSRs is also required. In addition, the ETS requires 
operators to install lifelines to the surface from working face areas 
through both escapeways. Finally, the ETS requires mine operators to 
notify MSHA of mine accidents within 15 minutes.
    Question. Are the current laws and regulations sufficient to 
address the issues, or do we need to take new action?
    Anwer. The Mine Safety and Health Act of 1977 is a strong and 
effective worker protection statute that has contributed to the 
remarkable reduction in accidents, injuries and illnesses in our 
nation's mines. The Department has proposed to amend the Act to 
increase the maximum civil penalty for flagrant violations from the 
current $60,000 to $220,000 to address egregious violations.
    MSHA's comprehensive safety and health standards are constantly 
being reviewed and adjustments made to improve them or address newly 
recognized hazards. MSHA issued an Emergency Temporary Standard on 
March 9, 2006, to address underground supplies of oxygen generating 
breathing devices, training, lifelines and immediacy of accident 
notification.
    Question. Often we hear of mines being fined and then having those 
fines reduced. Is this common practice? Does this affect the amount of 
negligence at the mine?
    Answer. If a penalty is contested by the mine operator, it is tried 
before an administrative law judge of the Federal Mine Safety and 
Health Review Commission, an independent adjudicatory body. 
Approximately 6 percent of MSHA's proposed civil penalty assessments 
are contested each year and 2.5 percent of proposed civil penalties are 
reduced as a result of this litigation. For the period 1996-2005, the 
$225 million in proposed civil penalties were reduced by $24 million 
through litigation. I do not believe the reduction in civil penalties 
that occurs through litigation materially affects the amount of 
negligence at mines.
    Question. How often does MSHA have enforcement personnel at a 
facility?
    Answer. The Mine Act requires that MSHA make inspections of each 
underground mine in its entirety at least four times a year and of each 
surface mine in its entirety at least twice a year. The Mine Act also 
requires additional spot inspections and investigations to monitor and 
address conditions such as the liberation of certain quantities of 
methane, investigations of accidents, and responding to hazardous 
condition complaints. Therefore, MSHA enforcement personnel are 
frequently at mines carrying out the inspections and investigations 
required under the Mine Act.
    In 2005, enforcement personnel spent over 685,000 hours at mining 
operations--about 434,000 hours at coal mines and 251,000 hours at 
metal and nonmetal mines. There were MSHA enforcement personnel at mine 
sites every day last year.
    The amount of time enforcement personnel spend at any particular 
mining operation is contingent on a host of factors, including the size 
and type of the operation, the condition of the mine, and the number of 
days the mine is open or in production. Because underground coal mines 
are generally more hazardous work environments than other types of 
mines, MSHA enforcement personnel spend more time at underground coal 
mines than at other types of operations.
    In 2005, MSHA spent, on average, over 60 days at each underground 
mine that produced coal during at least part of the year. On many of 
these days there were multiple inspectors at these mines. There were 
over 100 underground coal mines at which MSHA enforcement personnel 
spent over 100 days on site, with 16 of these mines having MSHA 
personnel on site more than 200 days during the year.
    Question. Explain the process for enforcing regulations and any 
subsequent appeals process for fines levied.
    Answer. MSHA conducts frequent inspections of the Nation's mines to 
ensure compliance with MSHA's safety and health standards. If an 
inspector finds a violation of a standard, he must issue a citation or, 
in certain specified circumstances, a withdrawal order. In addition, 
MSHA, applying statutory criteria, proposes a civil penalty for every 
violation. If a citation, order, or penalty is contested by the mine 
operator, it is tried before an administrative law judge of the Federal 
Mine Safety and Health Review Commission, an independent adjudicatory 
body. The judge's decision may be appealed to the 5-member Commission, 
and the Commission's decision may be appealed to a U.S. Court of 
Appeals. Although the judge and the Commission are required to apply 
the same penalty criteria as MSHA, they are not bound by MSHA's penalty 
proposal. In contested cases they frequently reduce MSHA's penalty 
proposal, often significantly.
    Question. What is MSHA doing to ensure that we continue to reduce 
the risk of mining to miners?
    Answer. MSHA is relying on all its statutory tools to ensure that 
we continue to make mining a safer occupation. MSHA is committed to 
strong, fair and consistent enforcement. For example, last year MSHA 
completed 99.6 percent of its statutory coal mine inspections and 
continued to respond to every hazard complaint received. MSHA also has 
an active training and education program including an office devoted to 
assisting small mines with their compliance responsibilities. In 
addition, compliance assistance is available to all mine operators who 
face challenges in understanding mining's risks and dealing with 
compliance issues across the board. MSHA is also supporting research 
efforts by NIOSH into new technologies for improved mine safety and 
health. In addition, MSHA is using its own resources to test and 
develop new technologies for mine emergencies including use of robots 
and sensing devices that will de-energize equipment before it can 
injure or kill a miner working in close proximity.
                                 ______
                                 
             Questions Submitted by Senator Robert C. Byrd

    Question. Why did it take two hours for MSHA to be notified of the 
Sago explosion?
    Answer. The mine operator is responsible for immediately notifying 
the appropriate MSHA district office once an accident occurs, as 
required by MSHA regulations at 30 CFR 50.10. It should be noted that 
MSHA provides a 24/7 emergency answering service number for the 
operator to use if the operator is unable to reach appropriate district 
personnel. The emergency number is listed in 30 CFR 50.10 and on MSHA's 
website and is regularly made known to mine operators through various 
methods. When the mine operator calls the emergency toll-free number, 
the answering service immediately contacts the appropriate MSHA 
headquarters official who then immediately notifies the district 
office. This service has been provided to the mining industry for many 
years.
    The management at the Sago mine had the telephone numbers of 
persons in the district that had jurisdiction over the mine. According 
to MSHA's telephone logs, the mine operator began to call MSHA 
personnel at approximately 8:15 a.m. on January 2, 2006. At 
approximately 8:30 a.m. the mine operator was able to speak with an 
MSHA field office supervisor. Sago management also did not contact 
MSHA's emergency answering service number. The issue of immediate 
notification is currently being examined by MSHA's accident 
investigation team and enforcement action will be taken if appropriate. 
On March 9, 2006, MSHA published an emergency temporary standard that, 
among other things, defined the requirement to immediately notify MSHA 
of accidents by clearly stating that such notification must be without 
delay and within 15 minutes.
    Question. Given the 2.5 hours it took MSHA to be notified of the 
underground Alma fire, what is wrong with MSHA's notification system?
    Answer. Under MSHA regulations, the mine operator is required to 
immediately notify the Agency when an accident occurs. If the operator 
fails to immediately contact MSHA, the Agency's response and rescue 
efforts suffer accordingly. The management at the Alma mine had the 
telephone numbers of persons in the district that had jurisdiction over 
the mine. As noted above, Alma mine management also had access to the 
Agency's 24/7 emergency answering service number. MSHA's accident 
investigation team has begun a careful and thorough process to examine 
this issue and will issue a report of its findings and enforcement 
action will be taken if appropriate. On March 9, 2006, MSHA published 
an emergency temporary standard that defined the term ``immediate'' to 
mean ``at once without delay and within 15 minutes.'' In the event that 
communications are lost due to an emergency or other unexpected event, 
the term ``immediate'' means ``at once without delay and within 15 
minutes of having access to a telephone or other means of 
communication.''
