Jun 27 2002
Chairman Lieberman and Senator Thompson, I thank you for the opportunity to address this Committee today about various proposals to reorganize our government in order to create a Department of Homeland Security (DHS) to take the lead in protecting Americans from terrorist attacks within the United States. As I have pointed out many times, more Americans were killed by terrorists on September 11th, 2001 than died in Japan's infamous sneak attack upon Pearl Harbor on December 7th, 1941. It is both necessary and fitting that we do everything in our power to ensure that the United States never again suffers such a catastrophe - a third "Pearl Harbor." For this reason, I support the creation of a Department of Homeland Security.
As in so many important endeavors involving legislation; the devil is always in the details. We also know all too well that legislation alone can not meet all the challenges we face. One of the biggest risks we face in the world of intelligence collection is risk aversion. Our intelligence bureaucracies have over time become adverse to risk taking, partly because of internal institutional pressures and partly because of external criticisms. No bill, rule or regulation can reverse that. What we can do is address an immediate need. To do so, we need to create a new Department, but it is important that we create it right - and that in creating it, we do not simply replicate the mistakes of the past. Accordingly, I am grateful for the opportunity to discuss the intelligence aspects of homeland security, a topic with which I have been greatly concerned and closely involved for several years on the Senate Select Committee on Intelligence (SSCI), both as its Chairman and currently as its Vice Chairman.
I. The Centrality of the Intelligence Function to Homeland Security
In introducing his legislative proposal for a Department of Homeland Security, President Bush declared that the "top priority" of the Department will be "preventing future attacks." This emphasis is picked up in the text of his legislative proposal itself, which stresses in § 101(b) that the "primary mission" of the Department of Homeland Security will be to "prevent terrorist attacks within the United States." As the President's proposal recognizes, this fundamental mission highlights the importance of intelligence. First among the list of the new Department's "primary responsibilities," the President's proposal lists the crucial function of conducting "information analysis" related to terrorist threats. The intelligence function is absolutely central to the President's proposal, as it should be. It is therefore doubly important that we get the intelligence aspects of the Department right.
The President, in his proposal, assigns appropriate emphasis upon ensuring that this intelligence function is carried out properly, by making the Information Analysis and Infrastructure Protection office the first of the new Department's key components. If done right, the creation of such a national-level center for true all-source intelligence "fusion" of terrorist related threat information would be of huge value.
Most Americans would probably be surprised to know that even nine months after the terrorist attacks of September 11th, there is no federal official - not a single one - to whom the President can turn to ask the simple question: "What do we know about current terrorist threats against our homeland?" No one person or entity has meaningful access to all such information the government possesses, no one really knows "what we know," and no one is even in a position to go find out. This state of affairs is deplorable, and must end.
II. A Limited, Focused Agenda for Reform
In the wake of a well-publicized series of significant intelligence failures - including the failure to prevent the bombing of the World Trade Center in 1993, the failure to prevent the bombing of Khobar Towers in 1996, the failure to anticipate the Indian nuclear tests in 1998, the failure to prevent the bombing of our embassies in Africa that same year, the accidental bombing of the Chinese embassy in 1999, the failure to prevent the attack on the USS Cole in 2000, and, of course, the failure to prevent the attacks of September 11th - there has been no shortage of proposals to reform the U.S. Intelligence Community. Most of them have involved variations on the theme of empowering the Director of Central Intelligence (DCI) to exercise more real authority within the mostly Defense Department-owned Intelligence Community. Other proposals, such as one floated this week, would empower the Pentagon by creating an Undersecretary of Defense for Intelligence.
All of them, so far, have gone nowhere. When such ideas do not founder upon the rocks of inter-departmental rivalry and what the military calls "rice bowl" politics, they simply fail to elicit much interest from an Intelligence Community that, even to this day, insists that nothing is fundamentally wrong. Too often, serious reform proposals have been dismissed as "a bridge too far" by Administration after Administration and Congress after Congress, and have simply fallen by the wayside. While very modest attempts at reform have been enacted, they have been ignored by succeeding Administrations and openly defied by our current Director of Central Intelligence.
