Hearing Type: 

Full Transcript

[Senate Hearing 108-656]
[From the U.S. Government Printing Office]

                                                        S. Hrg. 108-656




                               before the


                                 of the

                          UNITED STATES SENATE

                      ONE HUNDRED EIGHTH CONGRESS

                             SECOND SESSION




                             JULY 20, 2004

96-109                      WASHINGTON : 2004
For Sale by the Superintendent of Documents, U.S. Government Printing Office
Internet: bookstore.gpo.gov  Phone: toll free (866) 512-1800; (202) 512�091800  
Fax: (202) 512�092250 Mail: Stop SSOP, Washington, DC 20402�090001


                     PAT ROBERTS, Kansas, Chairman
          JOHN D. ROCKEFELLER IV, West Virginia, Vice Chairman
ORRIN G. HATCH, Utah                 CARL LEVIN, Michigan
MIKE DeWINE, Ohio                    DIANNE FEINSTEIN, California
CHRISTOPHER S. BOND, Missouri        RON WYDEN, Oregon
TRENT LOTT, Mississippi              RICHARD J. DURBIN, Illinois
OLYMPIA J. SNOWE, Maine              EVAN BAYH, Indiana
CHUCK HAGEL, Nebraska                JOHN EDWARDS, North Carolina
SAXBY CHAMBLISS, Georgia             BARBARA A. MIKULSKI, Maryland
JOHN W. WARNER, Virginia
                   BILL FRIST, Tennessee, Ex Officio
              THOMAS A. DASCHLE, South Dakota, Ex Officio
                      Bill Duhnke, Staff Director
               Andrew W. Johnson, Minority Staff Director
                    Kathleen P. McGhee, Chief Clerk

                            C O N T E N T S

Hearing Held in Washington, D.C., July 20, 2004..................     1
Witness statements:
    Feinstein, Hon. Dianne, a U.S. Senator from the state of 
      California.................................................    19
    Hamre, Dr. John J., President and CEO, Center for Strategic 
      and International Studies..................................    44
    Odom, Lieutenant General William E., U.S. Army Retired, 
      Senior Fellow, Hudson Institute............................    54
    Woolsey, Mr. R. James, Vice President, Booz Allen Hamilton...    57
Additional Materials: Prepared Statement of Senator Olympia J. 
  Snowe..........................................................     8



                         TUESDAY, JULY 20, 2004

                              United States Senate,
                   Senate Select Committee on Intelligence,
                                                    Washington, DC.
    The Committee met, pursuant to notice, at 2:42 p.m., in 
room SD-106, Dirksen Senate Office Building, the Honorable Pat 
Roberts (chairman of the committee) presiding.
    Committee Members Present: Senators Roberts, Hatch, DeWine, 
Bond, Lott, Snowe, Hagel, Chambliss, Warner, Rockefeller, 
Levin, Feinstein, Wyden, Durbin and Mikulski.
    Chairman Roberts. The committee will come to order. The 
committee meets today in open session to begin what I hope will 
be a continuing discussion of the state of our intelligence 
    In the wake of this committee's report on prewar 
intelligence and the upcoming release of the 9/11 Commission's 
report, intelligence reform should be, and once again is, 
center stage.
    I say ``once again'' because we have been down this road a 
number of times. Congress has, on a number of occasions, either 
of its own volition or in response to a specific event, 
attempted to reform the intelligence community. We have created 
new positions. We have made existing positions subject to 
Senate confirmation. We have reallocated resources. We have 
attempted to terminate and, at times, we have terminated 
programs and directed specific actions be taken.
    We have not, however, undertaken a major reexamination of 
the intelligence community's mission and structure. That effort 
starts today. And today's hearing will be the first in this 
series where we will not only hear from Members of Congress on 
specific proposals for reform, but also from noted experts 
including current and former intelligence community experts and 
    I have said many times that while the committee's Iraq 
report cites out for reform, we will approach the issue with 
deliberation and responsibility. We will examine closely all 
proposals for change and subject them to the rigors of 
pragmatism and reality. It is far more important to do this 
right than to do it quickly.
    The committee's recent report on the U.S. intelligence 
community's prewar assessments regarding Iraq identified 
significant problems. Unfortunately, the problems we have found 
in the prewar intelligence are not unique. As stated in the 
Joint Inquiry's report of 2002, and with the impending release 
of the 9/11 Commission's report, it is also clear there were 
significant problems with U.S. intelligence before 9/11.
    Unlike most congressional or commission reports, however, 
our report contains no reform recommendations. I believe very 
strongly that the issues involved are so complex and of such 
import that it is incumbent on the committee and the Congress 
to think very carefully and deliberately about the question of 
    We must base whatever actions we ultimately take on facts 
and considered judgment, not expediency or media-generated 
momentum or politics. I intend to examine closely, as does the 
vice chairman, all proposals for change, certainly keeping in 
mind that we should, first, do no harm and avoid as best we can 
the law of unintended consequences.
    A number of important reform ideas are being widely 
discussed. These proposals range from creating a director of 
national intelligence separate from the CIA to giving more 
power and resources to the CIA to splitting off the FBI's 
national security mission and creating a domestic intelligence 
agency similar to MI-5 and MI-6 in Great Britain.
    Over the next several months, we will endeavor to identify 
the universe of problems and challenges, and then, only then, 
craft the appropriate legislative solutions. Not only must we 
be prepared to act legislatively to address these problems, we 
must also be prepared to accept the fact that many of the 
solutions may not be within our reach. In those instances, we 
will make recommendations to the president and strongly urge 
that the appropriate action be taken.
    One of the important issues that will guide this 
committee's intelligence reform is the question of resources. 
Contrary to conventional wisdom, I do not believe that reform 
necessarily means more resources. It involves issues such as: 
information sharing, better management and leadership and being 
more aggressive and innovative in the way the intelligence 
community actually does its work.
    The intelligence community has had a significant increase 
in funding since 9/11. And the questions is now less, as far as 
I'm concerned, a matter of, ``Do they have enough?'' than ``Are 
they spending it wisely?''
    For example, both the president and Senator Kerry have 
proposed improving our human intelligence, what we call HUMINT. 
I agree that better HUMINT is needed. Everybody on the 
committee agrees that better HUMINT is needed. It is always 
needed. It is important to note, however, that our HUMINT 
programs have already had a significant increase in funding 
since 9/11.
    Our report points out that the CIA had zero sources in Iraq 
reporting on WMD after 1998 when the inspectors left. How much 
funding and personnel do you need to develop one source? Our 
report notes that many of the problems we've found in the 
HUMINT efforts that were directed against Iraq stem from a 
broken corporate culture and an overabundance of caution that 
will not be solved by additional funding and personnel.
    It is important to note that our report does not say that 
individual CIA officers are timid. It says that they work in an 
organization with a risk-averse and a broken corporate culture. 
There's a very big difference between the two. If the CIA had 
asked for volunteers in 1998 to infiltrate Baghdad, I don't 
think there would have been any shortage of volunteers. Whether 
they would have been properly trained and capable of executing 
that mission is another question entirely.
    The President has also talked recently about making sure 
that the intelligence community has the most advanced 
technology and remains at the cutting edge of change. I agree, 
but it's important to think carefully about where in the 
intelligence process we invest the funds for improved 
technology. We continue to spend money on increasing 
intelligence collection when we still don't have the ability to 
fully analyze what we already collect.
    The intelligence community has made strides in developing 
automated tools to help the analysts manage the mountains of 
intelligence that we collect every day, but we must do more. If 
we are serious about keeping the intelligence community at the 
cutting edge of change, we can no longer budget unnecessarily 
billions--and I mean billions--of dollars for new collection 
programs while dedicating far less for the analytical tools.
    The president has also recently called for improved 
coordination among the intelligence agencies. We all agree with 
that. The most pressing coordination problem is the issue of 
information sharing. There has been much improvement in 
information sharing since 9/11, but we still have a long ways 
to go.
    Our report describes a stunning lack of information sharing 
by the CIA's Directorate of Operations. The CIA has made some 
improvements in this area, but the next DCI needs to explain to 
the Directorate of Operations that it has an obligation to 
share its intelligence appropriately with cleared people across 
the entire intelligence community, not just within the CIA.
    The public debate over intelligence reform has focused 
lately on the creation of a director of national intelligence, 
or what we call the DNI. We will hear from Senator Feinstein 
shortly on her proposal. And we thank her for her leadership 
and her presence.
    I believe a number of questions--and I don't mean to 
perjure this legislation--but I believe a number of questions 
need to be answered before we consider such an approach in 
final form.
    For example, what problems have we identified that will be 
solved by such an approach? What is the current DCI unable to 
do that he or she will be able to do as a DNI? Why should we 
move authority away from the Secretary of Defense, who is 
responsible for fighting and for winning wars, to a DNI, who is 
not? Why is it a good thing to strip the departmental heads of 
the authority to manage their own intelligence elements? What 
is it about the intelligence mission that compels us to elevate 
it to a Cabinet-level representation? Is the intelligence 
community really a community, and should we further try to 
facilitate this construct, or should we embrace a different 
    I look forward to my distinguished colleague's testimony.
    Regardless of what form the next leader of the intelligence 
community eventually takes, reform can begin with the naming of 
a new Director of Central Intelligence. Under current law, the 
Director of Central Intelligence has significant authority to 
shape the intelligence mission if he or she chooses to do so.
    So the committee is prepared to expedite the nomination of 
a new DCI as soon as the president makes his decision and/or 
his selection.
    I believe that there is a consensus on this committee that 
the intelligence community needs fresh thinking and fresh 
leadership. Many in the intelligence community agree and 
understand where they can and must do better. There are others, 
however, that have yet to appreciate, I think, the full extent 
of the problem.
    To say that ``We get it,'' and then imply that the problems 
with the intelligence community's WMD assessments were 
reasonable at the time, or to state that the problems with the 
prewar Iraq assessments were isolated shortcomings, says to me 
that there are still those that don't--don't--get it.
    We need fresh thinking and a willingness to look in the 
mirror and honestly examine the community's performance over 
the last decade. It is my hope that the appointment of a new 
DCI, our Iraq report, and the 9/11 commission's report later 
this week will help really facilitate that process.
    This committee's examination of intelligence reform will 
not focus exclusively on the executive branch, however. We will 
also look carefully at the way Congress really conducts our 
oversight of the intelligence community. We took an important 
first step on this front with a provision in our intelligence 
authorization bill this year that eliminates term limits for 
the committee members. This will end the practice of forcing 
members off the committee just as they are becoming 
knowledgeable enough to serve as effective overseers of the 
intelligence community.
    There is wide bipartisan support for this measure, and I am 
hopeful the intelligence authorization bill will be brought up 
and passed by the Senate as soon as we can do that.
    As we consider the reform of the intelligence community, I 
feel strongly we must ensure that we institutionalize change as 
a continuous process. We can't make the mistake of rearranging 
the organizational chart to meet the current threat and simply 
stop there. We must leave in place a system that will continue 
to adapt to the new threats that face us.
    Whatever course the committee takes eventually takes on the 
question of reform--and we will not take it unilaterally--we 
will work with the executive branch and our counterparts in the 
House of Representatives and our colleagues in the Senate to 
construct an intelligence capability worthy of the men and 
women we ask to do this difficult and, many times, dangerous 
work and to better safeguard our nation's security.
    The distinguished vice chairman, Senator Rockefeller.
    Vice Chairman Rockefeller. That's a good statement, Mr. 
Chairman. I congratulate you on that and welcome that.
    We are going to be undertaking potentially the most 
importantpart of what we have done in the last several years, 
and that is the reorganization of the intelligence community. Today is 
the first of a series of hearings. I hope we have many, not only with 
wise people, but that we also go off by ourselves and talk over a 
period of a weekend on several retreats so that people don't sort of 
get stuck on one particular idea and then become associated with it, 
and feel they have to defend it even if it doesn't stand the test of 
    It's terribly important that we be right on this. And I 
agree very much with the chairman. This has nothing to do with 
whether there's going to be an election or not. It has to do 
with that there has been 50 minus three years of the 
intelligence community and not once has there been really any 
kind of change made by the Congress.
    And that's ridiculous. I mean, you're going back to the 
days when, you know, ships did Morse Code by light to each 
other. And just everything in the entire world has changed. 
However, the organization of the intelligence community has 
    I think, and I know the chairman would agree, that our 
recently-published report in the Intelligence Committee was 
probably the most devastating report on the analytical work of 
the intelligence community ever leading up to the decision to 
go to war.
    When the report's conclusions are added to those of our 
Joint 9/11 Inquiry in 2002, which people forget took a long 
time to do and a large staff, and then the findings of the 
independent 9/11 commission to be released this Thursday, it 
leads to the inescapable conclusion that change is needed to 
address documented shortcomings within the intelligence 
community. And I think we would all agree that the time for 
that is long overdue.
    Timely and accurate intelligence is the tip of the spear 
that protects Americans here and abroad. That is said so often 
that it is a cliche. Often cliches are not listened to very 
carefully and, therefore, aren't paid attention to. People 
better listen to this one. As such, intelligence successes can 
save lives, and intelligence failures, as we now tragically 
know, can result in the loss of life.
    The failures detailed in our past investigations were, in 
some cases, individual failures as opposed to systemic ones: in 
some cases, failures of leadership; and in some cases, failures 
of organizations unwilling to share information or to take 
    Therefore, I believe it's important to approach the call 
for reform with an appreciation that there are not silver 
bullets. There is no instant answer, one solve-all approach. 
And there aren't panaceas when it comes to the improving the 
intelligence community's ability to meet the national security 
challenges of today and in the future.
    It need not be said that the world has changed dramatically 
since the Security Act creating DCI was passed in 1947. I was 
10 years old. The enemy has changed. The threats have changed. 
The technology has changed. One of the key questions before us 
is whether the organizational management structure established 
over almost a half century ago is the right arrangement of 
authority and personnel today. I think not.
    Or to put it another way, if we were given the chance to 
create an intelligence community from scratch today, would we 
end up creating a DCI, Director of Central Intelligence, in its 
current form and with its current limited budget authorities to 
loosely, loosely manage 15 agencies, while at the same time 
being responsible for the daily operations of the Central 
Intelligence Agency? I think it is unlikely we would choose 
such a complex and split organizational setup. But that is what 
we have. If that is a consensus, than the most difficult 
question remains what to do about it.
    I support the idea of centralized authority over the 
intelligence community. The creation of a Director of National 
Intelligence is one possibility. But I am open to ideas on 
exactly how to structure such an office. And, believe me, there 
are many approaches to such a concept.
    And that is not the only thing which I am willing to listen 
to. I think all us, to be fair and to be accurate, have to be 
open to new ideas. There are some incredible books that have 
been written on this. We have to do a good deal of that and a 
good deal of talking with experts as well as with ourselves.
    I think that most observers agree that whoever leads the 
intelligence community needs more authority than the Director 
of Central Intelligence has today in order to get the current 
array of intelligence fiefdoms working together.
    The leader of the intelligence community needs more 
authority over budgets, personnel, tasking of collection assets 
and the appointment of intelligence agency heads. That person 
also needs to be separated from running the daily operations of 
the Central Intelligence Agency.
    We must also further integrate the work of the different 
agencies by creating true joint operations, as the Defense 
Department did through its Goldwater-Nichols reforms, which 
have been declared by all to be thoroughly successful.
    We must find a way to do this without ripping the different 
agencies out of the Defense Department and elsewhere. There are 
legitimate reasons that these agencies are where they are. And 
we don't want to lose the benefits of those arrangements, but 
neither do we want to leave them without being questioned, 
neither do we want to do that.
    It is true that the special operations of the CIA may be 
less flexible--Dr. Odom, you discussed this in your book--than 
does the defense services. Combatant commanders have to be 
responded to immediately. And they cannot sort of go through an 
enormous process unless we can invent one which works. And I 
don't think you think that we can do that.
    We need to continue and to accelerate improvements that we 
have made to our human intelligence collections programs. That 
we agree on.
    In improving analysis, there are some actions that we can 
take right away, including the use of red teams, which is 
analysts whose entire job is to be contrarians and to try to 
pick holes in arguments and to do what could have happened a 
great deal more during these past number of months and also to 
challenge the assumptions of the national intelligence 
estimate. That would be a specific hope on my part.
    One of our biggest challenges is finding a way to insulate 
theintelligence community and its head from the kind of 
political pressure that we may have seen. Intelligence must be 
completely objective, regardless of the past, and beyond the reach of 
politicization. One possibility is a set term for the head of the 
intelligence agency, as we have for the FBI director. It's a relatively 
simple matter. And I think it's isolated the FBI director rather well 
and effectively and gives some comfort to the American people and to 
the people working for the director of the FBI.
    In the end, we need a flexible intelligence community that 
works well during both times of peace and times of war. We need 
an intelligence community that can fight the clandestine war 
against terrorists and, at the same time, support overt 
military operations.
    We should not be timid in addressing the need for reform. 
We cannot dodge the difficult questions and the tough choices. 
We owe it to the families of those who died on 9/11 and those 
who have answered the call to arms in Afghanistan and Iraq to 
rise to the challenge before us.
    Today's hearing is the start of an ongoing dialogue--and I 
hope it is truly that--on the committee, where all viewpoints 
should be presented and discussed. I am hopeful that the 
exchanges of ideas will yield in the end to a bipartisan and 
enthusiastically supported reform that we can send to the 
president for his signature.
    Following Senator Feinstein's testimony on her bill to 
establish a Director of National Intelligence, the committee 
will hear from three very distinguished individuals who are 
eminently qualified to address the future of U.S. intelligence. 
And we're lucky to have them. And I expect we will be badgering 
    I am pleased that they were willing and available to be 
with us this afternoon, et cetera, et cetera.
    Thank you, Mr. Chairman.
    Chairman Roberts. I thank the Vice Chairman for his most 
pertinent comments.
    It is my personal privilege to recognize a very valuable 
member of our committee to discuss her legislation. It is also 
co-sponsored by several members.
    Please proceed, Senator Feinstein.
    [The prepared statement of Senator Snowe follows:]

