Fourth public hearing of the National Commission on Terrorist Attacks Upon the United States

Statement of John Deutch to the National Commission on Terrorist Attacks Upon The United States
October 14, 2003

"Strengthening U.S. Intelligence"

Mr. Chairmen and members of the commission, during President Clinton's first term, I served successively as Undersecretary of Defense for Acquisition & Technology, Deputy Secretary of Defense, and Director of Central Intelligence. As Undersecretary of the Department of Energy in the Carter Administration my responsibilities included our nuclear weapons program and the Department's nonproliferation efforts. During 1997 to 1999, I was chairman of the Commission to Assess the Organization of the Federal Government to Combat the Proliferation of Weapons of Mass Destruction; our report was submitted to Congress on July 14, 1999. I am currently an Institute Professor at the Massachusetts Institute of Technology.

After the attacks of September 11, we owe the victims' families and all American citizens the assurance that the Intelligence Community (IC) is in the best possible position to defend against possible future catastrophic terrorism. In my judgment, we need major changes to accomplish this goal. My purpose today is to propose for your consideration a major realignment of intelligence community authorities and responsibilities that will lead to better intelligence to protect us from terrorist acts and to strengthen more broadly our national security.

I propose two major changes: first, to the relationship between the IC and the Department of Justice (DOJ) and Federal Bureau of Investigation (FBI); second, to the relationship between the IC and the Department of Defense (DOD). These changes require a consensus on a new balance of responsibilities and authorities among the agencies that make up the IC, as well as new procedures for executive branch and congressional oversight. These change to current policies and practices are not insignificant and therefore will meet considerable resistance from those parties participating in the present arrangement; neither the executive nor the congressional branches of our government readily yield even a small part of their writ, e.g. consider the struggle over the establishment of the Department of Homeland Security (DHS). However, without a significant restructuring of responsibilities, I doubt that sufficient improvement in intelligence is possible.

It should not be surprising that change is necessary. The basic alignment of responsibilities and organization of the IC has been in place for over fifty years, since the 1947 National Security Act established the Central Intelligence Agency (CIA). During the subsequent decades, the IC evolved as the Cold War dominated geopolitics, and the United States faced very different threats from those it confronts today. U.S. intelligence organization was based on several distinctions, which have become increasingly blurred: that there is a bright line between wartime and peacetime; that threats are either foreign or domestic; and accordingly, that threats should be considered either a matter for law enforcement or national security.

Catastrophic terrorism and advances in information technologies that permits intrusion into information networks from a distance, invalidate these distinctions, as non-state groups threaten to employ destructive means, potentially including weapons of mass destruction. The creation of the DHS, dedicated to protecting the nation from further terrorist attacks, is a step toward resolving this ambiguity in some respects - border security, crisis response, infrastructure protection -- but not with respect to intelligence.

At present the Director of Central Intelligence (DCI) has very limited executive authority over the IC. I define "executive authority" (or line authority) to include (1) planning and allocation of resources, and (2) management of day-to- day operations. The DCI has executive authority over the CIA but has no executive authority over any of the other intelligence agencies involved in combating terrorism. The DCI has limited budget authority over intelligence expenditures [Most of the intelligence programs are in the Department of Defense (DOD) budget.] and the DCI has few tools to compel integrated planning of intelligence operations.

I believe the DCI needs greater authority over domestic and foreign counter terrorism intelligence activities. In particular, I believe the DCI should be given executive authority for domestic intelligence activities related to terrorism and other national security matters as well as planning and budgetary authority over intelligence activities of the DOD. Such additional authority will over time permit much more effective integration of our intelligence efforts by agencies of the IC.

The single most important change I propose is to establish a new Domestic Intelligence Service (DIS) for intelligence activities undertaken in the United States and for collection of intelligence where U.S. persons may be involved. The core of the capability for this new organization would come from the transfer of the Intelligence and Counter Terrorism Divisions of the FBI (These activities were part of the former National Security Division of the FBI.) to the DIS, including an appropriate number of FBI special agents. The DCI should have executive authority over the DIS, just as the DCI has control over the CIA today.

