By William E. Odom

3 October 2002

Good morning, Mr. Chairman, and members of the committee. It is an honor to appear here today.

You have asked me to share my views on the role and responsibility of the Director of Central Intelligence, the Secretary of Defense, and the FBI in dealing with terrorism. This is a very large set of topics. I have submitted a copy of an intelligence reform study, which I chaired and drafted a few years ago, as a comprehensive answer to your questions. The analysis and recommendations it puts forward, in my judgment, are all the more compelling in light of the events of 11 September 2001. I hope that this study, or parts of it, can be used as my written testimony. To be sure, I am also submitting a short additional written statement prepared especially for today to adjust the emphasis in the study to your specific interests in this hearing.

Those interests seem to be directed toward the structure of the Intelligence Community. If I am correct about that assumption, then I am encouraged. While it is important to know the details of how the intelligence failure of 11 September occurred and to assign some responsibility for it, it is far more important to take the opportunity to fix longstanding structural problems within the Intelligence Community. I certainly can offer nothing on the events leading up to 11 September of last year.

The issues of structural reform are too complex to explain comprehensively in a short statement, but it is possible to highlight three overarching issues for your attention.

The first concerns the orchestration of the intelligence process within the Intelligence Community. The second concerns management of resources, i.e., getting more intelligence for the dollar, and the third concerns counterintelligence, which is key for dealing with terrorism as well as hostile intelligence services.

Changing technology has produced a general trend in the Intelligence Community that has been delayed and blocked by bureaucratic turf concerns. Each of the three collection disciplines - signals intelligence, imagery intelligence, and human intelligence - is very different. Each needs a national manager to orchestrate collection operations.

The trend most advanced is toward a national manager in signals intelligence. The director of NSA comes close to having the authorities and means to be its national manager for signals intelligence. In imagery intelligence, the director of the NIMA is the proper candidate for that job, but his agency is very new, and his authorities and means are not yet adequate. Turf fights prevent the trend coming to fruition in imagery intelligence. In clandestine human intelligence, the CIA's Directorate of Operations has long had the authorities but shown no interest in being the national manager of the capabilities within the Defense Department.

As long [as] the DCI is double-hatted as the director of the CIA as well, he cannot stand above the Intelligence Community and carry through the creation of fully empowered national managers for all three kinds of collection.

Turning to the second issue, getting more intelligence for the dollar, the DCI is the program manager for all the budgets within the Intelligence Community. This is potentially a very powerful authority, but given legacies within the CIA, dating back to 1947 and earlier, the CIA does not want to see its authority used for more efficiency.

Lacking national managers for the three collection disciplines and also for counterintelligence, the DCI has no subordinates who can rigorously relate inputs of resources to outputs of intelligence. His executive management organ, the Intelligence Executive Committee, includes the senior intelligence managers, but none have the control over programs that allows the DCI to hold them accountable for presenting and "Planning Program Budget" analysis, the kind that has been used in the Pentagon for forty years. If there were three national managers of the collections disciplines with full program authority over the resources spent in their disciplines, they could present a proper program budget to the DCI that shows the effects that various cuts and increases will have.

The biggest stumbling block to achieving this kind of national manager system is the National Reconnaissance Office. As a procurement organization, it spends a large part of the money allocated for signals and imagery intelligence, thus preventing the directors of NSA and NIMA from being able to trade off NRO projects against other signals and imagery projects. As long as this is the case, the waste in intelligence spending will be very large.

Finally, to the third issue, counterintelligence. It is in the worst shape of all. Five organizations run counterintelligence operations with no overall manager - the FBI, CIA, and the three military services. The parochialism, fragmentation, and incompetence are difficult to exaggerate in the US counterintelligence world. This has become publicly clear to anyone following the reporting on the FBI and CIA over the past several months. It is not new. It has long been the case, right back to World War II andthroughout the Cold War. The combination of fragmentation - which leaves openings between organizations for hostile intelligence operatives to exploit - and lack of counterintelligence skills insures a dismal performance. And terrorists, like spies, come through the openings.

The skills problem for US counterintelligence derives from mixing law enforcement with counterintelligence. Spies will always beat cops. The record of the FBI during its entire existence is a painful and irrefutable evidence of that truth. The same is true for Navy and Air Force CI, which are inside criminal law enforcement agencies.

The first step in creating an effective counterintelligence capability, therefore, is to take the CI responsibility out of the FBI, leaving the Bureau with its law enforcement responsibilities, and to create a National Counterintelligence Service (NCIS) under the DCI and with operational oversight over the CI operations of the CIA and the three military departments in the Pentagon.

This proposal has been called the "MI-5" solution, modeled on the British MI-5 organization. My version is not. It is quite different. First, a NCIS would have oversight over the CIA CI and the Pentagon CI operations, which MI-5 does not have over MI-6 and the defence ministry. Second, I would not give the NCIS arrest authority. That can be left to the FBI and other law enforcement organizations. Third, it would be under the DCI for overall program management and direction for providing CI support to all agencies of the US government, including, of course, the Department of Homeland Security.

To sum up, I propose three major reform directions:

First, separate the DCI from the Director of Central Intelligence, giving him organizational support, and create national managers for the three collection disciplines.

Second, implement a Planning Program Budgeting System within the intelligence community that better relates dollar inputs to intelligence outputs.

Third, create a National Counterintelligence Service under the DCI.

I will be delighted to take your questions and fill in details for this general picture of which there are a very large number.

PDF Version