    [The prepared statement of Senator Feinstein follows:]


    Senator Feinstein. Thank you very much, Mr. Chairman, and 
let me thank you for keeping your word and allowing this 
hearing to take place. It is very much appreciated.
    I also want to thank my co-sponsors--the ranking member of 
the committee, Senator Rockefeller, Senator Lott, Senator Snow, 
Senator Mikulski, Senator Wyden, and Senator Graham. I'm very 
grateful for their support on this matter.
    Several of us on the Intelligence Committee are very 
troubled by what we see as fundamental, structural flaws in the 
community. In short, there are two basic problems, and our 
legislation aims to address both in the most direct way.
    The first flaw is one of leadership structure. Under 
current law, one person holds two separate and critical jobs: 
head of the entire intelligence community and head of the 
Central Intelligence Agency. That may have been acceptable in 
1947 with a vastly smaller intelligence community and a 
simpler, bipolar post-war world. Today, however, the fact that 
we don't have an independent head of all of our intelligence 
assets has become a significant problem.
    To use a nautical metaphor, we have a fleet of 15 ships 
without a full-time admiral. Instead, the captain of just one 
of those ships is trying to run his own crew and oversee all of 
the other ships in the fleet. It just doesn't work. Even if one 
extraordinary person could manage the workload of both jobs, 
they are inherently incompatible.
    Secondly, to the extent current law provides for a leader 
of the intelligence community, the position of DCI is poorly 
equipped to manage and lead the community. It lacks any 
meaningful statutory and budgetary authority. It lacks the 
basic tools to carry out the job. The result of these two 
fundamental flaws is that there's one person burdened with two 
incompatible jobs and without the authorities to do either of 
them well.
    I made these points in 2002, again in 2003, when I first 
introduced legislation to split the DCI away from the CIA and 
replace the one job with two. The Joint Senate and House 
Intelligence Committees investigating 9/11 made similar points. 
I worked with Chairman Graham of Florida. We redrafted my 
legislation, and we introduced it again. And that committee 
recommended its passage. And it is likely that the 9/11 
commission, whose report is due in a few days, will make 
similar recommendations.
    The report recently issued by our committee is a chillingly 
detailed account of failed process and analytic judgments. Let 
me call your attention to one conclusion. It's a conclusion 
entitled ``Conclusion 7'': ``The Central Intelligence Agency, 
in several significant instances, abused its unique position in 
the intelligence community, particularly in terms of 
information sharing, to the detriment of the intelligence 
community's prewar analysis concerning Iraq's weapons of mass 
    Our findings go on to say, and I quote: ``The CIA in 
several significant instances abused its unique position in the 
intelligence community. The fact that the DCI is head of the 
CIA and the head of the intelligence community, the principal 
intelligence adviser to the President,and is responsible for 
protecting intelligence sources and methods, provides the CIA with 
unique access to policymakers and unique control of intelligence 
reporting. This arrangement was intended to coordinate the disparate 
elements of the intelligence community in order to provide the most 
accurate and objective analysis to policymakers.
    ``The committee found that in practice, however, in the 
case of the intelligence community's analysis of Iraq's weapons 
of mass destruction programs, this arrangement actually 
undermined the provision of accurate and objective analysis by 
hampering intelligence sharing and allowing CIA analysts to 
control the presentation of information to policymakers and 
exclude analysis from other agencies.''
    Now, Members, if you'll recall, every time there was a 
difference of opinion between agencies, it was the CIA view 
that prevailed.
    Now, our approach is relatively straightforward.
    First, we turn to the problem of leadership structure. 
Since we've identified the key problem as being the fact that 
two very different jobs are held by one person, we redrafted 
the operative sections of the National Security Act of 1947 to 
split those two jobs into two positions.
    Secondly, because we also recognized that the head of the 
intelligence community needed more authority to properly 
coordinate activity within that community, we changed those 
authorities and responsibilities which need changing.
    Primary among them is the lack of meaningful budget 
authority. Today, the DCI has only limited budgetary and 
management authority over the myriad agencies that include the 
CIA, the Defense Intelligence Agency, the National Security 
Agency, the National Reconnaissance Office, the National 
Geospatial-Intelligence Agency, Army, Navy, Air Force and 
Marine intelligence, State's intelligence, Treasury's 
intelligence, Energy's intelligence, Homeland Security's 
information assessment, FBI and the Coast Guard.
    Together, these agencies make up a huge network, tens of 
thousands of employees and a significant secret budget. In 
practice, the DCI currently only controls the budget of the 
CIA; 80 percent plus, of the intelligence budget is under the 
control of the Secretary of Defense. This is untenable if we 
want a true leader of the entire community.
    Secondly, the lack of effective personnel authority further 
hamstrings the current structure.
    And thirdly, the lack of staff and resources to really lead 
the community also prevents effective management and control.
    The result of our changes is a package that combines 
leadership structure with statutory and budgetary authority, 
but leaves room for the detailed change and reform which will 
be needed in the coming years. Those changes will be the number 
one job for the first Director of National Intelligence.
    We are in the process of vetting this legislation, the bill 
language, with a number of different individuals, past and 
present, who have played significant roles in the intelligence 
community. And I believe we will have a substitute amendment 
when this bill comes to markup.
    Suffice it to say that the current structure of our 
intelligence community is a relic of last century's conflicts. 
It is a Cold War solution to Cold War problems. The structure 
dates to 1947. And this is important to understand. Our 
adversary was different back in the days of spy vs. spy, CIA 
versus KGB, U.S. versus Soviet Union. In a bipolar world in 
which the task was to anticipate and track armies, tanks, 
planes, governments, our system worked. And I believe it worked 
    But it's unsuited for our current world of asymmetric 
threat, fast-paced changes and a shadowy and brutal adversary 
in a different culture with heavy language issues.
    In many ways, the old adversary, Communism in its many 
forms, was a distorted mirror image of ourselves with similar 
tactics, weapons and structures. But the new adversaries, 
amorphous terrorist groups, proliferators and rogue nations do 
not fit that image, and our intelligence services must change.
    Secondly, much has happened here at home. The 21st century 
intelligence community is much larger than it was in '47. With 
the addition of National Reconnaissance Office, the National 
Geospatial-Intelligence Agency, the expansion of the National 
Security Agency to deal with an exponentially larger 
technological world, the intelligence community is much more 
complex than it was even a few short years ago.
    The DNI we propose would be responsible for leading this 
entire community, working within an independent office which 
could be housed in an agency, or separately, aided by a deputy 
director of national intelligence and equipped with meaningful 
budget and personnel authority. This director should and can 
provide focused, independent and powerful leadership and 
management the intelligence community badly needs.
    The DNI would be responsible for all of the functions now 
performed by the DCI and in his role as head of the 
intelligence community. A separate individual would be director 
of the CIA, and he would retain his role and its role as the 
central analytic element of the intelligence community and the 
lead agency for human intelligence.
    Nominated by the President, confirmed by the Senate, the 
DNI will be a member of the cabinet and address such important 
issues as assessing the balance between expensive technical 
collection platforms, such as satellite systems, and human 
source collection and analysis. In this new world, how much of 
a role do we have to place on large satellites tracking armies 
across continents versus heavy penetration of human 
intelligence? That's a balancing role. Somebody needs to 
evaluate it.
    Second is developing mechanisms to enhance our ability to 
collect foreign intelligence within the United States and 
setting the priorities and strategies in a new nonstate 
asymmetric world--in other words, someone that determines the 
scope of a mission across the entire community and sits down on 
a daily, weekly and monthly basis with the managers of those 
separate agencies and holds them accountable to specific goals.
    Third is evaluating and implementing a human intelligence 
capability with language and cultural knowledge in critically 
important new and growing areas and, most importantly, 
reforming a broken analyticprocess to ensure effective peer 
review, red teaming and adding analytic integrity to prevent the use of 
false intelligence in policymaking.
    The new DNI would not only have the statutory and 
structural position of leadership, he or she would have the 
authorities and tools necessary to accomplish these 
    First, the legislation makes substantive change to the 
current budget authority now vested in the DCI. The DNI would 
have clear authority to formulate and execute the budget. 
Spending would be under his control. He would move funds and 
people between agencies and accounts, subject to congressional 
oversight and in coordination with, but not subject to the 
control of, the Secretary of Defense.
    In addition to these authorities, the new Director of 
National Intelligence will have the staff to carry them out, 
including a community-wide general counsel to advise and assist 
in setting an implementing policy and ensuring compliance with 
law; a community-wide inspector general to guard against fraud, 
waste and abuse; a full staff based on what is now the 
community management staff of the DCI to assist him; a set of 
deputies including ones for administration, collection, 
community management and analysis to assist in making the 
community work together, breaking down the stovepipes; and, 
finally, directing control of the National Intelligence Council 
to ensure that community-wide intelligence products, such as 
the national intelligence estimate, are really community 
products and not biased on the product of CIA dominance.
    I recognize that this bill will certainly not solve every 
problem within the intelligence community, but I believe it's 
an important, even critical, first step.
    Let me add that none of the provisions are sacrosanct. I 
certainly am open to change, and I believe our co-sponsors are 
as well. My earlier legislation, first introduced in 2002, was 
intended to begin the conversation. It's not meant to be a 
final word. So we're open to any thoughts or ideas that members 
of this committee or other members of Congress, such as 
Congresswoman Harman, or intelligence community experts, such 
as Mr. Kindsvater, may have, as exactly how we construe this 
position. The goal is to make sure that we have the best 
possible intelligence community under the best and strongest 
    Finally, in summary, the DNI would determine, manage, and 
carry out the scope of a mission throughout the entire 
intelligence community, break down the stovepipes, set a 
structure and methodology for communication across the chain of 
command, and be responsible to see that collection and analysis 
of dots reached the most accurate product possible.
    The bottom line is that leading the United States 
intelligence community is a full-time position. And if it's to 
be done right, we cannot expect the person holding that 
responsibility to run a separate agency simultaneously. It's 
time to put somebody in charge of the entire intelligence 
community and give that person the budgetary and statutory 
authority to accomplish the job. Unity of command and tools to 
do the job are critical for the tasks ahead.
    I want to thank you very much, Mr. Chairman, and members of 
the committee.
    Chairman Roberts. Well you're certainly welcome, Senator. 
And thank you for your very comprehensive statement.
    I have three questions. I'll try to make them quick, 
because I know we have a full membership and they do want to 
ask questions.
    First of all, your bill makes an effort to give the DCI 
certainly much greater authority to set and enforce policies 
for the entire intelligence community. And we all see where 
this would be necessary, setting a community-wide standard for 
information sharing, for competitive analysis and even 
information technology, standards that would all help address 
the problems that both the Iraq review and the Joint Inquiry 
have identified.
    Now, the question is how the DCI doesn't have that 
authority now. For example, the DCI, just two weeks ago, 
promulgated DCID 8/1. I'm not sure how many members of this 
committee were even aware that two weeks ago that that 
    This new directive sets very high-level policy on 
information sharing for the entire intelligence community. It 
may be a day late and a dollar short, but clearly the DCI can 
set policy.
    So what about enforcement? What if, for example--and I'm 
being the devil's advocate here, and this is entirely 
hypothetical--say the NSA refused to share information under 
the directive. The DCI is not helpless under current law. He 
could modify the NFIP budget request for the NSA. He could 
penalize them for not complying. He could call the Director of 
the National Security Agency or even the Secretary of Defense 
and say, ``Hey, get in line.''
    Your bill does not give the DNI any direct operational 
control of all the 15 intelligence community elements 
equivalent to what the DCI has now over the CIA. So even if 
your bill was enacted, it seems to me like the DNI is in the 
same position as the current DCI on policy enforcement.
    Would you cut the budget or call the department head? What 
are your thoughts on this latest development?
    Senator Feinstein. Well, my thoughts are, first of all, the 
intent is to give the DNI the authority. And in reading a CRS 
analysis of this, I believe it does. We would welcome comment 
if there is additional language to strengthen it further.
    Chairman Roberts. So you would give the DNI the direct 
operational control of all 15 intelligence community elements?
    Senator Feinstein. That's correct. Now, clearly, the heads 
of the various agencies are going to run their own agencies. 
But just like anybody else, there has to be some central 
directive to management. I don't know how we can get 
intelligence in this new world we're in unless we have it, 
unless we understand what the mission is across the community, 
and then see how that mission is going to be carried out.
    One of the things that bothered me very much as a member of 
this committee was seeing how, every time there was a 
difference of view, the CIA view prevailed against the views of 
the intelligence components in other agencies.
    Chairman Roberts. Okay. Very quickly, I want to talkabout 
personnel. This issue, there are certainly few, if any higher, 
priorities for the intelligence agencies other than support for the 
    Under current law, the DCI can transfer personnel from one 
intelligence community agency to another for periods of up to a 
year only if the head of the department which contains the 
affected element or elements of the intelligence community does 
not object to such a transfer. Under the proposed legislation, 
the DNI can transfer personnel and the ability of the head of a 
department to object to such a transfer is simply eliminated.
    I am told that the DNI's new authority would not apply to 
personnel funded under the Joint Military Intelligence Program 
and the Tactical Intelligence and Related Activities accounts. 
However, under the proposed legislation, the DNI could, over 
the objection of the Secretary of Defense, transfer military 
personnel funded under the National Foreign Intelligence 
Program from a DOD combat support agency and their forward-
deployed elements to positions that would support strategic 
intelligence requirements.
    Is this possible under your proposed legislation? And, if 
so, why is that a good thing?
    Senator Feinstein. Well, there are many different questions 
wrapped up in that. And I certainly think, you know, what is 
existing in the field in a different country in a military 
organization is somewhat different. But the thrust of the bill 
is clearly to say to the Secretary of Defense: You detail 
people to these missions. And the mission is essentially 
controlled by the director of national intelligence.
    That's the thrust of this bill.
    Chairman Roberts. When we had--and I apologize to my 
colleagues--but I wanted to bring this up in regards to Mr. 
Fingar and Mr. Kindsvater, and we've reached agreement with the 
Vice Chairman that we can speak of this and that it's 
unclassified. And we had Mr. Fingar up in regards to being the 
Director of INR--i.e., the arm of intelligence of the State 
Department--and also Mr. Kindsvater regards to the Director of 
Community Management with the DCI.
    Both indicated some concern about being torn between two 
masters. In other words, if you have an event that pops up like 
say the USS Cole, you have the captain of the Cole who could 
certainly back that ship out of there and should have. He has, 
you know, unique control of that mission. That would come from 
Central Command, Secretary of Defense, the intelligence from 
the DIA, which it did not, and should have.
    And how does the DNI work into that? Or, say, if you had 
the embassy bombings in regards to Africa, does that analyst, 
do those personnel do they work for Secretary Powell or whoever 
is Secretary of the State, or do they work for the DNI?
    Senator Feinstein. One of the things about this place, if 
you will, is territorial imperative. You know, whether it comes 
to inner servicing of bases or anything else, everybody has got 
to have their own with a fence around it. And I don't see that 
solving the problems of this new world.
    I think common sense prevails on many of these questions. 
The movement of the Cole clearly would be a military decision. 
What kind of intelligence to set up, I think the DNI should 
have a role in that.
    But I think when you break it down into the specific 
happenings in a combat theater, clearly the military would 