    Question. Why did it take six hours for rescue teams to arrive at 
Sago?
    Answer. It was the responsibility of Sago mine management to 
provide the services of mine rescue teams for the rescue and recovery 
effort. MSHA regulations at 30 CFR 49.2(f) state that the mine rescue 
station(s) with which the two teams required to be available at each 
mine are associated must be located within two hours ground travel time 
from the mine. In addition, 30 CFR 49.2(g) requires that the teams 
present themselves at the mine within a reasonable time once 
notification is given. The explosion occurred at approximately 6:30 
a.m.; MSHA was notified at approximately 8:30 a.m.; and the first two 
mine rescue teams arrived at approximately 11:00 a.m., or 4.5 hours 
after the explosion. The teams were prepared to go underground at noon. 
However, even with mine rescue teams on the premises, the conditions 
within the mine had to be determined safe enough for their entrance 
(e.g., because of possible elevated carbon monoxide levels and the risk 
of secondary explosions due to high methane levels). At approximately 
5:20 p.m., teams were sent into the mine to begin rescue work after 
MSHA and State officials determined that this work could safely 
commence. The time that the two mine rescue teams were notified will be 
determined during the accident investigation. Each distinct response 
issue is currently being investigated by MSHA's accident investigation 
team, which will produce an independent report of its findings, and 
enforcement action will be taken if appropriate.
    Question. Why is a rapid notification and response system not 
available?
    Answer. Under MSHA regulations, the mine operator is required to 
immediately notify the Agency when an accident occurs. If the operator 
fails to immediately contact MSHA, the Agency's response and rescue 
efforts suffer accordingly. The management at the Sago mine had the 
telephone numbers of persons in the district that had jurisdiction over 
the mine. As noted above, Sago mine management also had access to the 
agency's 24/7 emergency answering service number. This emergency number 
has been published in 30 CFR Part 50 for many years.
    A mine rescue team's availability in relation to mine emergency 
response is governed by MSHA's regulatory requirements. Coupling a 
reasonable period of time to get to the mine rescue station with a 2-
hour ground travel distance from the mine rescue station, one could 
realistically expect a 2- to 3-hour response time by mine rescue teams. 
Underground fires and explosions result in conditions that must 
stabilize over a period of time before rescuers can safely enter the 
mine. Under MSHA's mine emergency response requirements, mine rescue 
teams almost always arrive at emergency sites before the conditions are 
safe enough for them to enter the mines. MSHA's recent Request for 
Information (RFI) asks for comments from the public on mine rescue 
procedures and mine technologies.
    Question. Why does MSHA not require more than two rescue teams at 
every mine in case of emergency?
    Answer. The legislative history for section 115(e) of the Mine Act 
indicates that Congress considered the ready availability of mine 
rescue capability in the event of an accident to be a vital protection 
for miners. In responding to the direction of Congress in section 
115(e) of the Act, the rulemaking process addressed, and the final Part 
49 rule required the ready availability of two mine rescue teams for 
each underground mine. In the discussion and summary of the final rule 
for Part 49, MSHA addressed hundreds of comments and statements.
    Since its inception, compliance with Part 49 has been accomplished 
by company teams, contract teams and state-maintained teams. 
Historically, the industry's response to accidents has included sending 
additional mine rescue teams, equipment, and personnel to emergency 
sites. This additional response is voluntary and not mandated. Notably, 
there was no shortage of well-trained mine rescue teams at the Sago and 
Alma mines. Following the Sago and Alma tragedies, MSHA is evaluating 
the current needs and response capabilities for effective mine rescue.
    It is important to remember that mine rescue teams using breathing 
apparatus perform a very specialized function and are only used in 
conditions where there is irrespirable air as the result of a fire or 
explosion. The vast majority of mining accidents are handled without 
the need for mine rescue teams.
    Question. Given how many rescue teams were borrowed at Sago, how 
prepared is a coal operator who meets MSHA's minimum requirement of two 
rescue teams per mine?
    Answer. Two mine rescue teams are able to conduct the initial 
phases of a rescue and recovery operation. One team conducts mine 
exploration while the other team backs them up at the fresh air base. 
If the emergency event extends beyond a period of 8 hours additional 
teams are essential. Historically, most mine emergency events last well 
beyond 8 hours. The industry presently relies on the rest of the mining 
community to provide additional teams necessary to respond to emergency 
events. This has proven to be a reliable process in the past.
    Question. What would have happened if other operators were not able 
to loan rescue teams?
    Answer. Historically, the mining industry has demonstrated their 
willingness to assist other companies in an emergency situation. 
Inability or refusal to lend rescue teams to other mines has simply 
never occurred. If other operators had not responded to the Sago 
disaster, the company would have had to rely on State teams in West 
Virginia and other state maintained teams. In addition, MSHA's Mine 
Emergency Unit (MEU) would assist these teams in the rescue operations.
    Question. Given the number of rescue teams used at the Sago and 
Alma Mines, how stressed is the system right now? What happens if 
another mine rescue should be necessary in the coming days?
    Answer. An adequate number of teams responded to the Sago and Alma 
accidents. If another accident were to occur that required mine rescue 
teams to intervene, history has shown the teams that responded to Sago, 
Alma, and/or other teams would be able and willing to respond. There 
are almost 190 mine rescue teams throughout the United States, 
available to respond to mine emergencies when needed.
    Question. In 2002, why did MSHA acknowledge the need to increase 
the number of mine rescue teams, but then withdraw its assessment of 
current regulations?
    Answer. The number of mine rescue teams has declined over the 
years, as has the number of mines. MSHA looked at regulations that 
would increase the number of these teams and held a public meeting in 
March 2002 in Barbourville, Kentucky to gather current ideas and 
suggestions concerning mine rescue capabilities and preparedness. Both 
labor and industry stated that cost is the major factor considered in 
establishing a mine rescue team. Recommendations to MSHA focused on 
incentives, particularly reducing penalties for violations if a mine 
had a mine rescue team. Legally, MSHA could not adopt that approach. 
Therefore, MSHA withdrew the mine rescue agenda item and issued two 
Program Information Bulletins that addressed mine rescue cost concerns 
related to training and technical assistance.
    Question. Without regulations, how can MSHA address recalcitrant 
coal operators who comply only with the minimum requirements for mine 
rescue?
    Answer. Mine operators who comply with the legal requirements of 
the Mine Act and MSHA regulations are meeting the requirements of the 
law and are providing the degree of protection required by the Mine 
Act. MSHA is actively developing regulations to ensure that miners are 
protected during a mine emergency. For example, on March 9, 2006, MSHA 
published an emergency temporary standard requiring operators to 
immediately provide an additional SCSR for each miner and to provide 
caches of SCSRs for each miner at strategic locations in the mine if it 
takes longer than one hour to exit the mine. Training in evacuation 
procedures and use of multiple SCSRs is also required. In addition, the 
ETS requires operators to install lifelines to the surface from working 
face areas through both escapeways. Finally, the ETS requires mine 
operators to notify MSHA of mine accidents within 15 minutes.