Currently, the FBI has responsibility for intelligence activities undertaken within the U.S. and for counter terrorism targeted against U.S. persons and the U.S. homeland, while the CIA has responsibility for intelligence collection abroad about foreign security threats to the United States. But, since the terrorist threat is likely to involve both non-U.S. and U.S persons and activities at home and abroad, the traditional distinctions between foreign and domestic no longer hold, and the need for closer cooperation between the FBI, the CIA, and the other IC agencies becomes more urgent.

The pivotal issue is whether FBI intelligence activities should be viewed as a national security or law enforcement mandate. These two objectives are quite different. If the purpose is law enforcement, then the effort is investigative, aimed at building a case to present in a court of law, in order to bring the perpetrator of a crime to justice. Information collected in the course of a law enforcement investigation is held in custody, as evidence to be presented in court. In contrast, the national security objective is to protect the American people. Information is collected with the aim of providing warning or supporting an action to interdict terrorists before they strike. The procedures and practices differ quite significantly between these two approaches. The expertise and methods necessary to build a criminal case are quite different from the expertise in collection and analysis necessary for warning.

Though I understand that progress has been made in recent years, barriers to coordinating the planning of intelligence collection and to the smooth flow of information between the FBI and the IC remain significant. And the underlying tension remains. The FBI fears a compromise of on-going criminal investigations, and the CIA fears that sensitive sources and methods may be revealed in court proceedings. Significantly, the Terrorist Threat Integration Center (TTIC), established in May 2003, promises to be colocated with the FBI Counter Terrorism Division and the DCI Counter Terrorism Center, but the director of the TTIC has limited authority over the two agency centers. Moreover, the FBI has no tradition or capability for intelligence analysis or technical intelligence collection and analysis (DNA testing is an exception.).

The result of this difference in approach and organization is that the U.S. government is not in the best possible position to integrate both domestic and foreign intelligence to provide warning and protection from terrorist attack. The apparent lack of effective exchange of intelligence information obtained before the September 11 attacks is testimony to this shortcoming.

The fight against terrorism demands that the higher priority be placed on intelligence for the national security purpose of warning and defense then for law enforcement.

The DCI should be responsible for setting the collection priorities for the DIS and for receiving and disseminating information to policy users for the purposes of warning or intelligence threat analysis. The DIS would not receive information directly from grand jury or other U.S. court proceedings, nor would DIS information be directly usable in criminal prosecution of U.S. persons. The DCI and the new DIS would have authority to collect information about and maintain data bases on U.S. persons, but only when following rules established and enforced by the Department of Justice. In effect, the role of the DOJ would be changed from active involvement in intelligence collection and analysis to assuring that the rules bearing on the rights of Americans are scrupulously followed.

The strength of this proposal is that all counter terrorism intelligence activities would be centralized under the DCI, thus permitting a better integration of collection efforts and intelligence analysis of information from a security, not law enforcement, viewpoint. The disadvantage of the proposal is the risk that the IC will misuse information collected about Americans (as occurred before the reforms of the mid 1970s). But, this shortcoming is shared, to some extent, by all proposals to strengthen domestic intelligence and underscores the need for strong, independent oversight of intelligence activities by the Department of Justice.

An alternative approach would be to house the DIS in the DHS or create it as a separate agency. The disadvantage of both these approaches is that integration of domestic and foreign intelligence planning and operations would be as difficult to accomplish with today's division of responsibility between the FBI and the CIA. A separate DIS agency would still need to report to an executive branch official, but it is by no means clear who that should be.

The second most important step to strengthen our intelligence capacity to combat terrorism and related national security threats (such as proliferation of weapons of mass destruction) is to change the balance of authority between the DCI and the Secretary of Defense. At present the DCI does not have the authority or responsibility to plan or allocate resources across the Intelligence Community. There is no integrated plan coordinating the efforts of the separate agencies, notably the CIA, FBI, and the numerous intelligence agencies of the DOD. This shortcoming is most serious with regard to the DOD's direct control of many of our most important collection assets, especially signals intelligence through the National Security Agency (NSA). Just as separation of domestic collection in the FBI and foreign intelligence in the CIA limits our intelligence capability, the separation in planning and resource allocation of foreign signals intelligence in the NSA and human intelligence collection and foreign intelligence analysis in the CIA (and DOD's Defense Humint Services) limits our capability to integrate all that we know and thus produce the most effective intelligence.