prevail. But we're setting basic parameters of intelligence 
collection among 15 different agencies. And it seems to me 
somebody has to manage that process, because under this very 
disparate, very different, very stovepiped systems, we're not 
getting the intelligence.
    Chairman Roberts. Senator Rockefeller.
    Vice Chairman Rockefeller. Thank you, Mr. Chairman.
    A couple of quick questions. The question of the Cabinet. 
Your bill proposes that the President's principal adviser on 
intelligence matters be, through a Sense of the Congress, 
should be a Cabinet-level officer of the United States.
    Senator Feinstein. Correct.
    Vice Chairman Rockefeller. Now that's very important and 
very interesting because what we are looking for is the non-
politicization in an increasingly either political through 
the--I don't think the Republican-Democratic sense so much--but 
turf sense, which you have referred to in your statement 
several times, that if you put somebody in the Cabinet, there 
is ingrained throughout our history a sense of loyalty to the 
President of the United States. In other words, people always 
rise when he comes in and--or she--whatever.
    Bob Mueller doesn't sit on that. And you know, Bob Mueller 
is a pretty brainy, efficient guy. I'd like to know what your 
sense of the balance is between having somebody with the symbol 
of being in the Cabinet and therefore purportedly having the 
ear of the President or being able to interrupt Cabinet 
discussions--which I guess aren't that regular--as opposed to 
the loss of perceived independence that he or should would have 
as a result of being on the Cabinet.
    Senator Feinstein. Thank you, Senator.
    Look, we live in a certain world. And the key is how the 
individual functions in the world. You can give the individual, 
if you choose, a 10-year term and make it a term appointment or 
you can make it a pleasure appointment--either way.
    But my view is that seminal to protecting against another 
attack on our homeland, seminal to be able to defeat the myriad 
of terrorist groups that are growing across the world, is 
someone that can really permeate the entire culture with what 
we need to know, the culture of the executive branch, that no 
longer is a President going to be the victim of a daily 
Presidential brief or a one-page national intelligence 
estimate. There is somebody right there with him all of the 
time that has the full feel and scope of the entire 
intelligence community.
    I think that is real important. I mean, what all of us know 
is happening out there, even in our own country, we've got to 
change the structure. And it really bothers me that you have--
sure, you can have a DCI send out a policy brief, but what I 
have found is that most of these agencies are pretty remote 
from what we do. They're pretty based in their own kind of 
bunker mentality. And I think having that strong manager 
thatholds everybody accountable and has the power to do it, I think it 
can break through in a number of different areas.
    At the very least, we should try it.
    Vice Chairman Rockefeller. And that leads me to my second 
question, which is also philosophical. There are kind of two 
schools of thought, and you are in one of them, I believe.
    Senator Feinstein. Correct.
    Vice Chairman Rockefeller. And I think you may be right. 
That is, one looks in Congress and if anybody is culpable, we 
certainly have to share in that, because if you talk about turf 
wars, we surely have them--what is it, 56 people that oversee 
intelligence? Sort of ridiculous--is that you either decide 
you're going to put forward a bill which challenges the 
traditions of Congress and the way things are done here; that 
is, let's say a top down, the way a corporation could do it 
through a vote of, you know, its 12 members. And it would 
become the operation of the corporation. That's why you have 
CEOs. It does make sense.
    In government, things get very much more complicated.
    So you have these sort of two decisions. One is, well, I 
don't think we can ever get that 80 percent of the Defense 
Department. And so, if we put up a bill which is predicated to 
do that in one way or another or to get parts of it or 
whatever, that they will fight it, that the President will back 
up the Secretary of Defense. And that it will go nowhere.
    And the other is, well--and we learned this from the 1991, 
1992 health care fights--you try to do too much and it fails, 
and so you then go incremental. When you go incremental, you 
kind of lose the attention of the Congress and the American 
people and the folks that report on such, and you don't get 
much done.
    So the question I would have for you is, it may be--and I 
think this is your thinking--that post-9/11, we've got to go 
for the whole thing and in a sense put all of those people who 
will judge us in the Congress and those who surround the 
President and finally the President online, that we need to 
make an unpopular, not entirely defensible, but wholly 
different change, dramatic change, and that this is the only 
time we'll be able to do that.
    And the other school is that that is where you start and 
then you come back from that a bit because ultimately what 
counts around here is making changes which improve the safety 
and security of the American people. And if that isn't the 
whole loaf, then surely it should be part of the loaf.
    And I'm just interested in your reaction to that.
    Senator Feinstein. Well, I happen to agree with you in 
that. You know, we have changed our laws. Senator Hatch, 
Senator Durbin, Senator Chambliss, Senator DeWine, I, we all 
sat for weeks reviewing the PATRIOT Act, making substantial 
changes in that Act, brought about by the need to change law 
for this new world that we're in, to try to bring it to the 
terrorist cells and be able to mix it up and get the knowledge 
that we need to prevent another attack on the homeland.
    But we haven't really changed the Cold War mentality that 
determined how these agencies grew. If I were President, 
Republican or Democrat, I would want to have a powerful head, 
one head, of the agency. I'd want him sitting at my right hand. 
I'd want him advising me all the time. I'd want him to talk at 
a Cabinet meeting and let departments know exactly what's 
happening, the things that he can share.
    And we don't have that today. And I don't understand how a 
George Tenet could run an agency like the CIA, be the main 
adviser to the President, come down and talk to us and still be 
head of the intelligence community. To me, there isn't enough 
hours in the day to do it right--to hold the meetings, to sit 
down with the managers, to say a month ago I gave you this 
charge. I haven't heard. What are you doing? Why aren't you 
doing it?''
    I think we've got to mix it up big time in this new world 
that we're in.
    Chairman Roberts. Senator Levin.
    Senator Levin. Thank you, Mr. Chairman.
    Senator Feinstein, first of all, let me thank you for your 
commitment, extraordinary testimony and knowledge. You really 
bring us a breadth of knowledge here which is important, and a 
willingness to shake things up. That doesn't mean necessarily 
we should shake them up in your direction, but a willingness to 
do that, if it's called for, is essential here.
    It seems to me that some of our goals ought to be as 
follows. One is independence. I do think we want somebody who 
will call them as he or she sees them in terms of intelligence, 
free from pressure from policymakers to support whatever the 
policy is, but a willingness and an ability to say this is all 
the intelligence that's available to us from all sources, and 
this is what we see.
    How would somebody who is a DNI, who is in the Cabinet and 
therefore rubbing elbows with policymakers every day, every 
Cabinet meeting, be more likely to be independent in terms of 
the decision as to what that intelligence says than would the 
CIA director?
    Senator Feinstein. Because by statute we would make him 
that independent. In one of the earlier drafts of this, this 
was a term appointee. I think some people felt it should not be 
a term appointment. It could be.
    I think we have to remember that a President appoints the 
Supreme Court independent, appellate court judges. We sit in 
the Judiciary Committee. We agree with it or we don't agree 
with it. We take a vote on the floor of the Senate. We approve 
it or we don't approve it.
    But that's kind of our world. And I think the ability to 
mobilize an administration, to have a clear chain of thinking 
across these very departments--you know, we haven't had 
Homeland Security, this humongously large department, for very 
long. If you have a Secretary of Homeland Security added to the 
Cabinet, it seems to me that somebody that heads 15 agencies 
ought to be there, too, because these agencies are increasingly 
vital to the survival of our country.
    Senator Levin. Your argument, then, is that somebody who 
sits with policymakers day in, day out and hears what they want 
in terms of policy is more likely to be able to call them as he 
or she sees them on intelligence matters than somebody who is 
separated and iscalled in to provide independent intelligence. 
That's your experience as a political figure.
    Senator Feinstein. No, actually it's a little different. I 
see this individual advising the Cabinet. I don't see the 
Cabinet advising this individual. What I want and what I hope 
we've created--and if we haven't we should strengthen it--is 
giving this official all of the authority they need to really 
be the definitive figure.
    Senator Levin. In addition to the independence issue, 
independence from the political pressure or political decision 
or policy decision, I think we want somebody who can consider 
all of the intelligence fairly and in a balanced way.
    And I think your point that the CIA in almost every case, 
as we analyzed it, favored a CIA position over the other 
components of the intelligence community, is an accurate and a 
perceptive comment about the current shortcoming of the current 
structure to be able to consider all of the pieces in a 
balanced way.
    Another goal that we have relates to who is going to be 
making decisions on allocations of resources. And this is an 
issue which particularly concerns me as a member of the Armed 
Services Committee.
    We've got commanders out there who need capabilities and 
need them quite promptly. They've got requirements for 
intelligence collection. You've got people, let's say, in 
Afghanistan or Iraq who need those airplanes. And you've got an 
intelligence DNI who says, no, we need those unmanned aerial 
vehicles, those UAVs, for our borders right now. And sorry, I 
look at the whole picture, the big picture, and we're going to 
take the UAVs we've got and put them on our border. And you've 
got a commander out there in Afghanistan who says, we're after 
bin Laden. We need those extra UAVs. And we need them now.
    Senator Feinstein. Well----
    Senator Levin. You, as I understand it, would give the 
ultimate decisionmaking, obviously, to the President. But 
putting that aside, you would give that DNI the correlation of 
all the resources. And that person sits at the elbow of the 
President and gives the decision as to how we would use those 
resources. And I think there's going to be some real 
nervousness in terms of some combat capabilities as a result.
    Senator Feinstein. I don't believe for a minute that this 
isn't a President's decision in that kind of a conflict--well, 
we can get bin Ladin, he's near Peshawar or here, and we can 
pick it up from the Khyber Pass or whatever it is, versus this.
    Senator Levin. If he knows about it.
    Senator Feinstein. And it might happen around a Cabinet 
meeting. I don't know where it would happen. But clearly in 
these kinds of decisions, the commander in chief is going to 
play a role.
    Senator Levin. If he hears about it, but you're giving----
    Senator Feinstein. And I don't think that's unhealthy, 
    Senator Levin. Okay. That is a concern.
    Senator Feinstein. I understand.
    Senator Levin. Thank you.
    Chairman Roberts. Senator, I apologize. I've been giving 
Senators seven minutes, but we have 11 Senators that would like 
to comment.
    Senator Levin. I'd like to apologize for going over my 
    Chairman Roberts. No, you shouldn't apologize.
    At any rate, if all Senators can be aware of the timeframe 
of five minutes. I apologize for that, but we do have a 11 
Senators that wish to as questions, and we have a very 
distinguished panel. And we do want to get to that.
    Senator Chambliss.
    Senator Chambliss. Thank you, Mr. Chairman.
    Senator Feinstein, thank you for bringing this issue to a 
head and for giving it your usual thought process that you 
always bring to the table.
    What has bothered me about the concept of the DNI from day 
one is that we avoid creating another level of bureaucracy that 
is simply one more filter between raw intelligence and what 
gets to the President of the United States, and also how that 
information is used and acted upon.
    As I understand what you have said your bill does, it 
creates a Director of National Intelligence and makes agencies 
now that may be responsible to the Secretary of Defense 
responsible to the Director of National Intelligence. It makes 
the director of CIA responsible to the Director of National 
Intelligence and probably these other 20 or so agencies that 
have some piece of the intel community responsible to that one 
    Now, that in and of itself, it bothers me a little bit that 
I think we do create another level of bureaucracy. But let me 
ask you this. Why wouldn't you just simply give the Director of 
Central Intelligence more power and authority, make that office 
the true director of national intelligence--if that's what you 
want to call him--move the domestic intelligence obligation 
from the FBI into the CIA, remove from the DIA any duplication 
that may be ongoing over there under the Director of National 
Intelligence or Director of Central Intelligence, whatever you 
call him?
    Why couldn't we coordinate all of these entities that 
currently gather intelligence under the CIA, give the Director 
of CIA that budgetary authority which you're alluding to--and I 
think that's critically important and I think you've hit that 
right on the head--and require that person to be responsible 
rather than creating another level over and above the current 
director of CIA?
    Senator Feinstein. I spent a lot of time reading our 
report, you know, particularly before it was redacted. And if 
you look at how judgments were put together, for example, the 
aluminum tubes where the Department of Energy analysts who 
really would be in a position to know about aluminum tubes for 
gas centrifuge enrichment of uranium and to hear that the 
Energy Department said that they were not suitable for that 
purpose, and yet the CIA view prevails.
    To hear that with respect to the unmanned aerial vehicles 
where those who would have the best knowledge, like the Air 
Force intelligence, who said they were most likely to be used 
for targets or aerial reconnaissance, and then to have the CIA 
view prevail, indicated to methat to have one person doing both 
of these jobs doesn't work.
    If the proof is anywhere, it's in this pudding. Because 
every time there was a disagreement, and there was no real way 
to reconcile these differences, so what we got was a document 
that was not accurate.
    And I think if you put every agency on fair and equal 
footing and you have one skilled manager--now, it is true a lot 
will depend who and the talent and the determination of the one 
individual--but assuming there is such an individual out there, 
and I believe there are probably more than one, then it's 
really going to be his management in this difficult world and 
the degree to which he brings real expertise in how to go about 
it. I just don't think you can do both jobs. And I think the 
NIE was the conclusion that said you can't do both jobs in this 
new world.
    Senator Chambliss. Under your bill, who would brief the 
    Senator Feinstein. The DNI.
    Senator Chambliss. Thank you, Mr. Chairman.
    Chairman Roberts. We thank you, Senator Chambliss.
    Senator Bond.
    Senator Bond. Thank you very much, Mr. Chairman.
    I'm going to try to give a little bit of time back. I 
appreciate very much the thoughtful presentation that the 
Senator from California has made. I think we've raised a number 
of very important issues. And I have some real problems saying 
that we ought to start out with going to a totally new 
structure with a DNI.
    Having had some experience running a much smaller executive 
branch, I find that the power usually goes to the people that 
have the bureaucracy. And with the Department of Defense being 
a very large part of this operation, I might say that a 
simplistic solution would just be to make the Secretary of 
Defense the head of all intelligence, because that person has 
the resources.
    I don't think that flies. I am concerned about trying to 
bring together all of the different agency intel expertise 
under a separate heading because I think there may be 
intelligence that is very important in the entity in which it 
resides. I think the Defense Department would say if all of 
their intelligence assets were directed by a Director of 
National Intelligence, they would be very much concerned about 
the flow of information from the tactical to the strategic on 
their side.
    So I appreciate your bringing it forth. I hope we don't go 
take one particular step. I think we need a lengthy review of 
the entire process, weighing all these issues.
    And, Mr. Chairman, if either you or the Vice Chairman can 
carry a tune, it is Senator Mikulski's birthday, but I won't be 
so bold as to begin the singing.
    Chairman Roberts. That information was classified, Senator 
Bond. [Laughter.]
    At any rate, the chair would like to recognize the birthday 
    Senator Mikulski. Well, thank you very much, Mr. Chairman. 
Certain aspects of that are classified. And if anybody reveals 
it, they'll have to report to me.
    But, first of all, Mr. Chairman, I really want to thank you 
for convening this hearing. I hope that it will be the first 
step of many, but you've really responded to our call for not 
only reform, but for transformation. So thank you very much as 
well as our ranking member, Senator Rockefeller.
    I'm going to come to Senator Feinstein in a minute, but I 
want to note our other panelists. And if I could, just an 
experience with General Odom. When I was a freshman Senator, I 
visited the National Security Agency, located in my own state, 
with regularity--it was during the Cold War; it was during the 
Reagan administration--but so that I could really be briefed on 
the issues related to national security and ending the Cold 
War. General Odom was a great teacher.
    And I look to him now, as we did then. And the service that 
he did for this young Senator at the time, really I think will 
be very valuable to us.
    Senator Feinstein, of course you're known for your 
thoughtfulness and your due diligence about the way you go 
