March 29,2005

The President
The White House
Washington, DC 20500

Mr. President,

You asked this Commission to advise you on the adequacy of plans produced in response to your memoranda to the Attorney General and the Director of Central Intelligence dated November 18, 2004. This letter conveys our views. In brief, we do not believe that either response is entirely adequate, and both responses show that these agencies remain too comfortable with a “business as usual” approach to intelligence gathering. They show, in short, just how important—and how difficult—Ambassador Negroponte’s job will be.

(In a separate Memorandum, you sought the advice of the Secretary of Defense and the Director of Central Intelligence on paramilitary operations and invited our comments on their report. This report was due in mid-February but has not yet been submitted; we plan to provide comments when we receive it.)

Federal Bureau of Investigation

Your memorandum to the Attorney General approved and called for implementation of the 9/11 Commission’s recommendation that the Federal Bureau of Investigation establish “a specialized and integrated national security workforce.” You specifically directed the FBI to “allocate sufficient resources and authority to the new Intelligence Directorate to perform its assigned mission” and take other measures to improve intelligence collection, analysis, and dissemination. You asked the Bureau to produce “a comprehensive plan with performance measures” for achieving measurable progress.

The FBI responded on February 16, 2005, with a Comprehensive Plan for the FBI Intelligence Program with Performance Measures (the “FBI Plan”). To all outward appearances, the FBI Plan is what you asked for. It contains a detailed account of all major aspects of its intelligence program, including progress reports, initiatives underway, performance objectives, and quantified metrics. However, from a broader perspective, despite its detail, the plan fails to create a truly “specialized and integrated national security workforce.” Thus, the FBI Plan is inconsistent with the core recommendation of the 9/11 Commission regarding the FBI.

We do not mean to say that the FBI has done nothing. Far from it. The FBI has taken some commendable steps in the direction of an integrated national security workforce. It has given the new Intelligence Directorate various authorities over policy, budget, training, and analysis. In essence, the Directorate of Intelligence functions as an overlay on intelligence activities that are managed by other elements of the FBI. It establishes intelligence requirements and passes them to FBI’s field offices for execution. However, unlike the counterintelligence and counterterrorism divisions, the Directorate of Intelligence does not control operational resources. Nor does it directly control the Field Intelligence Groups (FIGs) that, according to the FBI, “manage and direct all field intelligence operations.” (FBI Plan at 15.)

As our Report concludes, the Directorate’s lack of authority prevents the FBI from vertically integrating foreign intelligence collection, analysis, and operations. We believe that all three national security missions—intelligence, counterintelligence, and counterterrorism—should be jointly managed at the strategic level and fully integrated in planning, targeting, and operations.

The FBI recognizes what is needed to integrate these three national security missions, and rejects it. The first page of the FBI Plan advances a counterargument, stating that its “core principle” will be “integrating law enforcement and intelligence operations.” (FBI Plan at 1.) As we understand it, the FBI’s reasoning is that the intelligence and law enforcement disciplines are mutually reinforcing. Specifically, the Bureau contends that in the area of domestic intelligence, the criminal investigator’s knowledge of legal limitations and procedures is an important carryover to the intelligence discipline that will help protect civil liberties. Also, in the area of counterterrorism, the boundary between criminal prosecution and intelligence interests may be blurred and shifting.

We agree that cross-pollination between criminal investigation and intelligence disciplines is a good thing. We also agree that all of the FBI’s arguments have some merit; but they all too easily become reasons not to integrate the national security missions at all. As our Report concludes, the FBI cannot fulfill its intelligence mission without integrated intelligence capabilities, cohesively managed from collection to dissemination. Establishment of an integrated national security service within the FBI would in fact enhance opportunities for cross-discipline training and experience. But the FBI proposes to “integrate” law enforcement and national security in a way that makes it impossible to establish an integrated national security workforce, which you called for in November and which we believe is essential to the security of this country.

Central Intelligence Agency

Your memorandum to the Director of Central Intelligence approved several recommendations of the 9/11 Commission. It required improvements in the CIA’s analytic capabilities, a transformed clandestine service, stronger language programs, more diversity among operations officers, better coordination between human and signals intelligence, and a better balance between unilateral and liaison operations. You directed the Director of the CIA to submit within 90 days “a detailed budget and implementation plan, including performance measures, with timelines for achievement of specific, measurable goals.” You specified that the Plan should include actions to “increase as soon as feasible”: (1) “the number of fully qualified, all-source analysts by 50 percent”; (2) “the number of fully qualified officers in the Directorate of Operations by 50 percent”; (3) “the number of CIA officers tested and proficient in mission-critical languages by 50 percent”; and (4) “the number of officers who are engaged in research and development” by 100 percent.