The various Intelligence Community "centers," such as the counter terrorism and counter intelligence centers, are important mechanisms for improving coordination. But, without the DCI having greater authority, the centers have limited capacity to set collection priorities and make them stick. For example, the DCI has authority over the National Foreign Intelligence Program, (NFIP), but even this authority is limited. In the past, agencies receiving NFIP dollars have reprogrammed resources into activities with greater relevance to the agency mission, thus diminishing the national character of intelligence planning.

With respect to the Department of Defense, the DCI should be given planning and budgeting authority over the Joint Military Intelligence Program (JMIP) and parts of the Tactical Intelligence and Related Activities (TIARA) program. The DCI should establish the resource allocation plans for all intelligence activities that support national security objectives, such as counter terrorism and counter proliferation, as well as intelligence support to military operations. The DCI's authority should include collection activities from airborne, space, and naval sensors, but equally importantly, all activities related to tasking, processing, exploitation, and distribution (TPED) of this information, to assure that timely intelligence reaches appropriate levels of national and military command. Such a strong, integrated approach to intelligence is a necessary condition to capitalize on advances in technology that can give our national leadership and joint military commanders in the field a tremendous advantage from access to near real time intelligence.

Under this proposal, the DCI would have greater authority for the planning and budgeting of several DOD intelligence agencies, including the Defense Intelligence Agency (DIA), the National Security Agency (NSA), and the National Imagery and Mapping Agency (NIMA). The DCI would share with the Secretary of Defense the selection of the senior officials in these agencies.

However, responsibility for day-to-day management of the DCI approved programs should remain with the Secretary of Defense. These DOD intelligence agencies are directly involved in combat support functions and employ military personnel. For example, the NSA is responsible for cryptography and communications security (COMSEC), as well as signals intelligence (SIGINT). The DCI would be responsible for the latter program, not the former. NIMA and DIA produce target attack packages for operational military units from intelligence information. The DCI needs to approve the resources devoted to the intelligence programs of these agencies, but need not and should not manage the day-to-day operations of the agency activities that are an integral part of military operations. The importance of these agencies to intelligence support of military operations requires that the Secretary of Defense must remain in charge of day-to-day activities.

Conversely, there remains little justification for the DCI to manage the National Reconnaissance Office (NRO). The plans, priorities, and budget for the NRO should be set by the DCI, but there no longer is an advantage for the DCI to manage the research, development, and acquisition of satellite systems. Indeed, there is a potential for much greater efficiency, if the DOD managed centrally military and intelligence satellite acquisition.

If these recommended changes were put into place, the DCI would have responsibility for planning and budgeting for less than 10% of the DOD budget. The Secretary of Defense must retain the authority to concur in the resource allocation plan that the DCI proposes. In order for the Secretary to be well informed in his decision to concur, he or she will need staff to be knowledgeable about the program plans proposed by the DCI. Review by DOD staff should be done by those familiar with intelligence support to military operations, and the review process should avoid duplication of the financial program-budgeting process that would be the principal responsibility of the IC staff.

If the DCI's authority is expanded, then the positions of DCI and Director of the CIA should be separated. Ideally, the FBI's national security division should be separated from the Department of Justice and established as a separate agency reporting to the DCI. If the DCI's authority is broadened to executive control of the DIS, as well as the CIA, and to resource allocation for DOD intelligence agencies, the DCI's focus will shift to harmonizing the efforts of the entire community, rather than management of the CIA or any component agency. In order to support the DCI in this role, the IC community staff will need to be enlarged and to acquire capability for program analysis and evaluation that spans all community activities.