about legislation. And I think you're onto something here, and 
happy to be a co-sponsor.
    Here are my questions, because I think we'll all agree, we 
don't want more bureaucracy. We've had a dismal experience with 
the drug czar concept. But in the actual, day-to-day operations 
of a DNI, how would you see this interacting with the Secretary 
of Defense and particularly a big agency like the National 
Security Agency or NRO? Or is he or she just going to be 
someone trying to blow a feather across the table?
    Senator Feinstein. Thank you very much, Senator.
    In the first place, I know one of my own newspapers, the 
Los Angeles Times, had an article, ``Feinstein Touts 
Intelligence Czar.'' And it just makes me shiver, because 
there's a huge difference between what we're proposing and a 
drug czar.
    The drug czar had no real authority. You and all of us know 
that. It was an office, but it didn't have the budgetary and 
statutory authority that's really required to have any degree 
of clout here. And we have tried to change that.
    You see, I guess I look at governmental entities as having 
to be managed, and that management is all important. Each one 
of these agencies would have their own head, but the heads 
would meet periodically with the DNI, strategize, prioritize, 
discuss, have the interchange of information across the 
spectrum of agencies.
    And I really think that's important. In this business, you 
know, we're all very proud that the stovepipes are being broken 
down, but I would suspect that every one of us has come upon a 
situation recently where there were stovepipes and where 
information wasn't conveyed. And a lot of it is explained in 
this report.
    So I think if you get somebody that's going to actively 
manage the agencies, that that will go a long way to delivering 
a better intelligence product.
    Let me make one other quick point, and I can't go into this 
inany detail. But we all know when we do our budget and we look 
at satellite programs, that one or the other of us have questions about 
one or the other program. You've got to have somebody that crosses all 
of this and has the knowledge to take a look at what's in the pipeline, 
what should be followed through with, what shouldn't, what adjustments 
should be made, because the world we're in now is different than the 
world before 9/11. And I'm very concerned that, you know, these are 
tens of billions of dollars items, and we sure as heck better know what 
we're doing.
    Senator Mikulski. Well, thank you. I think that clarifies 
some questions.
    Thank you.
    Chairman Roberts. Senator Hatch.
    Senator Hatch. Well, I want to thank you for your 
thoughtfulness and the efforts that you've put forward. And I 
give you a lot of credit for that. However, I have a lot of 
difficulty seeing how this is going to change very much. And I 
also--having been here 28 years, I've seen great CIA directors 
and some who haven't been so great, but all of whom have tried 
and all of whom deserve credit for their devotion to our 
    I'm just very concerned that if we create another layer of 
bureaucracy, we may just be not getting anywhere at any time. 
Most people didn't realize that the CIA director only has about 
20 percent of the budget and yet has 100 percent of the 
criticism that comes. And I think a lot of people have come to 
the conclusion, too, that part of the problem has been through 
the years Congress itself--continually critical, but never 
really giving the backing that the whole intelligence community 
needs in so many different ways.
    And there's a risk aversion, in my opinion, that has arisen 
out of that that makes the bureaucracy even more bureaucratic.
    And I know that's what you're trying to overcome, in part. 
And I respect you for that. I'm going to really study this and 
hopefully we can arrive at a conclusion that will help our 
whole intelligence community to work better than it ever has 
    But I think we ought to re-evaluate Congress, too, and our 
role, and see if we're part of the problem. I can just 
guarantee you we are. And see if we can find a way of having 
the Intelligence Committee work in an energetic and efficient 
and effective way, compared to some of those times when it 
hasn't been as energetic, efficient or effective.
    But in any event, I want to thank you for your 
thoughtfulness and for your desire to make things better. And 
to the extent that I can help, I certainly will try. But I'm 
not sure that this is going to make the intelligence community 
any better. A lot depends on who's picked for these current 
positions and how well they manage those current positions.
    Like I say, I've seen them managed very, very well. And 
I've seen them where they haven't been managed quite as well. 
But all have been dedicated in their efforts and have tried.
    Thank you, Mr. Chairman.
    Chairman Roberts. Senator Durbin.
    Senator Durbin. Thank you very much.
    Senator Feinstein, thank you for your contribution to this. 
And I think most of us believe that since 9/11, as we've looked 
at the intelligence community and capacity, with our reports 
and others, that the wheels have changed and the wheels of 
reform grind exceedingly fine and exceedingly slow.
    And even our best intentions, creating new departments and 
trying to merge cultures, trying to bust up the stovepipes, I 
continue to feel great frustration that there's still so much 
more to be done, even in light of 9/11, which leads me to my 
next concern.
    What we're talking about with new legislation is building 
boxes, moving boxes, on the organization chart. And I'm 
wondering if, at the end of the day, even if we come up with a 
rational approach for reform and for restructuring the 
governance of intelligence, whether we still don't have a 
couple very fundamental problems.
    And let me point to one. This new Cabinet-level person, 
this DNI, is a person who will have to bring some extraordinary 
talent to the job. We have usually kept the director of the CIA 
in the shadows of the highest levels of government, with the 
exception of President Bush's father. I asked the staff to 
think back to someone who was clearly a political person who 
ended up in that position, and there haven't been a lot of 
them. Some of them may have had political skills, but may not 
have had the personal political experience.
    Most of us feel that we want to keep politics out of this 
equation. We want the CIA to be objective, dispassionate, 
honest, accurate, even in the face of political opposition. And 
yet it's hard for me to imagine someone sitting at that Cabinet 
table who wouldn't have political skills or need them to 
survive in that atmosphere.
    And I'm wondering if we're not creating here a tension, 
between professionalism and political skills in the person that 
we're seeking by going to a Cabinet-level position. That's one 
of the things that crosses my mind, and I'd like your thoughts 
on it.
    The other thing I would note is I'm just about finished 
with the biography of Sargent Shriver--which I recommend; it's 
very interesting--who with the ear of first President Kennedy 
and then President Johnson was given extraordinary 
responsibilities--create the Peace Corps and let me know when 
it's ready to go; create the war on poverty and tell me when 
you're ready to launch.
    And he brought exceptional skills, a lot of luck and a lot 
of trial and error before it finally worked. And even with the 
backing of the president, it was hard to create these agencies 
anew, fighting with existing agencies and the like. It just 
reminds me again of how much we are vesting in the person that 
we're looking for, what kind of skill set.
    The last point I'll make--and I think Senator Hatch 
referred to it--I think before we go looking for the enemy, we 
ought to look internally when it comes to reform. We need to do 
an extraordinarily better job here in oversight. We need more 
staff, more resources. This is a very demanding committee, but 
a very rewarding committee, and if we do our job right, it 
protects this country. And I hope that as we talk about 
reforming the intelligence agencies, we won't forget our own 
need toreform from within.
    But could you address this idea of the Cabinet-level 
person, political skills, and nonpolitical nature of the job?
    Senator Feinstein. I think whomever it is, even if you 
don't pass this, a new DCI will be appointed. And he will be 
appointed by a President. And that President will have a 
relationship with him, not necessarily other departments, but 
that President will.
    And the problem that various Senators are speaking of, 
including yourself, Senator Durbin, is inherent in the present 
situation. What the beauty of this is, not another layer of 
bureaucracy, but a strong central manager who has clout, who 
doesn't have to depend on wheedling his way one way or another. 
But he's got the budgetary and the statutory authority.
    He can come before us. He's a part of the Cabinet. He mixes 
it up. I mean, I just couldn't comprehend him being used in 
this situation. And if he were, I don't think he'd last very 
    So my vision is a very much strengthened figure, but a very 
knowledgeable figure. I mean, I happened to have liked George 
Tenet very much. I think he was smart. I think he was good at 
what he did. But to have said the intelligence, if in fact he 
said say that, was a slam-dunk either showed me he didn't know 
what it was or some very bad decisions were made.
    For example, how can you put the Secretary of State out 
before the world on mobile labs with four discredited sources? 
That alone, to me, is enough to change the whole bloody system. 
Before the world, at the United Nations Security Council--that 
can never happen again.
    Now this is, you know, a four-star general, who has an 
impeccable career, whose integrity and credibility is a 10 on 
every scale. How can this happen? But it did.
    And so what I want to do is see that doesn't ever happen 
again, that there is the red-teaming, the peer review, that 
agencies reach out, that there's adversarial discussion of 
intelligence and that we don't look like the gang that can't 
shoot straight.
    Chairman Roberts. Senator Snowe.
    Senator Snowe. Thank you.
    Chairman Roberts. Shooting straight.
    Senator Snowe. I'll try.
    Thank you, Mr. Chairman. And I too, want to thank you as 
well, for holding this very critical hearing, which I know will 
be one of many as we begin to evaluate the need for reform and 
the types of changes that are absolutely essential.
    And I couldn't agree more with my colleague, Senator 
Feinstein. And I want to applaud you for your longstanding 
leadership in spearheading this initiative for major reform. 
And I'm delighted to be able to join you in this effort. And I 
thank you for what you're doing in underscoring the fundamental 
need for a major overhaul and restructuring of our intelligence 
    I certainly was left with the inescapable conclusion, after 
we went through our year-long investigation on the prewar 
assessments concerning Iraq and stockpiled weapons of mass 
destruction and postwar realities, it left me with one final 
thought, and that is that we needed real reform within the 
intelligence community to function as a community.
    And the fact of the matter is, we need a Director of 
National Intelligence to provide the coordination and the 
information sharing, the analysis, in a systematic and 
synchronized way and not in an ad hoc fashion.
    I believe the DNI will facilitate an atmosphere of 
objectivity, connectivity, information sharing. And as you 
mentioned, some of the startling facts that were revealed as a 
result of our intelligence investigation, one is, there was not 
the sharing of information concerning the credibility of the 
biological weapons analysis and the sources that were used for 
that assessment; it wasn't shared with other analysts in other 
    We found that the Director was not informed of the 
dissenting opinions in other agencies with respect to the 
aluminum tubes. The fact is, that debate started internally 
back in 2001. It was almost a year, if not more, in 2002, that 
the Director became even aware of those conflicting opinions.
    And again, when you're talking about the CIA Director, who 
is the principal intelligence adviser to the President of the 
United States, he has to be totally informed.
    But those issues weren't joined at his level until the end, 
for a month or two before we voted on the Iraqi resolution or 
the assembling of the national intelligence estimate, which 
didn't happen until the request of Senator Durbin and our other 
colleagues on the committee, about three weeks prior to our 
consideration and votes on the Iraqi resolution granting the 
President the authority to go to war in Iraq.
    It seems to me that here we had a threat over a decade, 
went to war already, that we already should have had a 
collection, an accumulation of analysis with respect to Iraq. 
But we depended on foreign sources. We relied on other liaison 
sources that obviously, as we discovered, were less than 
    So all told, I think that frankly, for those who suggested 
that we can't have major reform, are opposed to this idea, 
really have the burden of proof to suggest why not, because I 
think the time has come that we really do have to overhaul the 
    So I appreciate all that you are doing in this effort. And 
I think, if anything, we have to explore the issues that will 
invite the kind of change that's going to be essential to 
moving this legislation in a direction that's essential.
    Would you not think that in terms of information sharing 
and accountability, that a Director of National Intelligence 
would enhance that approach, not lessen it?
    Senator Feinstein. Thank you, Senator.
    Mr. Chairman, just follow the women on the committee. 
    Chairman Roberts. I get mail saying ``Ms. Pat Roberts''--
you never know.
    Senator Feinstein. Thank you, Senator.
    Chairman Roberts. Senator Hagel.
    Senator Hagel. No questions.
    Chairman Roberts. Senator Wyden.
    Senator Wyden. Thank you, Mr. Chairman.
    Senator Feinstein, I think this is an important concept 
that you're advocating. And most of the opposition to what 
you're proposing seems to me to be mostly theoretical. And I 
want to get your opinion on this.
    When we are at war and our troops are in harm's way, it's 
clear we're going to support the troops and we are going to be 
very sensitive to the concerns of what the military is seeking. 
But what the Iraq review shows and why I think your bill is 
important it that America isn't getting the intelligence the 
country needs before we go to war.
    Is that really what you're trying to accomplish in this 
legislation? Let's set aside the theoretical. I think the 
points that Senator Levin and others have raised--the Chairman 
has raised--I think are very valid. And I certainly want to 
make sure that when our troops are in harm's way, they get 
everything they need and that we're very sensitive to the 
    So I think that the idea of conflicts with DOD are mostly 
theoretical. What's not theoretical is what our review showed, 
is that we absolutely have to beef up the intelligence we're 
getting before we go to war.
    And why don't you outline, using the short time I have, why 
you think your legislation would do that.
    Senator Feinstein. Well, I think that's right, because we 
had a deeply flawed process where intelligence was both bad and 
wrong. At least up to the present time, I think that's 
certainly the case. And we made a huge policy decision based on 
    I can only speak for myself, and perhaps this is why I feel 
so strongly. That's why I voted to authorize use of force. And 
if you really think about it, any doctrine of preemption is 
dependent upon accurate information, accurate intelligence. And 
if you don't have that accurate intelligence, and it isn't 
fleshed deep and actionable, you're behind the eight ball from 
the beginning.
    And I think we have to take very seriously what was a 
massive intelligence failure. I don't really think anybody is 
to blame for it. The more I looked into it, it really was the 
processes and the structures of how things were done in this 
new world.
    And again, that may have been all right in state to state 
kind of intelligence where you're kind of on level playing 
fields. But when you've got cultural, language issues, a 
dictatorial country, you can't place agents, you're dependent 
on the other countries for intelligence, Senator Roberts used a 
term groupthink, which really boils down to an over use of 
assumption--because X existed and the shell of X is still there 
like Fallujah 1 or Fallujah 2, therefore, these factories must 
be making X, Y or Z.
    We can't afford that any more.
    Senator Wyden. I would only say in conclusion, I think this 
goes to the point you were making, Chairman Roberts. I just 
think it's important not to see this as a zero-sum game. When 
you listen to the debate in the past, the idea is if you give 
the intelligence community to do the job right, somehow you're 
taking away authority from the military. I don't think that's 
right. I think this gives us a chance to have a win-win, do a 
better job of intelligence and making sure that when our troops 
are in harm's way, we're still sensitive to the military.
    Thank you, Mr. Chairman.
    Chairman Roberts. Senator Feinstein, we thank you very 
    Senator Feinstein. I thank you, Mr. Chairman.
    Chairman Roberts. We have some questions for the record, 
and I'm sure that we can work those out and get the appropriate 
answer, more especially in regards to budget and the existing 
authority that the DCI now has.
    We turn now to our distinguished second panel waiting 
patiently. It is the first of many and gentlemen, I would not 
be surprised if you were asked to make a return engagement. 
This panel is a who's who of expertise in national security. 
Please, just have a seat.
    Our witnesses are Dr. John J. Hamre, who is president and 
chief executive officer of the Center for Strategic and 
International Studies.
    Welcome, Dr. Hamre.
    Dr. Hamre. Thank you, sir.
    Chairman Roberts. Dr. Hamre has served in senior positions 
in both the legislative and executive branches, including a 
stint as the deputy secretary of defense in the Clinton 
    Dr. Hamre is joined by Lieutenant General William E. Odom, 
United States Army, retired. General Odom is a senior fellow at 
the Hudson Institute, but he has served in nearly every 
national security setting imaginable since graduating from West 
Point in 1954. He's best known to the committee as the director 
of the National Security Agency from 1985 to 1988. He is also 
the author of ``Fixing Intelligence for a More Secure 
    Finally, we're joined by Mr. Jim Woolsey, who is a vice 
president at Booz Allen and Hamilton. His prior service, 
however, as the Director of Central Intelligence during the 
Clinton administration gives him a unique perspective on the 
reform issues we'll be discussing today.
    Gentlemen, thank you for being here today. Thank you for 
your patience.
    Dr. Hamre, you may begin.
    [The prepared statement of Dr. Hamre follows:]