The CIA’s response (the “CIA Plan”), dated February 16, 2005, does not, in our judgment, adequately respond to your November 18 memorandum. Our specific comments are set forth below, but over all, the means and goals articulated in the Plan are too general to create accountability. The CIA Plan generally recites institutional aspirations that are well-known and most of which are contained in previous CIA plans. Our fundamental concern is that too little has changed as a result of your November 18 memorandum.

• Ensuring “diverse views” is critically important, but the section of the CIA Plan that deals with the topic lacks specific remedies, offering only general improvements in analyst training. This is a good thing, but it will not ensure that alternative views are reflected in analytic products. The Commission’s Report suggests numerous additional steps that can be taken to improve analysis and encourage analytic diversity, including routine critiques of finished intelligence, alternative assessments by outside experts, and “post-mortems” that in hindsight evaluate the accuracy and quality of estimates.

• With regard to numerical objectives, the CIA Plan adds a relatively small number of analysts in this fiscal year and estimates that the balance will be acquired in the “long-term” without further specification of what is meant by “long-term.” In our view, that is not a sufficiently detailed “timeline for achievement of specific, measurable goals.” The projections for an increase of “fully qualified” DO officers are even more troubling. The CIA Plan shows an insufficient increase in DO officers in FY 2005; it expects to approach the goal (without reaching it) by FY 2011.

• We are under the impression that by the term “fully qualified” DO officers, you meant case officers or operations officers who collect intelligence in the field. However, the numbers supplied in the CIA Plan include mostly Headquarters’ officers and other “non-DO support officers occupying DO positions.” (CIA Plan at 21.) We do not quarrel with the need for teams that support the operations officers; indeed, we expect that other costs—training, physical space, communications, and the like—will also increase. Although we would like to see improvement in the tooth-to-tail ratio, we question whether the CIA Plan is truly responsive to your direction. An increase of 50 percent in the DO’s numbers would still leave the CIA with a thin overseas presence, especially when there is need to surge in a particular area such as Iraq. This makes the failure to meet that goal all the more troubling. Aside from numerical objectives, the CIA needs to re-think the ways in which it deploys its operations officers and, further, it should examine alternative approaches to human intelligence that are less reliant on the traditional case officer. We have recommended that CIA establish an Innovation Center outside the DO. The Center would initiate, test, and evaluate new approaches to human intelligence and other new operational concepts. While the CIA has begun a small innovation initiative of this nature, we judge that it falls short of what is needed.

• The CIA Plan says little about open source intelligence. It promises that 30 percent of the work force will study things like “evolving [foreign] demographics and views of the United States” and “developments in [foreign] economies.” These are not secrets to be stolen by spies; they are classic open-source topics. The CIA must learn to use open sources more effectively. The Commission Report recommends a significant effort in this area, specifically, the creation of a new Open Source Directorate within the CIA.

• Some of the CIA Plan’s proposed metrics are troubling. While measuring inputs (e.g., staff and funding) has obvious limitations, measuring outputs can also distort incentives. For example, simply increasing the number of intelligence reports issued on counterterrorism and counterproliferation (CIA Plan at 25) may not improve intelligence on those topics. In contrast, and commendably so, the FBI Plan shows a serious effort to measure both the quality and quantity of intelligence products.

• The CIA Plan does not provide consistent measurable targets. Some parts of the Plan treat 2007 or 2008 as the target date; other parts set 2011 as the target. Others simply treat the target as the “long term.” To measure success, a single set of target dates should be chosen.

• There are some stronger sections in the CIA Plan. We found the responses related to R&D and to support services to be substantive. Notably, however, even these sections do not specify completion dates.

We recognize that CIA is pushing hard to do more of what it has done in the past and to do it better. While the Agency has begun to experiment with new approaches, it has done so timidly; its experiments are fragile and at risk. Rather than treat your Memorandum of November 18 as an opportunity to press forward with new ideas, the Agency has largely reported on what was already being done.

In sum, we believe that the incomplete nature of both responses illustrates the difficulty of bringing about real change in the Intelligence Community. Changing the way the CIA does business and bringing the FBI all the way into the Community are not goals likely to be achieved from within. As we emphasize in our report, these changes will require strong leadership from the DNI—and firm backing from above.

Our Commission’s report provides detailed recommendations aimed at achieving the goals of your Memorandum of November 18. We therefore suggest that you consider directing the DNI to oversee the development of follow-on agency plans, with budgets, timelines, and specific, measurable goals to implement any recommendations of this Commission that you may choose to accept.

Very respectfully,

(SIGNED BY)

LAURENCE H. SILBERMAN (Co-Chairman)

CHARLES S. ROBB (Co-Chairman)

RICHARD C. LEVIN

JOHN MCCAIN

HENRY S. ROWEN

WALTER B. SLOCOMBE

WILLIAM 0. STUDEMAN

PATRICIA M. WALD

CHARLES M. VEST

LLOYD CUTLER (Of Counsel)