Because the DCI would have greatly enlarged responsibility, it is reasonable to consider separating the dual hat appointment of a single individual as both head of the Intelligence Community and Director of the CIA. With separation, both the image and the reality will be that the DCI is primarily concerned with community issues and not CIA concerns. Given today's division of authority, it makes sense to have a single individual occupy both these positions, and the separation should not take place, until it is clear that the DCI indeed has broader effective planning and budgeting authority over the entire community, as well as executive control of the DIS and the CIA.

The realignment of responsibilities of the Director of Cental Intelligence is summarized in the following figure:

This proposed deployment of capabilities has the potential to improve greatly the collection and analysis intelligence on terrorism and other critical national security threats. Here are some expected improvements:

1. The realignment would strengthen human intelligence collection. Clandestine collection of intelligence from human sources is an important activity for foreign intelligence and a vital tool for penetrating terrorist organizations. The effectiveness of human intelligence is not fully captured by raw data, such as the number of case officers in the field or number of agents recruited. For example, we should emphasize disproportionately those sources that provide especially valuable information. Recruiting these exceptional sources requires ingenuity, professional zeal, and discipline, but also a profound understanding of the nature of the intelligence targets.

   The proposed realignment would enable greater integration of collection efforts. It would be easier to plan cooperative collection activities by the CIA from sources abroad and by the DIS from informants at home.

   The realignment would also improve the important synergy between signals intelligence and human collection in the planning and execution of collection activities. The combination not only reduces the risks in such operations, but also may provide important confirmation of information provided by a human source.

2. The realignment should improve intelligence analysis. Disciplined objective analysis is the fundamental underpinning of successful intelligence. Analysis is important for producing useful intelligence on the capability and intentions of adversaries for both policy makers and for collection planning. While these comments may seem self-evident, all too frequently proposals to strengthen intelligence focus on adding resources for collection and not analysis. Such an imbalance can lead to the flow of raw data overwhelming the capacity for thoughtful exploitation and analysis.

   Two aspects of intelligence analysis deserve special mention. First, combining foreign and domestic intelligence greatly increases the potential for incisive analysis. However, some information will be intertwined with information about perfectly innocent U.S. citizens - for example in mining of electronic data involving commercial transactions. An important advantage of the proposed realignment is that the DOJ will be primarily concerned with assuring that practices and policies of the IC do not inappropriately impinge on the rights and privacy of American citizens.

   Second, and independent of the realignment proposal, the IC should make greater use of outside experts, especially individuals with regional and economic expertise, to inform intelligence community judgments. Outside experts who have been recruited to spend several years on the National Intelligence Council have proven to be particularly useful, and this practice should be encouraged. This external expertise is especially valuable for understanding local social and cultural features that give rise to terrorists and shield their activities.

3. The realignment would give the DCI authority over resource allocation and make him accountable for IC performance. The DCI should have the responsibility for planning and harmonizing all community intelligence collection and analysis bearing on national security threats, whether foreign or domestic. Today there is no overall budget or multi-year program plan for intelligence activities. If the DCI has responsibility for constructing a budget for all intelligence activities that encompasses the NFIP, JMIP, parts of TIARA, and the part of the current FBI national security division expenditures that is currently included in NFIP, it will be possible to establish program performance milestones, costs, and schedules that will allow both the president and the congress to track the accomplishments of the IC. It is especially important to maintain a multi-year perspective, because intelligence success comes only from sustained coverage.

   The proposed realignment should also give the DCI specific authority to carry out four other important functions:

4. The realignment should strengthen covert action capability and oversight of these activities. Covert action capability is an essential tool for the U.S. in the fight against terrorism. Examples of contingencies when covert action is likely to be most important include destroying terrorist training bases or weapons of mass destruction development facilities. [These purposes are quite different from past efforts to de-stabilize or replace foreign governments.] There is little doubt that the American people support such covert action efforts. U.S. covert action capability diminished considerably in the early 1990s, but has strengthened in the last several years. It is important to continue to encourage this trend.