                      SECRETARY OF DEFENSE

    Dr. Hamre. Mr. Chairman, thank you.
    I'm a little embarrassed to be the first witness because I 
have two colleagues here who are far more expert than am I. 
I've always been a consumer of intelligence products, not a 
producer. Both of these gentlemen are genuine experts as 
producers. So it's a little embarrassing for me to begin. So 
I'll be very brief.
    Let me first thank you, Chairman Roberts and Vice Chairman 
Rockefeller, for holding these hearings. I honestly believe 
that we're in a near constitutional crisis right now, a crisis 
of confidence at least, about an intelligence community that we 
can't live without and one that we've no longer come to trust. 
And it's really going to have to be through your efforts that 
we recover a sense of trust and confidence in an institution 
that we can't live without. We count on our government and we 
count on you to both safeguard our lives and protect our 
liberties. And that's what I think is at stake here. And so I'm 
very grateful that you've taken the lead to do this, Mr. 
    Let me say, I personally have not used the words 
``intelligence failure.'' And I've done that intentionally, for 
a reason, because I think that it tends to lead us toward the 
mechanical solution to a problem. And I think that this is a 
much bigger problem than can be solved with a mechanical 
solution of rearranging the boxes on a wiring chart. I think 
that this is inextricably tied with how the intelligence 
community and the policy community have interacted and do 
interact. And I don't believe that they're easily severable.
    So simply starting by saying it's an intelligence failure 
tends to lead you to a path of trying to reform the supply side 
of an equation. And I'll speak to this in just a moment. You 
and, I think, Senator Hatch spoke to the issue of the reforms 
in the Defense Department, and how Goldwater-Nichols was so 
    But I remind everyone here--and I was lucky to be just a 
little staffer who was working on this at the time--that when 
we were dealing with Goldwater-Nichols, we didn't deal with the 
supply side of the problem, we dealt with the demand side of 
the problem. We didn't deal with the military services which 
are responsible for organizing, training and equipping the 
forces. What Congress did was to strengthen the organizations 
that were demanding better military capability, the CINCs and 
the Joint Chiefs.
    It was an improved demand side that really led to the 
reform of the Defense Department, that really brought truly 
integrated operations. Goldwater-Nichols put a much stronger 
premium on institutions which demanded outputs, not who argued 
for inputs.
    And I honestly think that's part of the problem we have 
when we're looking here at this question of intelligence 
reform. Far too much of the debate is really about rearranging 
the boxes of the supply side of the equation and far too little 
discussion about improving the demand side of the equation.
    I'm just as much culpable of failure as anyone in this, so 
I don't excuse myself. But we have not spent anywhere near the 
time that we should be spending on saying how good is the 
output of this community. We've been spending far too much of 
our time arguing about the inputs for intelligence.
    I have to say that oversight reform really has to start up 
here, on Capitol Hill. I think we have to honestly say that the 
committees are spending far too much time arguing with each 
other about the various attributes of the input to the 
equation, rather than the quality of what we're getting from 
our intelligence organizations. Oversight in general has 
deteriorated in Congress. This area, it has, too. It's not been 
insulated from it.
    I do not exempt myself. I was a deputy secretary for three 
years and failed to insist on a quality outcome from a system 
that I was living with every day. So I'm not exempting myself.
    I would ask that you spend time thinking about how do you 
get better quality by demanding better quality. How do you 
institutionalize better demand? I hope that you consider some 
changes here in the Congress. In a way, I think the 
Intelligence Committees have simply become too big and they are 
spending far too much time on arguing about the inputs to the 
intelligence process--what should the next satellite look like? 
What should the receiver dish be like? What signal structure 
should we use for the antennas? Far too much time is being 
devoted to those kinds of questions, and not enough to asking 
why our analysis systematically misses major developments.
    I think you should institutionalize that oversight in a 
more deep way across the community.
    Senator Snowe, you've introduced legislation on creating an 
inspector general. I think it's an important step. How you, 
this Committee, connect with an inspector general, how you 
interact with him, I think that's a very important thing. It 
will help far more in driving and shaping the quality of 
outputs from this community.
    I think we should ask ourselves, should the NIC, the 
National Intelligence Council, only have on it members who 
represent the supply side of the equation--that is members of 
the leadership of the organizations that produce intelligence? 
Should it have a bicameral approach, where you've got an equal 
body that's on the demand side, asking how good is this 
analysis, does this help me make a decision, does this help me 
execute a war plan? We aren't doing enough on providing demand 
    So I would just plead for the Committee to see that beyond 
the question of organizing, rearranging the organization of the 
departments, look to how good we are or how poorly we're doing 
in demanding better quality from a system that we have to have.
    Mr. Chairman, thank you for inviting me to be a part of the 
    Chairman Roberts. Dr. Hamre, we thank you very much. Again, 
we thank you for your patience, and thank you for your very 
sound advice.
    General Odom, it's a privilege to have you here, sir.
    [The prepared statement of General Odom follows:]

                    NATIONAL SECURITY AGENCY

    General Odom. Thank you, Mr. Chairman. And it's a pleasure 
and honor to appear before you and the Committee.
    I've already submitted written remarks, and I won't repeat 
those. I would like to pull out a point or two and give it some 
special emphasis.
    Chairman Roberts. That's a rare occasion, and we certainly 
welcome that, sir.
    General Odom. Concerning this issue we've been discussing 
about the Director of National Intelligence as a new position, 
I long ago concluded that separating the Director of Central 
Intelligence from the director of CIA is essential. Double-
hatting one person in these two jobs blocks an evolutionary 
process that's been under way for a long time, and in the 
intelligence community it began at least in 1970. And it 
desperately needs to be continued. I think it bogged down some 
time ago because of this double-hatting.
    Now, the separation could easily be done by executive 
order. It does not require legislation. In fact, given the 
resistance to making the change, however, I've come to support 
this by doing it by legislation since there doesn't seem to be 
any other way to push it through.
    That's my basic reaction. But when I look at the details of 
the bill in its present form, I see that they could create more 
problems than they can solve. And let me focus on those that I 
think are problems.
    The DNI's budgetary authority. The bill gives him both 
program budget management and budget execution. Now to achieve 
the spirit of control of the legislation--that is, really 
influence where the money goes and can move it around--he 
really doesn't need execution authority. I think it will cause 
him more headaches than he can imagine and he'll rue the day he 
ever had it. Let me explain why.
    Maybe everyone isn't aware, but there are three budgets 
operative all of the time. The first is the budget you're 
spending, which you have put in law. There is the budget that's 
being considered before your committees now. And there's a 
budget being built for the following year. If you're the 
Director of Central Intelligence or the head of NSA or any 
other, you're dealing mentally with those three simultaneously. 
So let's separate them so that we're clear what we're changing.
    The first budget, as I said, has been written into law. 
There isn't much room to move money around. You can come back 
to Congress for reprogramming. But otherwise, once the money is 
locked in, it's going to be spent like you asked for it to be 
spent. And each department--Army, Navy, Air Force, Defense 
Department, Treasury, State, Energy and others--have their own 
accountants, controllers, who run the details of this spending 
    If one of these agencies wants to move money around above 
the reprogramming threshold, it has to get the DCI's permission 
to come over here to the Congress to get your permission. So 
the DCI could cut that off right there if he wants to today. 
Thus, I don't think budget execution authority is really what 
Senator Feinstein wants, that is, in light of what I heard her 
say she wants to achieve with this bill.
    Now, the second budget, the one that's pending before 
Congress, that's already locked in as far as Intelligence 
Community agencies are concerned. The only way to change it is 
by the suasion of their arguments to you. And you know more 
about that than I do, and I won't try to elaborate on it.
    The third budget is a program budget being constructed each 
year in the intelligence community, in the Army, Navy, Air 
Force, State Department, et alia, for submission to OMB and 
then to Congress.
    Secretary of Defense McNamara, you may remember, introduced 
a planning, programming budgeting system, or PPBS, for 
constructing this budget in the Pentagon. It gives the 
Secretary much greater influence on the allocations than a line 
item budget does because he can line up the money inputs 
against combat outputs. Now, prioritizations go on all year 
long in the various programs he's established to see how better 
to get the money lined up behind the policy outcomes he wants.
    The real influence over resource allocation is exercised in 
this process, not in budget execution. If the DNI is to 
exercise budget influence, this is the place to do it. 
Curiously, the DCI has had this influence for a very long time. 
In principle, he had it in the 1947 Act. But since there was no 
PPBS idea at those times, he really didn't have a very sound 
basis for using it.
    But I know for certain that President Nixon signed the 
memorandum in 1970 giving him program budget authority, and 
that every president since then has reaffirmed that in a new 
memo. And I think Nixon probably did that because he was 
responding to the fact that McNamara had left this system 
embedded in the Defense Department. He probably used it for 
building the National Foreign Intelligence Program.
    The problem has been that DCIs have not used that authority 
effectively. They have never established a system of program 
managers who could relate program inputs to program outputs. 
And the big obstacle has been the NRO's independent budget, 
which means nobody can stand up and speak for and allocate and 
talk about trade-offs within imagery intelligence or within the 
signals intelligence, nor is there a program where a single 
program manager looks out for the analysts.
    We were discussing earlier the fact that poor analysts 
never get looked after. That is true because there's no program 
management above each independent part of the budget for 
analysts, nor is there a program manager for 
counterintelligence that has the authority to move the program 
monies around, seeking better output performance.
    Now, that's the obstacle. Until the DCI uses his management 
authority, I don't see how giving him more authorities on paper 
can do anything but worsen his relationship with others 
departments and agencies.
    The second problem I see concerns the DNI or the DCI's 
relationship with the military services. The military services 
are the largest and most critical consumers of intelligence in 
both peacetime and in wartime.
    And over the past 20 years, we have achieved something I 
never thought we could achieve when I was looking at the 
intelligence community in the 1970s. National level collection 
has been linked in through technology to support tactical level 
collection. This first happened in the SIGINT world. It should 
now be happening in the imagery world as well. It has not 
happened, to speak of, in the HUMINT world, and interestingly 
because CIA does not like to work for regional commanders under 
a CINC's operational control.
    Now there is integration over turf boundaries between the 
National Foreign Intelligence Program and TIARA, although 
probably not yet enough. There are people in NSA, probably in 
the National Geospatial-Intelligence Agency, who don't like 
that integration. I know when I was the director, I had to 
resist such internal opposition at NSA. But progress was made.
    Would a Cabinet-level DNI have the effect of fragmenting 
the national level from the tactical level in that regard? I 
don't know. Maybe not. I would pay attention to that 
    Another dimension is the connection between the DCI, CIA 
and the military services concerning special activities--i.e., 
covert action and paramilitary operations. I can't go into this 
topic in detail in an unclassified forum. Just let me say that 
anyone with experience in it, if he looks at the draft 
legislation, he would see potential for serious bureaucratic 
clashes and dysfunctional behavior, greater than what we have 
experienced without overcoming it in the past.
    Now, to some degree, these problems will also exist if you 
just separate the DCI from the Director of CIA. I don't think 
they are reasons not to do so. I think these are just areas you 
need to be careful about damaging to make sure that you solve 
them properly.
    Let me end saying that I don't think you can do effective 
reforms unless you separate these two jobs. I think the best 
metaphor for the problem of double-hatting is to see the DCI as 
having meant to have been the teacher on the playground and the 
directors of the other agencies as the kids.
    Now when the DCI is double-hatted, he becomes one of the 
kids and fights with them. What you're really asking for is, as 
I heard Senator Feinstein describe it, is for the DCI or DNI to 
stand up and be the teacher. And I don't think that adds 
another layer of bureaucracy. The bureaucracy is already there 
in the IC staff, the National Intelligence Council and a couple 
of committees. The problem is that this bureaucracy is just not 
doing its job.
    Now, let me add an additional point, which you didn't bring 
up in this discussion, but I want to take this opportunity to 
get on the record, and that is counterintelligence. The so-
called MI-5 model is not a good solution. MI-5 has serious 
flaws. A separate national counterintelligence service does 
make sense. Any counterintelligence agency that has arrest 
authority will never rise above being a police agency. Police 
agencies are users, not distributors of intelligence.
    You can see that in the Navy's Naval Investigative Service 
and the Office of Special Investigations in the Air Force where 
criminal investigation and CI are combined, and then you see it 
separated in the Army. The Army has poor performance cases, but 
it's also had some good performance cases, attributable in part 
to separation.
    Also, this relates to having a DCI or DNI separate from the 
Director of CIA. If you created a CI organization sometime 
later, there's no place to put it but in competition with the 
CIA, without a referee, making this present intramural game in 
the IC even worse. If the DCI is separate, you put the new CI 
agency at the same level as the CIA and the DCI or DNI.
    Let me end by saying I think organizational reform will not 
ensure competent performance, but dysfunctional organization 
can prevent good leaders from turning in good performances.
    Thank you.
    Chairman Roberts. General, we thank you very much for your 
very comprehensive statement based on a great deal of 
    Mr. Woolsey.