    In addition, on January 25, 2006, MSHA published a Request for 
Information (RFI) on issues related to mine rescue teams and mine 
rescue equipment and technology. MSHA is in the process of evaluating 
the comments in response to the RFI and has in-mine testing underway 
for several promising communication systems to aid in mine rescue.
    Question. Why did MSHA allow the Sago Mine to continue to operate 
with a total of 276 safety violations?
    Answer. MSHA does not have the authority to force a mine to 
permanently cease operations. Under certain circumstances, MSHA has the 
authority to suspend mining activity in the area affected by a 
violation by issuing a ``withdrawal order.'' MSHA aggressively 
exercised its withdrawal authority against Sago Mine in 2005, issuing 
18 separate ``withdrawal orders'' that shut down mining activity in 
specific areas of the mine until health and safety problems were 
corrected by the operator. Under the Mine Act, the operator of the Sago 
Mine has the right to reopen an affected area once all of the safety 
issues that caused the withdrawal order have been corrected. Absent a 
mine-wide hazard, MSHA does not have the statutory authority to shut 
down a mine simply because it has a large number of other violations. 
MSHA records indicate a total of 208 enforcement actions at the Sago 
Mine in calendar year 2005.
    Question. How did MSHA utilize Section 104 of the Mine Act in 
regard to Sago, which authorizes MSHA to close an area of the mine, 
after 90 days notice to a coal operator, if MSHA finds an unwarrantable 
failure to curb habitual violations?
    Answer. MSHA utilized section 104(a) citations, section 104(b) 
withdrawal orders, section 104(d) citations, section 104(d) 
unwarrantable failure withdrawal orders, and section 103(k) withdrawal 
orders at the Sago Mine. MSHA did not use its ``pattern of violations'' 
withdrawal order authority at the Sago Mine because the mine did not 
meet the criteria for a ``pattern of violations'' under the Agency's 
pattern regulation at 30 CFR Part 104. This regulation requires a 
review of the mine's compliance records at least once a year for: (1) a 
history of significant and substantial (S&S) violations; (2) use of 
increasingly stringent enforcement tools; and (3) evidence of the 
operator's lack of good faith in correcting the problem that results in 
repeated S&S violations. In the preamble to the pattern regulation, 
MSHA suggested that a prior two-year history be used in conducting the 
pattern review. The Sago mine did not operate continuously during 2004, 
making a review for the period of 2004 to 2005 problematic. In 
addition, the Sago mine changed ownership during 2005, with changes in 
company personnel. The accident incidence rate at the mine declined 
late in 2005, under the new owner. The fact that Sago received 208 
enforcement actions during 2005 would not, alone, trigger the pattern 
sanction. The compliance record had to be reviewed within the context 
of other pattern criteria. MSHA had appropriately increased enforcement 
at the Sago mine using the tools available under sections 103, 104, and 
l07 of the Mine Act.
    Question. How many pattern letters or notices did MSHA send in 
regard to the Sago Mine?
    Answer. None. The mine did not meet the criteria for a ``pattern of 
violations'' specified in 30 CFR 104.
    Question. How many pattern letters have been issued since you 
became Acting Assistant Secretary in 2004?
    Answer. Since I became Acting Assistant Secretary, MSHA has issued 
one pattern letter. Subsequently, the mine operator submitted an action 
plan to improve safety at their mine. Evidence showed that improvements 
did occur and that the operator was committed to improving safety and 
health conditions.
    Question. How did MSHA utilize Section 108 of the Mine Act in 
regard to Sago, which authorizes MSHA to institute a temporary 
injunction, whenever MSHA finds a mine operator engaged in a pattern of 
violations?
    Answer. No such injunction has ever been sought or issued in the 
history of the Mine Act. In any case, the mine did not meet the 
criteria for a ``pattern of violations'' specified in 30 CFR 104.
    Question. How many times has this authority been invoked since you 
became Acting Assistant Secretary in 2004?
    Answer. No such injunction has ever been sought or issued in the 
history of the Mine Act.
    Question. What influence does the Arlington office have on the size 
of penalties assessment by inspectors? Since 2001, have policies been 
enacted that allow the Arlington office to veto or reduce the size of 
penalties? Has the Arlington office ever vetoed or reduced the size of 
a penalty assessed by an inspector?
    Answer. MSHA's inspectors do not determine the amount of civil 
penalties for the citations and orders they issue. Almost 98 percent of 
all civil penalties issued are assessed by MSHA's Office of Assessments 
using its computerized assessment system based on criteria in 30 CFR 
Part 100. The inspectors establish the specific values for each of the 
five Part 100 criteria when they issue citations and orders. 
Thereafter, the citation and order data is entered into MSHA's 
electronic data system and subsequently transferred to Assessments for 
civil penalty determinations. The remaining 2 percent to 3 percent of 
civil penalties are for the more egregious violations that receive 
special assessments. These assessments are higher fines determined by 
special assessors in the Office of Assessments using procedures to 
evaluate the same five criteria given in 30 CFR Part 100.
    MSHA has not adopted any policies since 2001 that allow the 
Arlington office to veto or reduce the size of penalties. All penalties 
are determined by the Office of Assessments. The Office of Assessments 
does not seek a consensus among senior agency officials when 
determining penalties. I have never reduced an assessment proposed by 
the Office of Assessments.
    When a civil penalty is issued, the mine operator has 30 days to 
pay or contest the assessment. Once a proposed penalty is contested, 
the Mine Act mandates that the proposed assessment shall not be 
compromised, mitigated, or settled except with the approval of the Mine 
Safety and Health Review Commission (Commission). In addition, no 
penalty assessment, which has become a final order of the Commission 
shall be compromised, mitigated, or settled except with the approval of 
the Court. Therefore, reductions of civil penalties are made by or 
approved by the Commission and its administrative law judges.
    Question. MSHA announced on Saturday that it would reassess 
regulations related to mine rescue teams and technology. Does this 
announcement mean that MSHA made a mistake in abandoning its assessment 
of these rules in 2002?
    Answer. MSHA recently published a Request for Information on Mine 
Rescue Equipment and Technology. We did so because significant 
questions remain about the availability and reliability of current 
technology. Promising technologies are known to us, but practical 
experience with their use remains very limited. Our decision to explore 
these technologies and their potential application in the nation's 
mines is not inconsistent with the decision to remove the previous 
rulemaking agenda item, which was never published as a proposed rule. 
Recommendations to MSHA focused on incentives, particularly reducing 
penalties for violations if a mine had a mine rescue team. Legally, 
MSHA could not adopt that approach.
    Since 2001 we have had a continuing interest in improving mine 
rescue capability. We have supported NIOSH research into two-way 
wireless communications and have worked hard at developing robotic 
technology for use during underground emergencies. In addition, in 
March 2002, MSHA held a public meeting at the Mine Health and Safety 
Academy to gather ideas and suggestions from the mining community on 
the availability, quality and preparedness of mine rescue teams.