   There is no bright line boundary between covert action and para-military operations. In my view, the covert action capability that is most vital involves action in peacetime against specific targets or facilities, e.g. a suspect BW or CW production site, rather than actions that precede or accompany military operations. Activities that support contemplated military action are best planned and executed by Special Forces or other military units. For example according to news reports, CIA operations of the PREDATOR unmanned aerial vehicle (UAVs) armed with HELLFIRE missiles were highly successful in Afghanistan. The CIA apparently undertook this activity because the DOD seemed slow to do so. However, in general, the DOD should carry out this type of paramilitary activity. CIA covert action should be reserved for highly sensitive operations against targets that cannot be reached by conventional military forces.

   The proposed realignment should permit better planning and coordination of covert action operations that involve the CIA and the para-military capability of the DOD.

5. The realignment should also give the DCI authority to improve our capability to plan, coordinate, and control information operations. Signals intelligence, whether acquired from passive platforms or special collection operations, demands cooperation between the NSA, CIA, FBI, and other IC agencies, for example, the DOD's unified U.S. Space Command (SPACECOM).

   Advanced technology has blurred the difference between the capability required to collect signals intelligence, penetrate an information system, and interrupt or introduce deceptive information at the information source. These techniques will be of growing importance to both the U.S. (in both military operations and covert action) and to our adversaries, including major terrorist organizations.

   The Intelligence Community and the military services are pursuing information operations largely without coordination. An integrated plan and approach would significantly advance the development of effective information operations. Moreover, since information operations are, in effect, a precision weapon [As a former defense official said: "the electron is the ultimate precision guided weapon."], deployment and or use of offensive information operations amounts to covert action. The Secretary of Defense, in consultation with the Director of Central Intelligence, should establish a joint program office to coordinate and manage all offensive information operations.

   In addition, there are no common rules for obtaining approval for information operations, no common requirements for informing Congress of planned capability or on-going operations. In some cases, the information operation would be considered "covert action" and require both a presidential finding and reporting to congressional oversight committees. In other cases, establishing an information operation capability in peacetime might be considered part of normal preparation for dealing with a potential military contingency and require neither review by the National Security Council nor congressional notification.

6. The proposed realignment should also give the DCI authority to coordinate intelligence cooperation with other countries. U.S. intelligence efforts become more effective through cooperation with foreign intelligence services that have knowledge of their regions that we do not possess. Frequently, foreign intelligence services will have access that would be difficult, if not impossible, for the U.S. to gain. The assistance of foreign intelligence service is especially valuable in combating terrorism. Accordingly, it is important to use our bilateral cooperative agreements to advantage.

   Today, every agency in the intelligence community has its own bilateral relationship with its counterpart in every other country. In many countries, the U.S. may have half a dozen or more agreements in effect. Although, the IC staff maintains some oversight over all the different arrangements, there is no systematic effort to harmonize the liaison relationships of the different agencies. The DCI should have the authority and responsibility to set priorities and assure a coordinated inter-agency effort in our liaison relationships.

7. The realignment should also strengthen the IC's science and technology capability. Technical intelligence collection, especially communications intelligence, is a prime intelligence gathering technique for combating terrorism. As previously mentioned, human intelligence collection is made more effective when combined with communications intelligence. Other techniques of technical intelligence are key to obtaining intelligence on weapons of mass destruction. All the component IC agencies, including the Department of Energy, have an important role to play in developing and fielding new technical intelligence methods for both collection and analysis.

   Important technology trends affect intelligence gathering. First, the tremendous success of "upstream" collection technology, for example, from satellite and airborne sensors, means greater effort is needed on "downstream" processing, exploitation, and distribution of information.

   The CIA has been slow to move away from its historic and highly successful satellite technology activity to new areas. There are indications that the CIA's Directorate of Science & Technology is shifting priorities, and the current CIA leadership certainly appreciates the importance of doing so. But, my impression is that the pace of the shift is much too slow.

   Perhaps the most important new area involves technology that assists in the exploitation of information systems. The CIA creation of the quasi-public firm, In-Q-It, to identify relevant technology in the commercial sector, is a step in this direction. But, the CIA does not have the cadre of computer science and information technology experts needed to make significant progress.