    Mr. Woolsey. Thank you, Mr. Chairman, Senator Rockefeller, 
members of the committee.
    I would note one matter in my biography that may be 
relevant here. When I was general counsel of the Senate Armed 
Services Committee in the early 1970s for Senator Stennis, I 
was one of three congressional staffers who was responsible for 
the staff work on oversight of the intelligence community. And 
I know something of the difficulties of oversight and what an 
important job it is and what an important thing it is to do it 
    Let me just address two issues here this afternoon--first 
of all, the wisdom or lack thereof of splitting the current job 
of the Director of Central Intelligence into an official who is 
the head of the CIA and a separate official who is, let's say, 
the Director of National Intelligence; and, secondly, what the 
effect would be on our ability to do some things right that we 
have not done right in the past.
    It's about sort of a 60-40 proposition. I believe that 
splitting the current job would be a wise decision. I say 60-40 
because I don't think this is one of those issues which is 
absolutely clear-cut. I think there is a risk of layering and 
adding an added bureaucracy, as several Senators have 
    And I believe the job could be done by one individual in 
ideal circumstances. What I mean by ideal circumstances is if 
the currentDirector of Central Intelligence had a close and 
cordial working relationship with 10 people--the President, the 
Secretary of Defense, and the chairman and ranking members of the 
oversight committees and the Senate and House Appropriations 
    But you need all 10 of those. I had eight. I didn't have a 
bad relationship with President Clinton; I just didn't have any 
relationship at all. And I believe I got along reasonably well 
with seven of the eight senior members of the committee in the 
Congress. But it's no secret that Senator DeConcini, the 
chairman of this committee at the time and I agreed on 
virtually nothing over a two-year period.
    Partly as a result of that, Congress was in session 195 
days my first year in the job, in 1993. And I had 205 
appointments on the Hill. I was up here an average of more than 
once a day. I'd say three-quarters of that was attempting to 
restore funds for satellites or supercomputers at NSA or Arabic 
language speakers that Senator DeConini and I disagreed about 
and that had been cut from the budget.
    It takes a great deal of time if one is in less than an 
ideal situation. And one cannot count, as a Director of Central 
Intelligence under the current system, on having a close and 
profitable and useful working relationship with 10 individuals. 
Sometimes, it just will not work out that way. So in my 
judgment, dividing the job makes more sense than not.
    It, I think, should not be a czar. I hope we can banish the 
use of that word from discussion of this issue. To my mind, 
half a millennium of stupidity, rigidity and autocracy, 
followed by the victory of Bolshevism, is not a good model for 
the management of the American intelligence community.
    But I do believe that it is a better idea than not to 
establish an overall head of the community, someone who has 
important responsibilities in dealing with the President, 
dealing with the Congress and dealing with coordinating the 
community's work.
    And the system will probably work better under those 
circumstances if this committee and its counterpart in the 
House are alert to the risks of a bureaucracy and the like that 
have been discussed before.
    Should this person be a Cabinet member? I believe that for 
the overall Director of National Intelligence, yes, that makes 
sense. It's largely honorific. The difference between being an 
executive level two and executive level one essentially is zero 
in terms of pay. And the main difference is you get to take 
your chair at the end of the administration from the Cabinet 
    But when President Reagan named Bill Casey as a member of 
the Cabinet, he said something special about his relationship 
with Mr. Casey. And I think that it is a good idea for the 
Director of National Intelligence--let's call him or her--to be 
a Cabinet member and to have that status and authority.
    What would be the effect of such a change on the 
intelligence community's ability to provide sound intelligence, 
to do some of the things that Bill has talked about, such as 
integrate intelligence with battlefield requirements, for 
    I think there is no way around the proposition that the new 
Director of National Intelligence has to be a partner, 
particularly with the Secretary of Defense. Much intelligence 
goes to the Defense Department. Defense is a very important 
consumer. Much of the technical requirements have to do with 
    This system simply will not work--the new one or the old 
one, I would say, the current one--without a close partnership 
between the Secretary of Defense and the DCI. In this regard, I 
was very lucky. I first had Les Aspin and then Bill Perry, old 
friends of mine, and wonderful individuals, easy to work with. 
That relationship in my tenure, I think, worked well.
    Will this improve our ability to avoid 9/11s? I think 
probably a little bit, but not a great deal, because I don't 
regard 9/11 as principally a failure of foreign intelligence. 
The CIA did a very bad thing in neglecting to put Messrs. al-
Mihdhar and al-Hamzi, who had been at the 2000 January meeting 
in Malaysia that planned 9/11, on the watch list and, 
therefore, made it virtually impossible for the State 
Department to keep them out of the country or the FBI to find 
them once they finally started looking for them just before 9/
    But many of the other failures were failures either within 
the FBI, the failure to coordinate bright agents looking at 
situations in Minnesota with those looking at the situation in 
Arizona, the failure of the Air Force of not having fighter 
interceptors anywhere near either Washington or New York, the 
failure of the FAA in permitting flimsy cockpit doors.
    I think we need to face the proposition that the nation was 
asleep at the switch before 2001, and it's very important to 
realize that the terrorists know our system very well. Most of 
the planning and organizational work for 9/11 was done in two 
countries where the CIA doesn't spy, and where essentially the 
United States doesn't collect foreign intelligence--the United 
States and Germany.
    The terrorists knew exactly what they were doing. We would 
have probably had a better chance of finding out what they 
might have been working on if they had been operating out of, 
let's say, Syria.
    I think that it is also important to realize that 
information sharing is not the be all and end all. One wants to 
share with the right people, but sometimes the people one 
trusts are a Rick Ames or a Bob Hanssen or the Walkers. And one 
does certainly not want to share with them.
    So the answer is not just more sharing. It's care and 
precision, avoiding stovepipes that don't make sense, avoiding 
keeping intelligence within certain channels where it cannot be 
exploited effectively. But in a sense, one wants competition in 
intelligence collection. The way we at the CIA developed the 
Predator in 1993-94 was precisely because the Defense 
Department's way of going about building unmanned aerial 
vehicles was extremely slow and expensive. And so we competed 
with them and I think, on that one, did a better job.
    One also wants competition, I believe, in analysis. Getting 
two different sets of eyes on the problem is a good idea, not a 
bad idea. What one doesn't want is competition between 
stovepipes of data thatdon't get integrated and don't get 
shared with the right people. But it is not simply a matter of fusing 
everything and sharing it widely.
    Finally, I think our domestic intelligence collection and 
utilization will be an absolutely vital part of avoiding and 
limiting the effect of terrorist attacks in the United States.
    In some ways, the key relationship of finding out about a 
possible terrorist attack in the United States is a 
relationship, if I could figuratively say, between Mr. Hassan 
in Dearborn, Michigan, who runs the corner grocery store and 
Officer O'Reilly, who is walking the beat. If Mr. Hassan feels 
comfortable in alerting Officer O'Reilly to something that he 
thinks maybe the officer should know about and that information 
is dealt with fairly and decently by our local law enforcement 
forces and they coordinate properly with, say, the FBI, we will 
have a much better handle on the possibility of avoiding or 
limiting the effect of terrorist attacks than with most 
anything else that we could do.
    Should we move toward a British style MI-5? I don't think, 
yet, that is a good idea. I think the FBI deserves a chance to 
use the 25 percent or so of its personnel that it's set aside 
to look at counterterrorism, to work with its state and local 
contacts and the rest.
    I think it's important to watch how it does, with 
particularly the Islamist organizations in the United States, 
Wahhabi-funded organizations, organizations such as the ones in 
Herndon, Virginia, that have been investigated that have been 
supporting of terrorism. If they do a good job with that, I 
believe that they should continue to have the domestic 
intelligence collection job. But I think the jury is still out.
    Finally, human intelligence overseas. Do we need more of 
it? Yes. Do we need to be better at it? Of course. Was it 
unwise to cut out all those CIA stations in the early 1990s and 
close them down overseas? Yes. Should we have spent more money 
on Arabic and Farsi? Yes, of course.
    But keep in mind that Mr. Kay told us, after he stepped 
down last fall, that he believed that the individual generals--
and I believe he said all of the individual Iraqi generals--who 
had been in charge of battlefield units, and they had been kept 
separate from one another after they were captured, each one 
said that, no, his unit didn't have any chemical weapons, but 
he believed that the unit to his right and the unit to his left 
    Now, put yourself in the position of poor George Tenet and 
his Director of Operations. Suppose they had been the most 
skilled spymasters the world has ever seen, and somehow they 
managed to recruit a dozen or so senior Iraqi officers as 
informants, as spies. And those dozen sincerely believed that 
the units, each on his right and each on his left, had chemical 
    A country such as Iraq before the war was a wilderness of 
mirrors. And it is not going to be the case that just by doing 
better and having more effort on human intelligence, we are 
going to necessarily be able to do a better job of 
understanding when we are being deceived and when we are 
being--well, when misleading statements are being given to us.
    As a general proposition, I think bills of the sort that 
Senator Feinstein and the one I've had a chance more to review, 
Congresswoman Harman's, head in the right direction. And I 
would urge the committee to support a bill of that sort.
    Thank you.
    Chairman Roberts. We thank you very much for your 
testimony. Let me start by saying you mentioned the frequency 
with which you visited Congress, either heels dragging or not. 
I hope we've been a bit easier on the schedule for those that 
have followed you. As a former staffer and as a frequent 
visitor, I am interested in your take on the repeal of the 
committee's term limits in this year's Intelligence 
Authorization Act.
    Mr. Woolsey. I am delighted to see the repeal of the term 
limits, Mr. Chairman. I never thought they were a good idea. I 
think Frank Church was wrong that the CIA was a rogue elephant 
that was likely to seduce any congressman or senator into 
supporting whatever it did if they stayed on the committee more 
than a few years.
    I saw, as general counsel of the Senate Armed Services 
Committee, a great deal of expertise built up in individual 
Senators and individual staffer members' backgrounds over the 
years. I think congressional committees work well that way. And 
I would really be very much pleased to see the repeal of all 
term limits.
    Chairman Roberts. General Odom, do you have any view about 
    General Odom. I would be delighted to see the limits 
lifted. Re-educating or educating new members all the time, 
when I was Director of NSA, was a challenge. And I don't see 
how anybody can, with the busy schedules you have, divine the 
nature of this arcane community in the time you have available. 
I think it takes quite a while. And once you have people with a 
reasonably sound understanding of it, they're valuable. They 
shouldn't be pushed off the committee.
    Chairman Roberts. Dr. Hamre.
    Dr. Hamre. I support it. And may I also add that I'm a big 
fan of the seniority system. I frankly think that we ought to 
honor the seniority system. I think that it makes sure we have 
leaders in charge of the committees who have got a lot of 
experience and are more detached from the parochial interests 
back home.
    Chairman Roberts. Well, obviously, as you're here longer, 
the more you appreciate it. [Laughter.]
    Mr. Woolsey, how much has congressional concern over turf 
contributed to your difficulties, or in general the DCI's 
difficulties in carrying out the duties over the years?
    Mr. Woolsey. Well, I'm only really cognizant in detail, Mr. 
Chairman, of the two years I was DCI.
    And I would say then it was less, except with one 
exception, it was less a question of turf. Certainly turf 
between the congressional committees was not that big a problem 
as it was of substantive disagreement. Senator DeConcini and I 
just had different views on virtually everything, and it meant 
that it took a great deal of effort for me to work with other 
members and him, when I could, work out compromisesand the 
    In the aftermath of the Ames case, this committee drafted 
legislation which would have transferred all 
counterintelligence, including overseas penetration of foreign 
intelligence services, to the FBI. And, of course, we opposed 
that in the CIA because, although having a strong FBI role for 
counterintelligence in this country is important, for the FBI 
to try to undertake overseas penetration of foreign 
intelligence services struck us as a very bad idea.
    That was a turf issue in a sense, because the House and 
Senate committees, happily, saw that issue very differently. 
The CIA and the FBI saw it differently. Eventually, we ended up 
with a system that did not make that transfer. That was the 
only case I can recall of the two years I was DCI where we had 
so-called turf issues salient.
    Chairman Roberts. Let me ask this question of you three in 
the 11 seconds I have left. What would you do to improve 
congressional oversight of the intelligence activities, and 
more particularly, the Senate in this case? Is there a problem 
with the structure in regards to overseeing the intelligence 
activities in terms of authorizing and appropriating budgets?
    Because that really gets to the nub of it in regards to--we 
have 37 staffers. We have 17 members on this committee. We like 
to think we do a good job with many, many, many closed-door 
hearings in going over the priorities of the intelligence 
    We make recommendations. That goes to sequential referral 
to the Armed Services Committee. Then there are the 
appropriators. And we make every endeavor to shine their shoes 
and to clean their windows and to carry their suitcases and do 
things of this nature to gain some degree of influence.
    Then you have to deal with OMB, and you have to deal with 
the agency involved. And so this is the typical lament of the 
authorizer. I'm sort of leading you here in this regard, but 
would you have any comment in terms of how we can do our job 
    And we can just go down the line--Dr. Hamre.
    Dr. Hamre. Sir, this is going to take a lot longer than an 
11-second reply, and I'd like to come back and talk with you 
about it. And I'll only speak to my experience----
    Chairman Roberts. Glad to have you back.
    Dr. Hamre [continuing]. On the Armed Services Committee, 
not on the Intelligence Committee. But frankly we are caught in 
an endless competition to try to be relevant compared to the 
Appropriations Committees. And we define too much of our 
relevance as an Armed Services Committee by trying to do what 
the Appropriations Committee is doing. They have the upper hand 
because they have money. We have a hunting license; they have 
    If we're going to really find a relevant role for the 
oversight committees, it's got to be in that role providing 
oversight. The power of the committee rests in you and the 
Senators. It doesn't rest with the staff in their capacity to 
second-guess a lieutenant colonel who's coming over from the 
    So finding a way for you to put a higher premium on the 
kinds of questions that Senators ask--the deep-probing, very 
simple questions they're asking on behalf of citizens--and 
trying to find an institutional structure to help you, I would 
advocate that you create a task force for oversight inside your 
committee and set aside a small, special staff element. Don't 
staff it extensively. Staff it in a very small way, and have it 
ask big questions as a starting point.
    Appoint a board of visitors for each of the intelligence 
agencies and ensure the Committee meets with them. What are you 
finding? What are you hearing? How good is this? Why isn't it 
better? Why did we miss this? Those sorts of questions aren't 
being asked because we're spending too much time at looking at 
budget inputs.
    Chairman Roberts. I thank you for your comments. And I 
meant to say the committee is going to have the heads of all 15 
of the intelligence agencies, the acting director at the CIA 
and others, to indicate what they have done since 9/11 and what 
they have done since our report in regards to progress if, in 
fact, that much has been made.
    General Odom.
    General Odom. Mr. Chairman, I was always surprised that the 
committees did not force us to present a program planning 
budgeting structure for you so that you could see what you were 
buying with the dollars in and out.
    I knew we couldn't provide it, but I thought if you asked 
us, it might put the pressure on us to do it. And if you asked 
us to it, if you had asked me to do it in my day, and I had had 
the authority to program all of the SIGINT monies, my agency 
would have gone crazy with all the pressure I'd have had to put 
on it to organize properly a program budget. But I know at 
least NSA had the info necessary to produce a proper one.
    The other thing that occasionally did happen, which I 
liked, was that you Members of Congress can speak to the public 
and you can tell Americans whether the intelligence community 
is violating their rights or not. We in the intelligence 
community cannot convincingly reassure them that we do not. 
When I was the Director of NSA, I wanted you to see inside my 
agency and to know what went on there so that you could stand 
up and say, yes, your rights are being protected as Americans.
    And I think that's a major role you can play. And I would 
subscribe to some of the points that Dr. Hamre made as well.
    Chairman Roberts. Mr. Woolsey.
    Mr. Woolsey. You have pretty much unanimity here, I think, 
Mr. Chairman. If there are some 50 people I guess on this 
committee, involved in oversight, I imagine you're in the 
ballpark of 150 for the Congress as a whole, with the members 
of the House committee and the members of the two 
Appropriations Subcommittees and the staffs.
    And that's a lot of people. And particularly, frankly, on 
the staff side, people used to get into when I was DCI some 
extraordinarily detailed technological arguments. You know, it 
seemed to me much to the detriment of spending time on the 
sorts of things that both John and Bill have addressed and 
which seem to me to ought to be more of the focus of a 
distinguished committee of this sort.
    Chairman Roberts. Senator Chambliss.
    Senator Chambliss. Thank you, Mr. Chairman.
    Gentlemen, I could sit here and listen to you all all 
afternoon. Your insight because of where you've been is truly 
fascinating on this very complex issue that we're trying to 
deal with.
    I want to ask you just one question, and it's basically the 
same question I addressed to Senator Feinstein a while ago. 
Rather than creating this separate DNI with the CIA, FBI, DIA 
and every other agency still in place, why doesn't it make more 
sense to bring all of the intelligence community under one 
umbrella whether you--I don't care what you call it, but let's 
assume it's under the current Director of Central Intelligence 
and create a Ford Motor Company or any other major corporation 
with the departments out there of domestic intelligence--and 
defense intelligence so that we don't have the overlap?
    Right now, I envision that there is a lot of confusion. And 
obviously there is. That's why we missed some of the pre-9/11 
signs, and we all know that.
    But is there a way--John, I'll start with you--maybe to 
make the Director of Central Intelligence the Director of 
National Intelligence and bring all of these agencies under 
him, give him that power to hire and fire and the budgetary 
authority without creating another level of the bureaucracy?
    Dr. Hamre. Senator, let me just start by stating my 
overriding worry: Because we are so dependent on this community 
and we need it so much and we need it to be good, we can't 
afford to have it not have a richness of thought inside the 
community. I worry about trying to bring all the agencies under 
a single structure that has, you know, mission control in a 
budgeting sense. And that is likely to narrow how we understand 
problems, not broaden how we think about them. So I personally 
would worry about that side of it for you. But that's a more 
philosophic question.
    Could you pull out of the Defense Department the elements 
that do intelligence and put it under the Director of Central 
Intelligence? Well, ultimately you could, but this is like 
trying to remove the lymphatic system out of the body and not 
touch blood vessels. Military intelligence is deeply, deeply 
tied to battlefield operations. And how you then surgically 
remove that, put it in a different body and not bleed to death, 
is, I think, going to be the key question. I think we would 
have a lot of difficulties, frankly, doing that.
    I understand what you're describing is a central problem. 
Our system, you know, we envision a 100 percent perfection in 
execution, and we really usually get 20 cents on the dollar 
when we do it. I would worry that we'd lose a lot in the 
process, just to be perfectly candid to you, sir. But I 
understand what you're trying to do.
    General Odom. Senator Chambliss, my answer will be, yes to 
some things, no to other things. You can centralize certain 
kinds of technical activities like SIGINT, like imagery, et 
cetera. But you cannot centralize analysis. It must be like 
distributed processing, and everybody needs his own analysts in 