    Question. Technology has improved significantly since the days when 
miners used to carry canaries underground, but mine rescue efforts 
still are hindered by the inability to communicate directly with 
miners, to pinpoint their location underground in case of emergency, 
and to provide them with sufficient oxygen in case they are forced to 
barricade themselves and wait for rescue.
    How can we better coordinate the mine safety technology being 
developed in other Federal agencies, like the Defense and Energy 
Departments, and in the commercial world, to equip miners with longer 
lasting breathing devices and more advanced communications technology?
    Answer. In 2005, through two workshops, MSHA and NIOSH reviewed new 
technology that is currently under development in industry and 
government, which may have an impact on new Self-Contained Self-Rescuer 
(SCSR) technology. There are several promising technologies on the 
horizon including new oxygen sources and filter technology; however, 
new hybrid SCSR technology may have the most immediate impact on the 
development of smaller, lighter, longer lasting SCSRs. A hybrid SCSR 
would combine oxygen generation technology with new filter technology. 
At least one manufacturer is interested in building such an SCSR. In 
addition, MSHA and NIOSH would have to update respirator approval and 
certification standards in order to approve such a novel device. MSHA 
and NIOSH have found the workshop forum to be useful, and will continue 
to sponsor such workshops in the future, to further seek out new 
technology, which is being developed in both government and industry.
    MSHA is also working with NIOSH and equipment manufacturers to test 
and develop through-the-earth, electromagnetic, locator beacons that 
may be able to locate miners who are involved in emergencies, and who 
become trapped in the mine. This system may also be able to integrate 
with existing proximity systems that MSHA helped develop for 
underground mines to prevent miners from being injured by heavy mining 
equipment. This technology is promising.
    The Mine Act does not provide the Department of Energy (DOE) with 
statutory authority for research into new technologies for mine safety. 
However, DOE has partnered with the mining industry to foster the 
development and use of advanced technologies and processes to save 
money, cut emissions, and reduce waste. Research and development 
projects are conducted through this partnership in areas of extraction, 
beneficiation and processing, materials handling, and cross-cutting 
applications. MSHA is aware of active and completed projects that have 
an effect on the mining industry. One project being conducted by DOE, 
NIOSH, academia and the mining industry is the development of an 
emergency communication system. This project is one of six selected 
that MSHA will evaluate from a total of 70 proposals received in 
response to a request for information recently published. The six 
selected projects do not rely on a wire back-bone and have the greatest 
potential to remain functional in the event of a roof-fall, inundation, 
fire, or explosion.
    Security concerns can impair the ability of the departments and 
agencies within the Department of Defense to share technologies with 
civilian agencies, particularly in the area of communications.
    Question. Why not create an Office of Technology in MSHA to assess 
new technologies in other Federal agencies and the commercial world, 
and to see if they would be applicable to mine safety?
    Answer. MSHA already performs this function through our Office of 
Technical Support at MSHA's Approval and Certification Center in 
Triadelphia, West Virginia. I invite you to visit this facility and see 
for yourself the high quality of work performed there by MSHA's highly 
qualified and experienced engineers. Additional work investigating 
state-of-the-art equipment and technologies related to mine rescue or 
mine fires and explosions is performed by other MSHA engineers and 
scientists at the Pittsburgh Safety and Health Technology Center in 
Bruceton, Pennsylvania. The various technical Divisions within the 
Centers are constantly looking at new technologies and working with 
manufacturers, other Federal agencies, academia, and industry to 
identify, develop, test and promote equipment, systems and other types 
of products that have the potential to improve miners' safety and 
health.
    Examples of these efforts include a recent partnership with two 
equipment manufacturers, a mining machine manufacturer, NIOSH, and two 
mining companies to develop and test a proximity protection system that 
can be placed on a remote control continuous mining machine to prevent 
operators from placing themselves in the hazardous area of the machine 
as it is performing tramming operations. This project was precipitated 
by the high incidence of pinning and crushing fatal injuries (29) that 
the Agency has investigated over the last 17 years. The result of this 
effort will be two commercially available systems that have been 
approved by MSHA and can be installed on continuous mining machines to 
prevent these types of accidents from occurring.
    Other examples of the new technologies investigated by Technical 
Support include motion detectors that can be installed in surface 
mining pits to prevent haul trucks from traveling over berms when 
dumping, and development of a high impact resistant glass for 
installation on bulldozers working on surge piles at coal mine 
preparation plants As coal is removed by draw-off feeders from 
underneath large coal surgepiles, the coal does not draw to the 
surface, but rather, bridges over such that the surface appears to be 
solid. As machinery such as a bulldozer travels over the surface, the 
bridged-over coal can collapse, and the bulldozer will fall into the 
void. Fatalities have occurred as coal crushes through the bulldozer 
cab windows and suffocates the operator. Impact resistant glass will 
prevent this outcome.
    One of the significant obstacles that we often encounter in dealing 
with the currently available technology is that most systems are not 
designed for hazardous location use. Systems intended for use in 
underground gassy mines, if not properly designed, could themselves 
present an explosion hazard. The current availability of communications 
equipment that is designed for hazardous location use, for example, is 
very limited.
    In response to MSHA's recent Request for Information on mine rescue 
technologies, Technical Support has received over 50 proposals from 
manufacturers for evaluation of emergency communication and tracking 
systems. Additional proposals are still being received on a daily 
basis. MSHA is currently reviewing these proposals and arranging 
meetings with the manufacturers and field testing of the systems. MSHA 
has posted an internet request for manufacturers of advanced 
communication and tracking systems to contact MSHA to partner in the 
development of mine-worthy systems. We have already met with 
manufacturers of several very promising systems. We are focusing our 
review of emergency systems on those systems that do not rely on a wire 
back-bone and that could be destroyed in a fire or explosion. Once 
systems have proven to be mine-worthy and can provide the desired 
emergency operations, MSHA will assist the manufacturers in obtaining 
MSHA approval.
    Our most current findings on these and other communications and 
tracking technologies, including the PED and Tracker IV Tag systems, 
can be found at: http://www.msha.gov/Techsupp/PEDLocatingDevices.asp
    MSHA is also working with NIOSH and equipment manufacturers to test 
and develop thru-the-earth, electromagnetic, locator beacons that may 
be able to locate miners who are involved in emergencies, and who 
become trapped in the mine. This system may also be able to work with 
existing proximity systems that MSHA helped develop for underground 
mines to prevent miners from being injured by mining equipment. This 
technology is promising, but needs further development.
    Technical Support scientists and engineers are in constant 
communication with equipment and product manufacturers and participate 
in industry consensus standard setting committees to keep abreast of 
the latest state-of-the-art technologies that may offer improvements in 
miner's safety and health. We strongly believe that this approach leads 
to much improved technology being made available to mine operators to 
improve mine safety and health.
    Question. The 1977 Mine Act Committee report reads: ``To be 
successful in the objective of including effective and meaningful 
compliance, a penalty should be of an amount which is sufficient to 
make it more economical for an operator to comply with the Act's 
requirement than it is to pay the penalties assessed and continue to 
operate while not in compliance.'' How did the severity of the 
penalties change as the number of citations issued against the Sago 
Mine began to increase?