   The detection of biological and chemical agents under covert development is another area that should command greater attention, but this technology is even further from the CIA's traditional area of expertise. The bio-medical competence that exists in other government agencies and in the academic and commercial world should be mobilized to address this problem.

   At present the DCI's IC staff does not plan a multi-year intelligence S&T development effort that takes into account technology needs on the one hand and expertise of agencies on the other. The reason is that the DCI does not have the authority to develop such a comprehensive plan funded by the intelligence budget - addressing, for example, technical methods to obtain better intelligence about weapons of mass destruction - or to require adherence to an approved plan.

8. The realignment should also authorize the DCI to require rotational assignments to develop professionalism. The greatest strength of the U.S. IC is the quality and dedication of its people. Every effort must be made to recruit, retain, and encourage the advancement of each person who chooses to dedicate his or her professional life to intelligence in the service of this country. Requiring rotational assignment in different agencies in the community, especially in offices that have community wide responsibility, is to broaden the professional experience of the individual and enhance his or her career. The experience of working in other agencies provides an appreciation of what is possible using other techniques and adopting different points of view. This greater experience in joint assignments, taken together, will improve the performance of the entire cohort of professional intelligence officers. A requirement for promotion to the senior ranks of the intelligence service should be one or more such rotational assignments. (The Goldwater-Nichols Act established a similar requirement for promotion to flag rank for military officers and rotation to "joint" assignments has become an accepted part of the experience required for advancement by senior military officers.)

   A similar proposal to improve development of the human resources of the IC and to encourage greater appreciation of the importance of the joint undertaking, as opposed to that of a single agency, was put forward by the Aspin-Brown Commission in 1996.

In summary, I have offered a realignment of existing intelligence authorities and responsibilities that centralizes authority and responsibility with the Director of Central Intelligence. There are other alternative arrangements: (1) the FBI could be given paramount responsibility for counter terrorism; (2) the proposed DIS could be placed within the DHS; and (3) the DIS could be established as an independent agency reporting to the president. Each of these alternatives has net disadvantages compared to the proposal I have presented to centralize intelligence. However, experience and history suggest that such a proposal will meet with considerable resistance. The question therefore arises whether some of the benefits of centralization can be achieved through greater cooperation and collegiality among existing entities. Although some progress has been made, I doubt that it will be possible to obtain the intelligence capability this country and its citizens deserve without a dramatic realignment that creates an executive authority that places national security first.

John Deutch is an Institute Professor at the Massachusetts Institute of Technology. He served as Director of Central Intelligence from May 1995-December 1996. From 1994-1995, he served as Deputy Secretary of Defense and served as Undersecretary of Defense for Acquisition and Technology from 1993-1994. Dr. Deutch has also served as Director of Energy Research (1977-1979), Acting Assistant Secretary for Energy Technology (1979), and Undersecretary (1979-80) in the United States Department of Energy.

In addition, Dr. Deutch has served on the President's Nuclear Safety Oversight Committee (1980-81); the President's Commission on Strategic Forces (1983); the White House Science Council (1985-89); the President's Intelligence Advisory Board (1990-93); the President's Commission on Aviation Safety and Security (1996); the President's Commission on Reducing and Protecting Government Secrecy (1996-1997); and as Chairman of the Commission to Assess the Organization of the Federal Government to Combat the Proliferation of Weapons of Mass Destruction (1998-99). He was a member of the President's Committee of Advisors on Science and Technology (1997-2001). Dr. Deutch serves as director for the following publicly held companies: CITIGROUP, Cummins, Raytheon, and Schlumberger Ltd. He is a director or trustee of the Council on Foreign Relations, Resources for the Future, and the Urban Institute. He is an overseer of the Museum of Fine Art in Boston.

Dr. Deutch has been a member of the MIT faculty since 1970, and has served as Chairman of the Department of Chemistry, Dean of Science and Provost. Dr. Deutch has published over 160 technical publications in physical chemistry, as well as numerous publications on technology, energy, international security, and public policy issues.

B.A. - History and Economics, Amherst College
B.S. - Chemical Engineering, MIT
Ph.D. - Physical Chemistry, MIT