his own office so that he's interacting with them so the 
analysts know what he wants.
    You cannot sit at NSA with a big mainframe computer and 
have people on dumb terminals out in the field calling up and 
asking what's on the other side of that hill. You've got to 
distribute that task to the local scene. The answer is, you 
have to have both centralization and decentralization.
    Now, let me try to give you another way to see this issue. 
In the military, there are several staff functions. The J-1, 
the J-2, the J-3, the J-4 are the principals--personnel, 
intelligence, operations and logistics. It's the G staff in the 
Army. It's the S staff down below the general officer level. 
You have the same staff functions at all levels.
    The intelligence function, a staff function, is just like 
the operations functions where the war plans are written. Would 
you outsource the writing of the war plan from the J-3 to 
somebody outside the Defense Department? Would you outsource 
the J-1, the personnel function? Would you outsource all of the 
    You can outsource some parts of it, but finally, each 
commander must have a particularized war plan; for that reason, 
he has to have the people who design it directly on his staff. 
And because there are so many people making decisions at so 
many levels, and in different departments, they must be 
supported on a highly distributed, process-driven basis, both 
for intelligence analysis and operational planning.
    Try to use the analogy of thinking of intelligence as a 
news service. Think of SIGINT, say it's radio. HUMINT is print 
news. Imagery is television. Now if you are in the business of 
doing something for which you need news, you'll subscribe to 
all three, and you won't subscribe to every program, but you'll 
start picking and choosing the ones most relevant to your 
business. You'll put what you need together in a particularized 
    What NSA and an imagery agency and a HUMINT service at CIA 
can do is provide this kind of news service, allowing 
intelligence analysts in many user agencies to subscribe and 
receive that service.
    The director of this whole operation, the DCI, has to 
orchestrate it, make it responsive. The influence of the DCI or 
DNI on the President's thinking that is most important is not 
exerted by sitting in the Oval Office and talking to him. I've 
witnessed the process. I served four years in the Carter White 
House and watched the DCI come in at NSC meeting after NSC 
meeting. The intelligence influence that counted poured into 
the Situation Room as both raw intelligence and all-source 
analysis that was distributed to the NSC staffers, who 
integrated it into their analysis. The President daily is 
reading their integrated policy memos. Additionally, there were 
three or four intelligence summaries a day that the national 
security advisor took up to the President, explaining why and 
how they were important.
    There is no way that the DCI can come occasionally and 
whisper the whole intelligence picture into the President's ear 
and thereby supply all of his intelligence needs. If the DCI 
never showed up in the White House but had these feeds of 
intelligence into the Situation Room working effectively, he 
would have a powerful impact. I've actually seen a situation 
where my own agency reversed the President's position 180 
degrees in four days. I didn't go to the White House. The NSC 
staff received the steady flow of reporting and recognized what 
it meant for thePresident's policies.
    So we have some popular images of how this works that are 
at odds with the reality. If you go to an NSC meeting, don't 
think the Secretary of Defense, the Secretary of State, 
Chairman of the Joint Chiefs come waiting breathlessly to hear 
what the DCI is going to tell them about what's going on in the 
world. They know before they come, because their intelligence 
staff have made it available to them. The great service the DCI 
has done is to orchestrate all distribution, to have 
collection, analysis and all of these news services clicking 
and getting their information to the right people at the right 
    And as we sit here and are critical of our intelligence 
community--and I'm one of the sharpest critics--we should 
remember there isn't any other country in the world that has an 
intelligence organization that works anything like as well as 
ours. So in a comparative sense, you're dealing with a 
Cadillac, not a worn-out Chevrolet. And the question is whether 
it needs a 50,000 mile checkup, whether it needs a new ring job 
or a few things like that.
    That's the perspective I would put on it.
    Mr. Woolsey. Senator Chambliss, three quick points.
    First of all, having this same individual, the current DCI, 
who runs the CIA and has some authority in the community, 
continuing in that job, as I said, is a question really of time 
and attention. One of these books that has come out about the 
1990s quotes someone anonymously from the CIA saying Woolsey 
wasn't really interested in us. He was always worrying about 
    Well, I did have to worry about satellites some because 
otherwise the country would have gone blind if the cuts had 
gone through. And so there's just so much time in the day if 
you're trying to have one person do this job.
    Second, with respect to foreign intelligence coordination, 
I think it is a problem if the Director of Central 
Intelligence, also the head of the CIA, is thought to be sort 
of the referee between, say, CIA and NSA. NSA won't really 
trust that. And so that's one problem with having the job all 
in one person today and having it all underneath the single 
individual. And finally, if you put domestic intelligence 
collection under it, something like the FBI or if you move to 
an MI-5, you really don't want the public believing, I think, 
or it to be the reality, either or both, you don't want the 
public thinking, accurately or inaccurately, that, if I can be 
blunt, the dirty tricks of foreign intelligence are being used 
against them.
    CIA officers are effectively trained to lie, cheat and 
steal for the United States. They go abroad to lie, cheat and 
steal in order to obtain intelligence. That's their job.
    And the whole tradition and culture of criminal 
investigation on the part of the FBI of chain of custody of 
evidence, of the way in which they have to question witnesses, 
all of that is a completely different culture. It may not be 
the right culture from the point of view from maximizing the 
ability to run domestic intelligence collection. But we ought 
to remember that the FBI ran successful long-term penetrations 
both of the American Communist party and the Mafia.
    Now those were not normally the skills that they used in 
kicking down doors and grabbing the 10 most wanted and helping 
prosecutors get convictions. But they had some bright and able 
people who were able to do that.
    I think we're better off trying under the current system to 
let Mr. Mueller give it his best and see if he, operating 
through the Justice Department and so forth, under the current 
system, can set aside this 25 percent or so of the FBI and work 
with state and local and run domestic intelligence collection, 
in a sense, outside the aegis of the foreign intelligence 
    I realize that means it's going to have to be coordinated 
and you're going to have to have much more seamless 
communications than we have now. But some of the barriers to 
that seamless communication have been knocked down in the 
PATRIOT Act. And I think the communication is better between 
the Bureau and the Agency than it was before 9/11.
    Chairman Roberts. Senator Durbin.
    Senator Durbin. Thank you very much.
    I have how many minutes, Mr. Chairman?
    Chairman Roberts. I beg your pardon?
    Senator Durbin. How many minutes do I have?
    Chairman Roberts. Well, we've been operating under the 
four-minute rule that has been stretched to eight.
    Senator Durbin. Well, I'm going to try not to abuse that. I 
have four questions. So let me, if I might, just pose these 
four questions, one more particularly to Mr. Woolsey, to see if 
I can receive a reply from the panel and any person who feels 
encouraged to respond.
    Mr. Woolsey, if you start with the premise that the reason 
that our intelligence agencies cannot share information is the 
fear of an Ames or a Walker or a Richard Hanssen, how in the 
world will we ever get cooperation? Why would the FAA tell the 
FBI about the people who are on the passengers' list for fear 
that perhaps there's a Richard Hanssen sitting in the FBI who's 
part of it? That seems to me like a perfect recipe for no 
communication. I wonder if you could expound a little bit on 
your statement on that.
    Secondly, you said--and perhaps others feel the same--we 
were asleep at the switch on 9/11. And you pointed out three or 
four specifics. Where are we asleep at the switch now? Where do 
you look at the current situation of the security of the United 
States and feel that it's obvious that we're asleep at the 
switch again, we're missing another vulnerability? I'd like to 
know if you could share.
    Third, has the world changed in intel with the preemptive 
strike doctrine that this administration has suggested? Are we 
now in a shoot first, ask questions later situation where we 
don't have the time to sit down and carefully go through the 
intelligence for fear that 9/11 occurs before we reach the 
right conclusion?
    I recall what happened on September 9 before the invasion 
when I sent the letter and asked the CIA where is the national 
intelligence estimate. We're getting ready to vote on a war and 
we don't have an NIE. And they said they scrambled then in 
three weeks to put one togetherwhere it ordinarily took six 
months. And it struck me that they were not ready to ask the questions 
and answer them. They were moving forward on an agenda and with a 
process that I guess is unprecedented.
    The final question is this. What are we losing with an 
acting director of the Central Intelligence Agency? There's a 
lot of consternation. Senator Chambliss and I were on a show 
over the weekend. When they asked us the question are you going 
to fill this spot, I'm not sure we could if we wanted to. It's 
a pretty tough spot to fill--George Tenet's shoes. But what are 
we losing today in terms of the defense of this nation by 
having an acting Director of Central Intelligence as opposed to 
someone permanently appointed?
    Those are the four questions.
    Mr. Woolsey. I'll take a quick stab at them, Senator 
    First of all, fear of a spy, a penetration: how can we 
share material across agencies if that fear exists? That fear 
will have to continue to exist. It's a realistic fear.
    I think we need to look for better technological ways to 
enforce need to know. It has to be the case that one can 
confidently share a bit of sensitive information about, say, a 
source with one or two key analysts in another agency without 
feeling as if it's being opened up to a lot of people. And I 
also think for security purposes, frankly, we put far too much 
stress on the polygraph. I think much of it is not only a waste 
of time but is counterproductive in that it gives too many 
false positives. I would highly commend to the committee the 
excellent National Academy of Sciences study of this about a 
year, year and a half ago.
    Where are we asleep now? I think that one of our biggest 
vulnerabilities here is not understanding the impact that the 
Wahhabi ideology out of Saudi Arabia has had on creating the 
infrastructure of terrorism and people who are willing to 
support it, including organizations, including organizations 
    I do not mean to suggest this is true of the whole Saudi 
state. I think the crown prince, for example, is basically a 
reformer, a moderate reformer who is trying to make some 
positive changes in Saudi Arabia. But the Wahhabis, their 
religious ideology suffuses what they send out into the world, 
and they spent some $70 billion since 1979 suffusing their 
ideologies in to the madrassas of Pakistan and the prisons of 
the United States. And I think we really need to focus very 
hard on that kind of threat.
    Yes, the world has changed, I think, with the preemptive 
policy the administration enunciated in '02. I think there was 
a reason behind that policy. We may have to move in some 
circumstances because of a joint concern about weapons of mass 
destruction and ties of a dictatorship to terrorist groups. And 
that's a much harder situation than we faced in the Cold War 
with deterrence and containment.
    And what are we losing with having an acting director? I 
think for the number of months having John McLaughlin as an 
acting director is a perfectly reasonable thing to do. He's an 
extraordinarily able, loyal man.
    I think it would be difficult to move toward having a new 
DCI under the current circumstances until one decides whether 
one is going to keep the DCI structure you have today or move 
to something more like the DNI split job that Senator Feinstein 
and others have suggested. I don't know exactly what one would 
be confirming a new director for.
    So I think that John McLaughlin is an extraordinarily able 
individual and the country is very well off in having him in 
that acting position. And I think he could do the job for a 
substantial period of time if necessary.
    Dr. Hamre. May I just speak to one thing?
    Chairman Roberts. Certainly.
    Dr. Hamre. I don't think we pay enough attention to the 
interplay of criminality and terrorism. In general, what we're 
now learning is that criminal networks become the logistics 
backbone for terrorist activities. We see it in the Balkans. We 
see it in South America. We see it in the Middle East. And 
unfortunately that falls in the cracks of the fault line that 
exists in our organizational consciousness. Law enforcement 
tends to be a law enforcement activity, not an intelligence 
activity and vice versa.
    We tend to send--you know, the FBI has its attaches 
overseas now. We reinforce that fault line. But if there's one 
single thing that I think is the most serious thing we ought to 
be looking at now is this interplay, this heavy, heavy 
interplay of transnational criminality and transnational 
security threats.
    Chairman Roberts. So General Odom, do you have any 
    General Odom. First, concerning the preventive war 
doctrine, I must find it hard to square with the U.S. 
Constitution and our whole tradition. I can understand that if 
you know an attack is eminent, that's one thing, but when you 
wage preventive wars, which it seems to me what we did in Iraq, 
the cost of that, you know, this gets beyond the intelligence 
world. And so that's a bigger issue.
    What are we asleep on? I think we're asleep on 
understanding the bigger issues. I think we have terrorism 
prioritized way above where it ought to be and that we should 
be looking at intelligence to help us understand how to achieve 
stability in this region of the Middle East, Southwest Asia, 
and then decide where dealing with the terrorism fits into it 
and where proliferation fits into it.
    And failing to understand the impact of a number of our 
policies and actions, recent ones, old ones, and how they're 
changing things, and how they're going to present us with 
surprises, and we'll say well, why didn't we wake up for that? 
Well, we're out causing some surprises by our wrong-headed 
    Finally, on appointing a new DCI now, if you're going to--I 
couldn't imagine anybody really wanting the job until he knows 
how the election will come out and he knows whether you're 
going to legislate reform. So it's kind of a moot issue, I 
    Senator Durbin. Thank you.
    Chairman Roberts. Senator Snowe.
    Senator Snowe. Thank you all for being here today. It 
certainly provides some exceptional insight into many of the 
challengesthat we're facing as a country and within the 
intelligence community.
    First of all, I mean, obviously the great impetus here is 
the report that we disclosed. And obviously it was extremely 
troubling from the standpoint that most of the major 
conclusions were wrong in the final assessment.
    Can you give insight to this Committee what your reaction 
was to this report and any particular areas where you thought 
it was unusual that problems manifested themselves either with 
the collection, the analysis, the failure to share information, 
the credibility of sources, the lack of knowledge, the way in 
which the NIE was assembled? Was there anything that really 
emerged as particularly troubling?
    Because I hear that the problems aren't new, but the threat 
is new. And we're living in a different world and we have to 
develop an infrastructure within the intelligence community 
that is especially agile to address the asymmetric threat.
    And what I see is interorganizational misfirings. And, you 
know, I just believe that we've gotten to a point that we're 
bogged down within a bureaucracy. And I frankly think the 
creation of a DNI is a way of breaking down the bureaucracy and 
the barriers that currently exist.
    We have 15 agencies across the board, and you have one 
agency that, oh, by the way, oversees the rest of the 
intelligence community, and yet its hands are tied to 
exercising any authority over the other 85 percent of the 
budget. And so I just don't see how within the current system, 
given all of the problems that have been revealed as a result 
of our report, how you can change it--maybe on the margins, but 
not in a fundamental way that will bring about the need for 
real reform.
    And I think that is my concern, is that what I found was 
most revealing is that across the board there were some serious 
systemic problems that I don't believe can be changed by, sort 
of, you know, making some marginal restructuring efforts 
without total reform. Not to mention, I think, that the 
Director of the CIA being dual-hatted, has to manage the day-
to-day operations of an agency and then also the entire 
intelligence community and be the principal adviser to the 
President of the United States.
    So I would ask you, how did you see this report? Was there 
anything, you know, that was particularly unusual to you, 
troubling, surprising?
    Mr. Woolsey. Senator Snowe, I have read the conclusions and 
I have read chapter 12, the Iraq-al Qa'ida issue, in which I 
have a particular interest. But I've only read the conclusions 
on the weapons of mass destruction issues. So let me just say a 
couple of words about the Iraq-al Qa'ida question, because I 
think the committee dealt with this subject quite well in this 
    It is a very important issue. And it has become suffused, I 
think, with a lot of confusion in the press. The committee 
talked about a dozen or so reports of chemical and 
bacteriological warfare training of al Qa'ida by Iraq and said 
that this was ``the most disturbing set of reports.''
    It had other reports about evidence of combat training, 
bomb making, chemical, biological terrorism, false passports, 
safe haven. And it assessed the relationship between al Qa'ida 
and Iraq as not an alliance and not formal, but that ``Saddam 
was not averse''--and these are quotes--``to enhancing Usama 
bin Ladin's operational capabilities, although he didn't 
endorse al Qa'ida's overall agenda. It was not a close 
relationship, but it was a tactical one. The mutual suspicion 
was suborned by al Qa'ida's interest in Iraqi assistance and 
Baghdad's interest in al Qa'ida's anti-U.S. attacks.''
    I think that is a sophisticated and well-formulated 
explication of what has become a very, very messy issue in the 
public debate. People like Dick Clarke saying there was no 
connection of any kind, ever, between al Qa'ida and Iraq are 
clearly refuted by this. On the other hand, there is certainly 
no evidence the Committee points to of an Iraqi involvement, 
for example, in 9/11.
    But I would say the Committee has done a great service to 
this part, very important part, of the debate, by the way it 
expressed those issues.
    And as people begin to look now at the relationship between 
Iran and al Qa'ida in the 9/11 Report and elsewhere, the 
failure to stamp the passports for the eight terrorists and so 
on and so on, I think what we will find is that al Qa'ida had a 
tactical and occasional and training relationship and safe 
haven relationship of one kind or another with both of these 
    After all, in the Middle East, the enemy of my enemy can be 
my temporary friend. And I think that this Committee has done 
an excellent job of setting out the nuances of that kind of 
relationship and will, as more and more journalists finally 
read the report rather than just reading the conclusions, I 
think it will have a very positive and enlightening effect on 
the overall debate.
    Senator Snowe. General Odom.
    General Odom. Well, I would quickly say, I don't think the 
report is quite as tilted in the direction that Jim Woolsey 
says. I read it to say that there's really no operational 
    Mr. Woolsey. You really ought to read it more carefully, 
    General Odom. But let me go over the other points that I 
would answer more directly to your question. The surprising 
thing to me about the report is, as you put it, not the 
substance. The same substance could have been found in the 
1980s. I can think of a case to fit in almost all the places 
cited in the report. If you really want to shock yourself, get 
into a lot of case history in the FBI, and you will make the 
CIA look really good.
    Structural reform versus policy reform, and I think that's 
getting mixed up in the discussion here, and I would, 
therefore, like to make a distinction between the two.
    Some of the things you're saying we need to do concern 
policy changes, and yet you go back and expect structure to 
repair them. I think there is a connection, but let me clarify 
the difference.
    Let's suppose we have a ship and it's not in good repair. 
If we want it to sail well, we'll put it in the dry dock and 
we'll fix it. Or we can say, the Cold War's over; therefore, we 
need to sail to a different port. Sailing to a different port 
is a policy issue, not a structure issue.
    If the ship will sail to one port, it will sail to another 
port. So I don't think the terrorist problem or these other 
things are the structure issue. A good news service is not 
broken because the news stories change. It can shift. And the 
intelligence community can do this.
    Now, some of the shifts are difficult. Take the ones that 
Jim Woolsey cited earlier--language training, et al. It takes 
years to change some of these things. It takes years to develop 
capabilities in other regions. So you want to keep that in mind 
and realize that structural changes can help you with some 
things, but they can't help you with others.
    To me, if I had to simplify, I would say that there are two 
places where structural change is imperative and can have a 
positive effect. I repeat, it won't ensure it, but it'll make 
it possible.
    Separating the DCI from the Director of CIA role is 