    Answer. From the beginning of calendar year 2004 through the end of 
calendar year 2005, the number of citations and orders issued to the 
Sago mine increased. More importantly, the severity of the enforcement 
actions issued to the mine increased significantly later in calendar 
year 2005. During the last half of 2005 thirteen section 104(d)(2) 
withdrawal orders were issued to the Sago Mine. Only three section 
104(d) citations and orders had been issued to the mine during the 
previous 18 months. As a result, the amount of civil penalties also 
increased for the mine from $9,515 during CY 2004 to over $20,300 
during the first three quarters of CY 2005 and the majority of the 
unwarrantable failure withdrawal orders have not yet been assessed a 
civil penalty.
    Question. Given the increase in coal prices and profits, is it more 
economical for coal operators to comply with the Mine Act or to pay a 
penalty?
    Answer. The Administration believes that the statutory cap on civil 
penalties is too low to deter repeat and flagrant violations of the 
Federal Mine Safety and Health Act, and has urged the Congress to 
increase the statutory cap from $60,000 to $220,000 for flagrant 
violations. MSHA has also announced its intention to revise the 
decades-old civil penalty system.
    Question. If penalties are large enough, why did the number of 
citations at Sago Mine continue to increase?
    Answer. MSHA inspectors are instructed to cite all violations of 
the Federal Mine Health and Safety Act, and, in the last half of 2005, 
13 unwarrantable failure withdrawal orders were issued. Compliance at 
the Sago Mine was improving in the last quarter of 2005, but we were 
still not satisfied with their safety performance and were devoting 
considerable MSHA management time to addressing the problem with 
company officials.
    Question. The Mine Act provides mine operators with the right to 
contest a fine, and penalties often are negotiated. How often are 
penalties negotiated?
    Answer. Civil penalties are proposed by MSHA and are not 
negotiable. The citations and orders can be contested to, and the 
penalties reduced by, the independent Federal Mine Safety and Health 
Review Commission. For the period 1996-2005, roughly 6 percent of all 
proposed assessments issued by MSHA were contested, and the Commission 
reduced 2.5 percent of all proposed assessments.
    Question. What is the difference between the fines assessed and the 
amounts collected?
    Answer. For the period 1996-2005, approximately $225 million in 
proposed civil penalties were assessed to mine operators and 
contractors nationwide, that number was reduced by $24 million through 
litigation at the Federal Mine Safety and Health Review Commission, and 
MSHA has received payments totaling over $158 million, which is an 80 
percent collection rate. When a civil penalty is contested and 
subsequently reduced through litigation, the collection rate is over 98 
percent.
    Question. To what extent are the negotiations a matter of public 
record and to what extent do off-the-record negotiations take place?
    Answer. All settlements in contested civil penalties must be 
approved by the Federal Mine Safety and Health Review Commission and 
are therefore a matter of public record. All records related to the 
rare cases when the Agency reduced an assessed civil penalty due to 
adverse effect on an operator's ability to stay in business are 
available to the public in MSHA's case files.
    Question. What was the chain of command at the Sago and Alma Mines?
    Answer. The chain of command at any disaster, including Sago and 
Alma, changes as senior officials arrive on property. Subsequent to the 
initial arrival of MSHA first response personnel, the District Managers 
arrived on the properties and assumed their roles as senior MSHA 
officials for the emergency operations. The District Managers remained 
in charge of MSHA's efforts until recovery operations were completed.
    The Administrator for Coal was at the Sago and Alma mine sites 
during the emergency operations. He provided guidance and assistance to 
the District Managers during the mine emergency operations.
    The Acting Deputy Assistant Secretary arrived at the Sago property 
on the second day and was the senior MSHA official on property. He 
attended briefings for families, met privately with families, and 
provided briefings for the media and headquarters personnel.
    During both incidents, I was located at MSHA's headquarters command 
center where I received hourly updates from on site personnel and 
participated in decisions as needed and provided coordination with 
senior DOL officials.
    Question. How were decisions made?
    Answer. A three-party group was established at each of the 
operations. The group consisted of designated personnel from the 
Company, State, and MSHA. As issues arose, the group discussed options 
and came to consensus on decisions. A senior mine official, MSHA's 
District Manager, and the head of West Virginia's mine safety 
department had oversight responsibilities for this group to ensure that 
appropriate, timely, and safe decisions were made. As appropriate, 
MSHA's District Manager consulted with other senior officials prior to 
presenting an option to the three-party group.
    Question. Who had the final authority to make decisions?
    Answer. The Mine Operator was responsible for making decisions 
concerning the mine and the emergency operations. MSHA issued a section 
103(k) order for the mine and the State of West Virginia issued a 
similar order, requiring the operator to submit a plan to MSHA and the 
State for recovery. The plan had to be approved by MSHA and the State 
before implementation. By its very nature, this process created a 
problem-solving environment. Decisions at the two mine emergency 
operations were made through discussion and consensus among the 
participating parties, with each bringing their experience, knowledge 
and skills to the process.
    Question. What do we know about the MSHA rule issued in 2004 
regarding belt entries and the fire at the Alma mine?
    Answer. We know that the last approved map for the Alma mine 
indicated that belt air was being used to ventilate the longwall 
section, but not the section where the two deceased miners worked. In 
May 2000, the mine operator submitted a petition for modification 
seeking permission to use belt air to ventilate workings. The petition 
was granted in June 2000, with stipulations. Most of these stipulations 
were included in the 2004 belt air rule, which then rendered the 
petitions granted prior to the effective date of the rule unnecessary.
    While it is generally known that smoke from the fire contaminated 
the escapeway being used by the crew from number 2 Section at Alma 
Mine, MSHA is not able at this time to confirm the reasons for this 
contamination because the matter remains under investigation, and all 
facts are not yet known. There is no evidence that the proper use of 
belt air, as defined in the Alma mine's approved ventilation plan, in 
any way hindered the escape of the miners.
    Question. How many mines use belt entries to ventilate workplaces?
    Answer. Currently, there are 41 coal mines in the United States 
that use air from a belt air course to ventilate a working section. 
MSHA began granting petitions for modification to permit the use of 
belt air for this purpose in 1980. Sixty-seven petitions were approved 
between 1993 and 2001, each with a specific finding that the practice 
was safe. An additional 27 were approved between 2001 and the 
publication of the Belt Air rule in 2004, which included the major 
stipulations of previously granted petitions and rendered those 
petitions invalid with the application of a nationwide rule.
    MSHA's 2004 belt air rule not only codified what had become a 
routine and safe practice with the safeguards of technologically 
advanced atmospheric monitoring systems (AMS) and multiple fire 
suppression systems, but increased miner protection by adding various 
requirements that were not included in all the petitions. For example, 
all sensors used must be listed by a Nationally Recognized Testing 
Laboratory; the trunk lines for the communication system and the AMS 
must be installed in separate entries; carbon monoxide sensors must be 
installed in the intake escapeways; sensor spacing must be reduced to 
1,000 feet (as opposed to 2,000 feet for mines using older petition 
requirements); and lifelines are required when returns are used as 
alternate escapeways. MSHA records indicate that there has never been a 
fatality caused by belt air carrying fire or contaminants to the 
working section under the current rule or any of the preceding 
petitions for modification.