essential before you can ever develop a program budgeting 
system that allows you to see what the monies buy. It will also 
allow you to pull the Director of Intelligence out of CIA and 
to put it inside this Director of Central Intelligence, or DNI. 
They will be less likely to compete with other intelligence 
agencies for analysis, but acting as their supervisor, their 
mentor, leading a collective and constructive effort.
    The other essential structural change concerns the FBI. No 
one has ever shown me how a law enforcement agency can create 
an intelligence culture. The FBI's poor CI performance is not 
the fault of the people there. To demand that the FBI carry the 
CI responsibility is like asking the Redskins to play in the 
American Baseball League. You're just asking FBI personnel to 
do something that they're not trained to do. And I understand 
the reluctance, politically, to step up to this problem, 
because people say, well, we won't tolerate a spy agency, a 
domestic spy. We have one. It's the FBI.
    It's not whether we have one, it's whether it's properly 
overseen and whether it's effective. And if you create a 
separate CI service, I think it ought to have dual oversight by 
the Judiciary Committee as well as this committee.
    So I actually think you could put us in a better position 
for domestic spying with a national counterintelligence service 
that's separate than we now have with the FBI. Thank you.
    Senator Snowe. Thank you.
    Dr. Hamre. Senator, may I speak to the question of the 
budget issues that you raised?
    Senator Snowe. Yes.
    Dr. Hamre. I used to run the budget for the Defense 
Department. I was the comptroller for four years. We had a big, 
very elaborate system. We produced around 40,000 pages of 
printout every night, you know, for the system--thousands of 
people working on it for six months. And it all built up to an 
hour and a half meeting with the Secretary to look at 12 
    So the system of oversight is not strong through the budget 
process. And, indeed, it's a system that lets you perfect 
whether or not your obligation rates are right, and we have too 
much outlays, or this sort of thing. But it doesn't really 
demand a qualitatively different performance on the 
    The system we had for budgeting hasn't changed for 50 
years. That wasn't what Goldwater-Nichols was about. Goldwater-
Nichols transformed the department when we elevated people in 
the department and made them demand better quality out of our 
military. That's what I think is missing in all our discussions 
about intelligence reform.
    Bringing the budgeting process around the DCI and let him 
manage a great, big elaborate process isn't going to produce 
better intelligence; it's going to provide a more elaborate 
system for our giving over inputs.
    You've got to get a system that reaches down and says how 
good is this analysis. Is it really meeting my needs? Why can't 
it be better? I have to have it better if I'm going to serve my 
responsibilities on behalf of the President.
    We need to democratize the demand process for intelligence 
product. That's what I would ask that you would consider.
    Senator Snowe. Thank you.
    Chairman Roberts. Senator Hagel.
    Senator Hagel. Mr. Chairman, thank you. Gentlemen, thank 
you. You have offered invaluable counsel and wise counsel. And 
as the chairman noted, we will call upon you often because you 
have helped--I can speak for myself only--frame these issues to 
get us a little more focused on what our responsibilities are 
as to how to implement whatever reform and change, 
restructuring we are going to implement. So thank you.
    I want to go back to a point in your testimony, Secretary 
Hamre, and just briefly read it back and then ask a question 
based on this, and then ask the three of you to respond.
    And here's what you said in your testimony regarding 
competition: ``To ensure competition among analysts is very 
important. To accomplish this we need redundant analytical 
capabilities in our intelligence community. We need competing 
organizations that report to different bosses in the federal 
government so that we profit from the competition that is 
inherent in bureaucratic politics.''
    Now, I think this is a very, very important point that we 
are gliding over the top of in our initial analysis here and 
the legislation that's been introduced. There will be more 
legislation introduced.
    But I'd like to go a little deeper into this, and 
particularly from the three of you--obviously I want to get the 
other two witnesses' thoughts on this, they may not agree with 
you on this. But the question would be: How does this 
competition and how does this interaction then get--and it's, I 
think, the central point here--get analyzed, processed and into 
the hands of the policymakers on a timely, real-time basis so 
that they can do something with this intelligence?
    I mean, after all, it is, as you note, and the three of you 
have said, it's the demand side. It's those who are charged and 
accountable to doing something with it. So how do we get it to 
them? And this competition piece that you think is important, 
how does that then work through whatever we may do or want to 
do in the future with our intelligence community?
    Thank you.
    Dr. Hamre. Senator, again, I need to defer to my colleagues 
who have been leaders in this system, and I have not. I've been 
an outsider looking into it.
    First of all, my goal is to get competition for strategic 
intelligence insights, not necessarily tactical developments. I 
don't think we want two competing SIGINT systems on the 
battlefield, that sort of thing. I'm not advocating that.
    But when it comes to making major national choices and 
decisions--is this an imminent threat, is it getting worse, is 
this something we have to act on now, what consequences would 
flow from it--that level of decisions, we need lots of 
different brains thinking about that, competing with each other 
for stronger ideas.
    We try to do that through the National Intelligence Council 
and by developing a national intelligence estimate. I think 
that has varied in its success. The NIC has on it members who 
are proponents of their organizations. There's a reason for 
doing that. You want to do that. They're the experts. And you 
want that competition.
    We probably also need some objective readers of the NIE. Is 
this really meeting my needs if I have to execute on this, if 
I'm trying to carry out a policy directive, do I know enough? 
Is this sufficient guidance? Is this adequate for me to make a 
decision and then direct my forces?
    I suspect that the process has had its ups and downs. I 
think that it's become a bit rigid, rather stilted in some 
dimension. And frankly I think the shortage of dollars through 
the 1990s led too much of the community to chase after the same 
rabbit, you know, hoping they could get a little bit more money 
from OMB with the next budget review.
    So I think part of it was scarcity of resources.
    But I think you need institutional venues where you have 
these multiple voices that are helping interact and come to a 
consensus on what should we be thinking about and what 
alternatives do we present to the policymakers.
    I think the National Intelligence Council is supposed to be 
doing that for us. We probably need to strengthen it. I'd like 
to defer to my colleagues who are actually inside that and have 
actually sat on the NIC to really answer that question.
    General Odom. The point that John Hamre made about time 
sensitive or non-time sensitive is important. And those are two 
different worlds. And you understand why you can't have a lot 
of competition out at that level.
    I have mixed feelings about the competition. The 
competition is okay, if you have open-minded people who will 
accept new evidence which bears on their conclusions. But what 
do you do when you have competition where both sides refuse to 
accept any new evidence that would change their bottom line?
    And my experience in the National Foreign Intelligence 
Board back in the 1980s was that DIA, CIA and other agencies 
were locked into positions so that new evidence was just not 
going to change their views.
    I'll give you an example. I was briefed once by DIA--we 
were going over either an NIE or something--about the Soviet 
Union, whether we could affect them by denying technology 
transfers. DIA's position was: of course we can. We have to 
block technology transfers. But at the same time, they insisted 
on saying that there's nothing we can do to keep them from 
producing as many missiles as they want.
    I said, well, you can't have it both ways. If they're 
unconstrained in resources for building these missiles, then 
cutting off the technology is not going to make much 
difference. So why oppose technology transfers? Well, analysts 
get locked into those kinds of positions. You had similar kinds 
of inconsistencies on the CIA side.
    I'll give you another example. People always argued about 
how big the Soviet defense budget was. I never understood why 
we asked the question. We're not worried about being attacked 
by rubles or dollars. We're worried about being attacked by 
rockets, airplanes, tanks and these sorts of things. The real 
question is how big the forces were and what their capabilities 
were, not the rubles that they put into it. And we had a pretty 
good track record on what we knew their forces to be.
    Another one that I think started over here with Senator 
Moynihan and the Congress was that we failed to predict the end 
of the Cold War. Well, it depends on why you think it ended. I 
happen to have done a fair amount of investigation into it, and 
I think it ended because Gorbachev decided to end it. And if 
he's a man that has free will, by definition we can't predict 
how he will exercise his free will. And, therefore, to have not 
known that the USSR would come apart on the 31st day of 1991, I 
don't feel the intelligence community can be blamed for that.
    There's another aspect to that question, though. What would 
we have done differently between 1985 and 1991, at the end of 
the year, if we had known that date in advance? We pulled off 
the largest strategic realignment in Europe, the reunification 
of Germany within NATO, the collapse of the Warsaw Pact, in the 
history of Europe without a war.
    So I don't know that the intelligence community's failure 
to predict the end of the Soviet Union is relevant to anything, 
but we've wasted a lot of time blaming CIA for failing to 
predict the collapse.
    I would just finish by saying in my proposal for how I 
would staff a separate DCI, I'd want him to take with him and 
slim down the Directorate of Intelligence out of CIA, because I 
think one smart analyst will always beat 12 mediocre analysts. 
In fact, intelligence insight is probably inversely related to 
the number of analysts at work.
    A smaller DI attached to the DCI's National Intelligence 
Council would be in a position where it would not be competing 
with DIA, INR and others, but act more like a dissertation 
supervisor, who will turn to DIA and INR or the Army and say, 
this is an issue you ought to be looking at. They may say, we 
don't have time. Well, the DCI can tell his own analytic 
element at the DCI level, go look at that issue. If it proves 
interesting, the DCI can turn it over to the appropriate all-
source analysis center, give it some additional money, and 
direct them to continue to handle it as a ``national'' level 
    Then I think you will get a more intellectually 
dispassionate climate within the intelligence community, where 
people are willing to hold the evidence up and say, does this 
really bear on our conclusions?
    Now, I've been a professor for a number of years, and 
I'veseen the academic debates. And sometimes you can wonder if 
political science departments, sociology departments and history 
departments increase enlightenment by competitive scholarship or 
whether they just bog it down into theological disagreements. So, yes, 
I want some competition, but I also cite these examples of competition 
as warning flags.
    Thank you.
    Mr. Woolsey. I think there is a fair amount of competitive 
analysis in the current system. There really was in the famous 
NIE. It's just that the key judgments of the shortened version 
eliminated a number of the caveats. And I haven't seen the 
classified version, but I understand from various reports that 
if one reads the whole NIE, one does get a much better idea of 
the issues on which people disagreed, such as the mobile 
laboratories or whatever.
    So the system in its current form does have some important 
competitive analysis. We tried to heighten that when I was DCI.
    I was very lucky to be able to persuade Joe Nye, then and 
later dean of the Kennedy School at Harvard, to come down to be 
the head of the National Intelligence Council for me. And Joe 
and I worked out a system of encouraging people to use 
gambler's odds, for example. Say they thought there was 
probably a 1 in 10 chance of something or a 2-1 chance, not 
that you can run history twice and tell whether--or more than 
once--and say whether you're right or not, but it gives a feel 
for whether you think it's a high probability or low 
probability, but still a dangerous situation.
    We tried to operate the National Intelligence Council in 
such a way as to get NIEs drafted so they were educating 
national leaders--for example, on the way the drug trade 
worked--rather than predicting the price next year of cocaine 
on the streets.
    Joe was very good at this, and I think the kind of 
leadership you have on the National Intelligence Council itself 
can help one come up with competitive analysis, even under the 
current system, that educates better than trying to reach 
single conclusions. It's that thirst for a single conclusion 
and giving an answer to a decisionmaker that often leads--it's 
almost always what leads to intelligence failures.
    I mean, I'd take one example from something Bill said. 
Contrary to him, I think it was very important what the size of 
the Soviet GDP was, because that would have enabled us, if we'd 
done a better job of assessing that, to have a better feel for 
how much stress the huge Soviet military expenditures were 
putting on the system.
    If one had a system that encouraged competitiveness, even 
if, let's say, the DCI agreed with Bill, rather than me, and 
said, look, I don't care about the size of the Soviet Union's 
GDP, what I just care about is its military capability, if you 
had a system with competitive analysis, someone in there would 
be saying, well, look, there may be a low probability that this 
is going to affect the overall outcome of the Cold War, but 
it's a very important question, even with a low probability, so 
here's my analysis of that.
    I think the current system can work to give you competitive 
analysis and a better job of educating the senior government 
leaders than apparently happened this time around. It would be 
wise to look for people like Joe Nye to be the head of the 
National Intelligence Council.
    General Odom. Can I just add one point? You know, not only 
did Joe Nye do what Jim Woolsey said--and I think that the 
cases that he described are really instructive--but he also 
farmed out to nongovernmental centers tasks to do parallel NIES 
to those done by the NIC. Some of them--I remember one on 
Europe--turned out to be really quite on the target. So there's 
another dimension there.
    And I also will accept the point here that there were 
really important reasons to know how big the overall Soviet GDP 
was. I would just emphasize that it too often was made the 
issue for either lowering or raising the U.S. defense budget. 
And I didn't think that made good sense.
    Dr. Hamre. And could I just say I think Ambassador 
Hutchings, who currently heads the NIC, is really trying to do 
a very good job of trying to bring strength and competition of 
ideas into the NIE process. So this is a place where you want 
it to happen. But it has to be there. You need a very strong 
competitive environment in my personal view.
    General Odom. But you did get an attitude from the Director 
of Central Intelligence that encouraged a diverse set of views 
in a non-hostile or non-threatening way. And that's what 
produces the competition in intelligence that I think will 
really be good.
    Chairman Roberts. Senator Hatch.
    Senator Hatch. Well, I want to thank you--the three of 
you--for being here. Mr. Woolsey, you were one of the better 
DCIs, I think, one of the best in my experience in 28 years. I 
appreciate your service.
    Mr. Woolsey. Thank you.
    Senator Hatch. And, General Odom, I recognize the great 
service you gave there and, Dr. Hamre, your abilities.
    Could we do a lot of this by changing some of the policies 
without overlaying another layer of bureaucracy that may or may 
not work? It's nice to be able to have change, but couldn't we 
change some policies? Wouldn't that be a quicker, better 
methodology? I don't know.
    General Odom. Can I answer your first one?
    Senator Hatch, I think you were out when I made the point 
    Senator Hatch. Yes.
    General Odom. I think you can separate the DCI from the 
director of CIA by executive order--not pass legislation. You 
don't create a new layer of bureaucracy. The bureaucracy is 
already there. The DCI has a staff--a community staff. If you 
look at the proposals I've laid out, I don't create a new level 
of bureaucracy. But I do separate it from CIA.
    And I would say that I don't think very important changes 
can continue or can be made if you don't separate the two jobs. 
But that's not to say you have to do it by legislation. And as 
I said in my testimony, I think there are dangers in some 
aspects of legislation. If you made it law as the draft bill is 
now written, I can see serious difficulties arising from it.
    Senator Hatch. I guess one question--and you probably 
covered this, as well. I'm sure you covered a lot of 
thingsbecause I had to go meet with the military for an hour.
    I guess what I'm trying to get to is this: I'm having 
difficulty trying to change the tops of these organizations 
when it seems to me the one part of intelligence that really 
needs a direct czar, who reports to the President and maybe the 
DCI, is a person who's over the analytical section--the various 
analytical sections--and who really is independent so that the 
analysis really--it has to be independent to be effective.
    And I'm just wondering if that wouldn't be a better change 
than trying to just put another person who's over everything at 
the top. I admit that budgetary considerations are very 
    But wouldn't it be important to kind of create an 
analytical approach that has an independence to it that 
transcends politics and has the leader over it that would have 
to report to other people, but still independently can run the 
analytical processes of intelligence?
    Mr. Woolsey. Senator Hatch, I think the independence won't 
come so much from having a separate reporting channel, as it 
will from the President and the Senate selecting people who are 
willing to call it the way they see it and let the chips fall. 
I realize that it's not always easy to tell who that's going to 
    But a number of DCIs have done that. Dick Helms, quite 
famously now from some of the memoirs, very early in the 
Vietnam War, went to President Johnson very privately with an 
extraordinarily critical assessment of our prospects in 
Vietnam. And President Johnson accepted it and thought about it 
and used it. It didn't ultimately make the decision for him of 
what to do. But a number of DCIs have called it the way they 
saw it. I tried to. I think a number of others have.
    And there's a big negative side to splitting the analysts 
out of the rest of their organization. I tried, in a way, to 
begin integrating the Directorate of Operations and the 
Directorate of Intelligence a lot more so they could share 
language training and understanding of sources. And we went so 
far as to distribute the art from different parts of the world 
around into their common areas in the CIA.
    I, for better or worse, tried to head things in a different 
direction. I tried to integrate analysis more with the people 
who had feet on the ground in a country in a region, quite 
frequently spoke the language well.
    And I think that there's ultimately not going to be any 
guarantee. One can try with the fixed terms and so forth, but 
ultimately, there's not going to be any guarantee other than 
having the head of these two or three key agencies and the 
overall head of the community be people who are willing to call 
it the way they see it. And I don't know that there's any 
organizational shortcut to make that happen if the propensities 
of those individuals are not that way.
    Senator Hatch. So you think it comes down to the choice of 
who manages these matters?
    Mr. Woolsey. I really do.
    Senator Hatch. And you don't think there's any analytical 
process or procedural way that we can help?
    Mr. Woolsey. Whether there might be, but I haven't been 
able to chance on one. The best way I can figure out how to do 
it under the system we've got now is to get someone, as I said, 
with the creativity of a Joe Nye to help pull all this together 
in a way that would push analysts toward educating 
decisionmakers so they can make judgments as distinct from the 
analysts sort of secretly going away and putting everything 
into a document which came up with a conclusion.
    Joe and I kept fighting against single conclusions. And I 
think sometimes people say, you darned intelligence analysts, 
can't you tell me what you think? Can't you tell me what's 
going to happen? And the answer is no. This is reality. Quite 
frequently one cannot tell people what's going to happen.
    In the example I used before, if we'd recruited a dozen 
Iraqi generals just before the war they might, each one, as 
David Kay suggested, tell us that they weren't going to have 
chemical weapons in their unit, but the unit to the right and 
the unit to the left did.
    There's not going to be a way to get through this and end 
up with an intelligence community that is going to save work 
for senior decisionmakers and give them an answer in different 
    What you hope, I think, the intelligence community can do 
is educate people about what the issues are a lot better than 
they do now, give them a judgment about what may be more likely 
than not and a judgment about the consequences of what would be 
less likely to happen but could still happen.
    I think that's the best you can do.
    Senator Hatch. Thank you. Thank you, Mr. Chairman.
    Chairman Roberts. I thank all the panelists for your 
patience and for your service. And as I've indicated again, we 
are likely to have you back. It might be a more informal 
setting, although this is now getting to be pretty informal.
    But we thank you for your contribution, and the hearing is 
    [Whereupon, at 5:48 p.m., the Committee adjourned.