                                 ______
                                 
               Questions Submitted to Bennett K. Hatfield
             Questions Submitted by Senator Robert C. Byrd

    Question. Why did it take two hours to notify MSHA of the explosion 
at Sago? Is MSHA's notification system adequate?
    Answer. My written statement presented to the Subcommittee on 
January 20 provides significant detail on this point. However, to 
address the question more concisely, the time lapse of approximately 
117 minutes consists of a 69 minute period (6:31 to 7:40 A.M.) when the 
notification process wasn't triggered because mine management was fully 
engaged in the initial assessment and emergency response, and a 48 
minute period (7:40 to 8:28 A.M.) during which notification was being 
actively attempted but many telephone calls were not being answered.
    The initial 69 minute period reflects the time during which Sago 
mine management was: (i) attempting to determine what had happened, 
(ii) safely evacuating the 15 miners from underground that were not 
trapped, and (iii) launching a spontaneous rescue effort to reach the 
remaining 13 miners. The 5-man team of mine supervisors and safety 
managers reached over two miles into the mine, reestablishing temporary 
ventilation as they advanced to replace the explosion-damaged 
infrastructure, before being turned back by a wall of thick black smoke 
and heightened explosion risk when they were only 1,000 to 1,900 feet 
from the trapped crew.
    The subsequent 48 minute period reflects the time required to 
successfully make direct contact with a Federal (MSHA) inspector--
although it should be noted that a state inspector was contacted within 
16 minutes. Multiple calls were placed during this period but state and 
Federal offices were closed and many people were traveling due to the 
holiday.
  --The first notification call went to the state office in Fairmont at 
        approximately 7:40 A.M. Subsequent calls to home telephones and 
        cell phones of state mine inspectors were finally successful 
        with a returned call from state inspector John Collins at 7:56 
        A.M.
  --The notification calls to MSHA contact numbers began at 
        approximately 7:50 A.M. with a call to the home of MSHA Field 
        Office Supervisor Kenny Tenney. This call went to an answering 
        machine. Subsequent calls to MSHA's District 3 office, other 
        inspector home telephones, and inspector cell telephones were 
        also unanswered. At approximately 8:28 A.M., MSHA inspector Jim 
        Satterfield was contacted via his home telephone and was 
        notified of the accident.
    Hence, the notification process for state mine regulators required 
approximately 16 minutes, for a total elapsed time from accident 
occurrence of 85 minutes. The notification process for MSHA required 
approximately 48 minutes, for a total elapsed time from accident 
occurrence of 117 minutes.
    With respect to the question as to whether MSHA's notification 
system was adequate, we note that January 2 was a State and Federal 
holiday so it was admittedly more difficult to reach officials than on 
a normal workday. A central emergency call center would have made it 
easier for the company to ``notify'' MSHA and the state. However, a 
central call center would probably have had the same difficulty we 
experienced in reaching individual officials outside of normal working 
hours. Thus, even with a fully staffed central call center, we cannot 
say that mobilization of state and Federal officials would have 
resulted in a quicker response. Likewise, earlier contact with those 
officials would almost certainly not have hastened entry of the rescue 
teams into the mine due to the trend of carbon monoxide readings 
confirming high risk of a second explosion.
    Put simply, based on air quality monitoring by state inspectors 
beginning at around 8:30 A.M., Federal and State officials were in 
agreement that it was unsafe to enter the mine. In its deliberation, 
the Subcommittee should not cast aside the lessons learned by painful 
experience: rescuers are often at risk from secondary explosions and 
additional lives should not be put at significant risk in a rescue 
effort.
    Question. Why did the Sago Mine have only two rescue teams 
available in case of emergency? What would have happened if other coal 
operators were unable to loan rescue teams?
    Answer. To begin with, it is important to recognize that the 
availability of mine rescue teams had no bearing on the tragedy that 
occurred at the Sago Mine. The mine rescue teams responding on January 
2 had to spend hours and hours waiting as State and Federal mine 
authorities monitored carbon monoxide levels to determine whether safe 
entry was possible.
    That said, each mine is required by applicable law to have 
emergency coverage by two mine rescue teams within two hours of the 
mine. The Sago Mine contracted with the Barbour County Mine Rescue Team 
to provide this service. These teams included several miners from Sago 
and its affiliated operations.
    To answer the hypothetical question about what would have happened 
if other coal operators had been unable to loan rescue teams, there are 
many mine rescue teams that are not associated with any particular coal 
company--including the Barbour County Mine Rescue and Tri-state Mine 
Rescue groups--both of which assisted in the Sago Mine accident. More 
importantly, we are aware of no instance where there has been any 
widespread refusal or inability of other coal operators to offer the 
services of their mine rescue teams. Indeed, the events of the past 
month have proven that, even in an environment of stiff competition in 
the coal industry, competitors quickly rush forward with their mine 
rescue teams to assist in the time of a mine emergency. For example, in 
the Sago Mine accident, a total of 15 mine rescue teams were assembled 
on site. Additional to the Barbour County and Tri-state teams 
referenced earlier, this group included several mine rescue teams from 
Cansol Energy, one of our primary competitors in the region, as well as 
ICG's own Viper Mine Rescue Team that flew in from Illinois. Indeed, we 
received more offers of mine rescue assistance than we were able to 
utilize--including those from two other ICG competitors--Massey Energy 
and Arch Coal.
    These demonstrations of volunteerism, along with other recent 
examples across the coal industry, clearly confirm that overwhelming 
response by the industry in the time of need is the norm and not the 
exception.
    Question. How large are the fines assessed by MSHA in comparison 
with your companies [sic] profits? What is more economical--to comply 
with the Mine Act, or to pay the fine and operate in violation?
    Answer. During the first nine months of 2005, ICG (including the 
earnings of subsidiaries it acquired from Anker Coal Group in November 
2005) had operating profit (EBITDA) of approximately $81 million. 
During that same period, the fines assessed by MSHA against the 
associated mining operations were approximately $116,000. However, it 
is important to note that the Saga Mine lost approximately $6.2 million 
during that same nine month period. Despite the poor financial 
performance, upon ICG's gaining management oversight, our company 
initiated approximately $5.0 million in capital expenditures to improve 
the Sago Mine. Many of those capital expenditures had a direct impact 
on safety. Those safety efforts were detailed in my written and oral 
testimony to the Subcommittee.
    With respect to the nexus your question places between profits and 
the size of fines, I wish to point out that profitable mines would more 
logically have the means to maintain the capital expenditures that 
would allow more safety enhancements. However, some unprofitable 
mines--like Sago--are subsidized by their owners to make such 
improvements. The Sago Mine currently operates solely to meet the 
contractual requirements of its customers. We certainly hope that the 
Sago Mine becomes profitable, but the facts clearly reflect that while 
it has operated (even at a loss), ICG has provided an influx of 
capital. Comparing fines with profits seems to draw the incorrect 
correlation--fines should be based upon the severity of the violation 
rather than the financial health of the company cited.
    The second question also reflects an incorrect premise. For ICG and 
its subsidiaries, the question is not whether mining can be conducted 
more economically by operating in violation of the law. ICG and its 
subsidiaries are committed to operating in compliance with the Mine 
Act. Unlike a factory floor, an underground mine is a dynamic 
environment, subject to ever-changing geologic conditions, in which a 
combustible fuel--coal--is mined. We proactively correct conditions 
without governmental oversight, as is shown by the steps we have taken 
as discussed in my testimony, but we also correct cited conditions 
immediately upon receiving a citation.
    Despite the apparently incorrect premise of the second question, 
safety and productivity go hand in hand. If one ignores the moral, 
social, and legal obligations that are the principal controlling 
factors for our company, the true economic disincentives to operating a 
non-compliant coal mine include: (i) Disruptions to production caused 
by governmental orders to cease production, even if limited in 
duration, and the loss of revenues due to that disruption; (ii) Expense 
related to lawsuits; (iii) Increased workers compensation and insurance 
premiums; (iv) Damage to reputation among customers and competitors; 
and (v) Inability to attract and retain skilled workers that clearly 
have other employment alternatives in the current labor market -all in 
addition to fines and penalties. Of course, the ultimate disincentive 
is the risk of injury or death to our employees.
    The true economic decision is not whether to operate a mine 
unsafely; rather it is whether a mining project is economically 
feasible. No mine operator will choose to operate at a loss 
indefinitely--nor will it choose to invest in new mines that are not 
economically justifiable. However, speaking for ICG, compliance with 
all applicable safety standards is the fundamental principle of our 
business; investments have been and will be analyzed only in a manner 
that presupposes such compliance.
                                 ______
                                 
                 Question Submitted to Cecil E. Roberts
             Questions Submitted by Senator Robert C. Byrd

    Question. At MSHA's Emergency Preparedness Conference in 1995, 
industry and labor representatives voiced concerns about the dwindling 
number of rescue teams allotted per mine.
    How many mine rescue teams should be maintained at each mine?
    Answer. The Union believes that every operator of a coal mine 
should be required to maintain two qualified mine rescue teams, which 
would be immediately available whenever miners are in the underground 
workings of the mine. (In practice, this would probably require the 
creation of at least three rescue teams so that whenever the members of 
one rescue team are working their normal assigned job underground, two 
other teams would be available on the surface, in event of an 
emergency.) All teams should be comprised of miners who are employed at 
that mine, as they are acutely aware of the mine's layout and 
conditions, to enable faster response and to afford their expertise to 
any additional mine rescue teams that may arrive to assist in an 
emergency. There should be no exceptions, even for small mines; this 
requirement must be recognized as a necessary business expense.
    In addition, specific training requirements such as drills one 
weekend each month at the mine site and mandatory participation in mine 
rescue contests each year should be required.
    Question. How do we increase the number of mine rescue teams?
    Answer. Allow a transition period of nine months from the date of 
the legislation, during which time each operator can identify and train 
sufficient numbers of miners to fill the needed mine rescue teams. 
Selected miners can be trained through dedicated classes at MSHA's Mine 
Academy in Beaver, WV and the Lake Lynn Lab in Fairchance, PA. The 
Rockefeller legislation would provide tax incentives (for a 3-year 
period) to promote the development of more mine rescue teams.
    Question. The 1995 MSHA report on its Mine Emergency Preparedness 
Conference reads: ``the industry's ability to learn from lessons of 
past accidents is limited by inadequate information sharing as well as 
by accident reports that do not always reflect what happened.''
    What else can be done to ensure a complete accident investigation?
    Answer. For all multiple-fatalities and life-threatening injuries, 
and other major disasters like Quecreek, in addition to MSHA's 
investigation of the accident GAO should review MSHA's conduct.
    After MSHA completes every accident investigation (and the GAO 
completes its report for multiple fatal accidents and other major 
disasters) the reports should be provided to employees at the District 
as well as National levels of MSHA, so that employees throughout the 
Agency can learn from the particular investigation; also copies should 
be made available to other organizations involved with miners' health 
and safety, as well as the affected operator and any miners' 
representatives at the affected operation. These should be generally 
available to the public, including posting on the MSHA web page.
    Also, to ensure all relevant information is relayed to MSHA's 
accident investigation team, the legally authorized representative(s) 
of the miners, in accordance with 30 CFR Part 40, should be included in 
all aspects of the investigation, including all interviews and the 
physical investigation. As MSHA itself and a Federal judge reviewing 
the role of miners' representatives at the Sago mine recognized, 
miners' representatives have the trust of affected miners and act as a 
valuable conduit of information to those conducting an investigation. 
The operator should be required to provide information, but not be 
present when witnesses provide information during MSHA's interviews; 
excluding the operator will enhance the likelihood that miners will 
provide complete and accurate information.
    Question. How long should it take for MSHA to be notified of an 
emergency, and how long should it take to organize and deploy mine 
rescue teams?
    Answer. MSHA should be notified ``immediately'' after a mine 
operator becomes aware that an emergency condition exists. Immediate 
should mean no more than fifteen minutes, absent an extraordinary 
explanation. If in doubt, the operator should provide notice to MSHA 
even before the scope of the emergency may be fully known. Emergencies 
for which notice must be provided includes: explosion, fire, 
innundation, massive roof fall, and a serious injury or fatal accident. 
The determination to send MSHA field personnel to the site after 
notification can and must be a determination of the Agency. The mine 
operator must be responsible for putting its own mine rescue teams into 
immediate service upon any emergency; it also may seek to call upon 
additional mine rescue teams from nearby locations even before MSHA 
arrives on site. However, MSHA must have authority to direct and 
coordinate the rescue.
    The representative of miners' should be notified as soon as notice 
is provided to MSHA.
    By requiring each mine to have two mine rescue teams immediately 
available, only minimal time would be required to deploy those 
individuals. Typically a team could be assembled as quickly as 30 
minutes, though it could take longer to assemble a team if its members 
were not all nearby and accessible by phone. Most miners live in the 
nearby communities and can respond quickly. If all operations 
maintained at least two teams at all times, then supplemental teams 
from the nearest mining operations would also be available to assist. 
Additional rescue teams can arrive at later times to bolster rescue and 
recovery efforts, and to trade out with the first responders. To give 
an operator quick access to such additional rescue teams, information 
about all rescue teams should be easily accessible through MSHA's web 
page--information about all rescue teams should be regularly updated 
and MSHA should ensure it is easy to find.
    Question. How do we create a rapid response mine rescue system?
    Answer. If every operation must maintain at least two mine rescue 
teams, as we suggest and as is required by the Mine Act, then a ``rapid 
response mine rescue system'' is well along to being in place. In 
addition, MSHA must make its personnel quickly available: phone 
numbers, including off-hours emergency contact numbers, must be 
provided to all operators for use in an emergency. All such contact 
information must be kept up to date. Also, a mobile triage/surgical 
unit located within a reasonable distance from mining operations could 
provide valuable resources when rescuers bring survivors to the 
surface. The Commonwealth of Pennsylvania currently has two such 
medical/surgical units readily available and can deploy them quickly in 
the event of a mine emergency. MSHA or the CDC could coordinate central 
placements of such units.
    Question. How do we insure that the size of penalties is large 
enough to encourage compliance?
    Answer. Penalties must be high enough to convince operators that 
full compliance is the best choice. The current base fine of $60 is far 
too low. For example, compare MSHA fines with those under OSHA: while 
OSHA deals with many far less-dangerous industries, fines for knowingly 
endangering workers begin at $5,000, while a coal operator 
demonstrating ``high negligence'' could be fined just a few hundred 
dollars. The base penalty under MSHA should begin at $1,000, 
irrespective of the size of the mine, the number of miners exposed or 
any other factor. For more serious violations, penalties should be much 
higher. After it assesses a penalty, MSHA should not negotiate or 
compromise that amount. The Agency must then collect the fines it 
assesses, on a regular and prompt basis.
    There needs to be an expectation throughout the industry that 
violations of the Mine Act will be cited with regularity, costly for 
the operator, and that abatement must be accomplished in a timely 
fashion. Pursuant to the existing penalty structure, the number of 
miners affected is one of the factors that is considered in assessing 
penalties. However, even the way MSHA applies this factor is 
unrealistically low: often MSHA counts how many miners are in the most 
immediate area without considering that a problem in one area can 
easily spread harm to all underground miners. Also, MSHA now considers 
the effect of a fine on an operator's ability to stay in business. We 
submit that factor is inappropriate. If an operator cannot satisfy its 
obligations under MSHA, then it should not be in the business. No miner 
should be subjected to unsafe mining conditions because his employer is 
on shaky financial footing. Penalties should be the same size for the 
same violation at all mines.
    Fines for a failure to abate should be immediately doubled, then 
tripled. Repeat violations should generally be assessed much higher 
amounts. MSHA should also have the authority to shut an operation for 
noncompliance and for a failure to pay outstanding fines. If an 
operator continues to ignore the law, it should lose the right to 
operate.
    Question. What do you think of assessing penalties as a percentage 
of profits rather than a flat rate?
    Anwer. This would not be good idea for a number of reasons: many of 
the smaller and more marginal operations pose some of the greatest 
risks to miners' health and safety; many coal companies are not 
publically traded so obtaining information about their profitability 
may not be available or verifiable; some companies may have 
artificially low profits because they pay extraordinary compensation to 
their top executives, or might be motivated to alter their financial 
records to avoid showing high profits if higher penalties could result; 
coal companies are notorious for having complicated and quickly-
changing corporate structures--figuring out which entity had what 
profit margin would be difficult to determine and could quickly change.
    Question. A miner knows his work area best. The Mine Act guarantees 
every miner union or nonunion the right to leave the mine whenever he 
feels his safety is threatened.
    How can we ensure that miners feel free to exercise that right 
without fear of retribution?
    Answer. This is a significant issue. In the same way that the NLRA 
does not prevent employers from firing employees when they try to form 
unions even though that is illegal, so too, do miners reasonably fear 
getting fired for exercising their rights under the Mine Act, including 
the right to withdraw. The right to withdraw is protected through the 
anti-discrimination provision of the Mine Act, at Section 105(c). 
However, enforcement of this protection takes too long and is not often 
pursued.
    While MSHA has the authority to seek temporary reinstatement for 
any miner who is fired unlawfully, this relief is rarely pursued. MSHA 
should regularly seek temporary reinstatements as soon as it finds 
prima facie merit to a discrimination claim. Also, it needs to act in a 
speedy fashion in these cases. MSHA habitually misses its own deadlines 
in investigating and prosecuting these discrimination cases. It needs 
adequate funding to accomplish this, and it must make this a high 
priority. MSHA needs to be more aggressive in protecting all of the 
miners' rights under the Mine Act.
    Note that the right to withdraw has been interpreted to be quite 
limited: only if a miner believes that his immediate safety is in 
jeopardy does he have the right to withdraw, and the withdrawal is from 
the particular area, not the mine itself. To exercise this right, a 
miner must bring the perceived danger to the attention of his 
supervisor stating expressly that the miner has immediate fears for his 
health or safety. The miner can only speak out for his own protection, 
and cannot seek to have any other miners withdrawn. This right is 
rarely invoked; to make it more meaningful the legal standard for 
withdrawal should be liberalized. Finally, a miner who pursues his Mine 
Act rights is often blackballed, whether or not the case is ultimately 
successful.
    Question. Why are you opposed to the MSHA rule that allows belt 
entries to be used for ventilation?
    Answer. The Union has always been opposed to using belt air to 
ventilate the working areas. We vigorously opposed implementation of 
this rule, and sued the Agency in Federal court when it was finalized 
in 2004; deferring largely to the government (under the applicable 
``arbitrary and capricious'' standard), the court overruled the union's 
challenges based on the present law.
    The belt entry is the only location in the mine where equipment 
remains in constant motion. This not only includes the conveyor belt, 
but also head rollers, tail rollers, belt take-ups and structural 
rollers used to support the conveyor belt. Stuck rollers, coal 
spillage, worn bearings, unaligned belt conveyor and other problems 
create extra friction, which can cause fires to ignite on the belts; 
with coal ever-present, this fuels such a fire. Further, when the belts 
themselves burn, toxic fumes are also generated.
    Conveyor belts extend for miles in underground coal mines, and 
these areas are not regularly monitored by personnel. The potential for 
fire in the mine is greatest in the belt entry; MSHA's data 
demonstrates that mine fires originate in belt entries most often. In 
addition, the propagation of fire throughout the entry by a burning, 
moving conveyor belt can quickly make a fire uncontrollable. Air 
coursing through the belt entry at high velocities further compounds 
the dangers posed by a belt fire. This is very similar to fighting a 
forest fire during a period of extremely high winds. The flames are 
fanned and even a small fire can become uncontrollable quickly.
    The end result is obvious. The smoke, fire, and toxic gases are 
directed by mine design toward the working areas of the mine and 
directly over miners working at the face. These events can happen 
quickly, rendering the belt and return entries not viable escapeways. 
Generally, there will be a third entry that also supplies air to the 
working section. However, because air pressures in the belt entry are 
permitted (by MSHA regulation) to be greater than in other entries, the 
fire and/or fire gases may bleed into other entries adjacent to the 
belt. Mines with this design allow the other entries to also become 
compromised when there is a fire on the belt. Miners have difficulty 
escaping in these cases. Indeed, we understand that t was along a belt 
entry that was used for ventilation, where the Aracoma Alma No.1 fire 
began, resulting in the deaths of miners Don Bragg and Ellery Hatfield 
in January, 2006.

                         CONCLUSION OF HEARING

    Senator Specter. Thank you all very much for being here. 
That concludes our hearing.
    [Whereupon, at 1:30 p.m., Monday, January 23, the hearing 
was concluded, and the subcommittee was recessed, to reconvene 
subject to the call of the Chair.]