With this in mind, last year I asked our committee's Technical Advisory Group (TAG) to undertake its own look at these issues. The TAG is a group of prominent scientists and technologists that volunteer their services to advise our committee on very difficult technical and program management issues. We worked with them over several months on these matters, and we came to some interesting conclusions. Rather than rest our hopes for reform upon plans destined to run headlong into vested interests wedded to the current inter-departmental division of intelligence resources or to be smothered by pained indifference from holdover bureaucrats satisfied by the status quo, the TAG proposed instead that the President create something entirely new - a small, agile, elite organization with the President's personal support, dedicated wholly and single-mindedly to conducting fusion analysis. This organization would draw upon all the information available to the federal government and use the resulting knowledge to achieve a single clear goal - dismantling and destroying terrorist groups that threaten the United States.
This, we hoped, might allow meaningful reform to take place without initially having to upset entrenched bureaucratic applecarts. We proposed, in effect, an intelligence-related version of the Manhattan Project that would take place to some extent outside the traditional chains of command and networks of vested interests. We suggested an approach modeled on the movie catchphrase "If you build it, they will come." If this new venture were successful, its progress would breed further successes by gradually attracting resources and support from elsewhere - and perhaps by stimulating the intelligence bureaucracies to do more to reform themselves when faced with the success of an alternative model. The private sector refers to this process as "creative destruction."
After the terrorist attacks of September 11th, we felt that it was time to present this proposal to the White House. If the mass murder of approximately 3,000 Americans could not drive meaningful reform in our Intelligence Community, we reasoned, what could? Accordingly, Chairman Bob Graham and I brought our TAG to meet with Governor Tom Ridge in the White House on November 29th of last year. We met with him for about 90 minutes, and talked in detail about our plan for the creation - for the first time - of a truly all-source, national-level intelligence analytical agency dedicated to knowing and assessing everything that our government knows about terrorist threats.
I think I can speak for Senator Graham as well as for myself and the distinguished members of our Technical Advisory Group in saying that I am very pleased to see that President Bush has seen fit to propose the creation of just such an organization within the new Department of Homeland Security. Unlike the Lieberman bill (S.2452) - which neglects the intelligence function and nowhere provides the new Department with a centralized threat-assessment entity capable of making up for the Intelligence Community's longstanding failure to provide government-wide "one stop shopping" for terrorist threat information and analysis, - the President's proposal puts terrorism-related intelligence front and center, making it the foundation for all other protective measures. I applaud the President's wisdom in making information analysis such a central focus of his plan.
III. The President's Plan for a Homeland Security Department
It is in that vein that I would like to offer a few constructive criticisms of the President's proposal. Precisely because the intelligence function is vital to every aspect of inter-agency coordination and planning for homeland security, we must ensure that these aspects of the President's plan are structured properly - and that they do not simply replicate past mistakes.
A. Access to Information
In this regard, I would like to point out that under § 203 of the President's bill, the Secretary of Homeland Security would have only limited access to information collected by the Intelligence Community (IC) and Law Enforcement Agencies (LEAs). That section provides that the Secretary would be entitled only to "all finished reports, assessments, and analytical information" related to threats of terrorism in the United States. Unlike information relating to "infrastructure or other vulnerabilities" to terrorist attack, - all of which the Secretary would be given access "whether or not such information has been analyzed" - information on terrorist threats themselves would be available to the Department only in the form of what is known as "finished" intelligence. Under § 203, the Secretary may obtain the underlying "raw" information only "by request" or when the President specifically provides for its transmission to the new Department.
To my eyes, these limitations are unacceptable, and seem designed to keep the new office of Information Analysis and Infrastructure Protection dependent upon the goodwill of the Intelligence Community and Law Enforcement Agencies and hostage to their perhaps incompletely-informed or self-interested judgements about what the Homeland Security analysts really "need to know." Already, we understand that Director of Central Intelligence George Tenet has no intention of providing "raw" intelligence data to Homeland Security intelligence analysts. As he sees it, they should be content to receive only "finished" reports - that is, to get no deeper access to Intelligence Community databases than we do in Congress as we receive the Community's periodic intelligence products.
To agree to such limitations would be, in my view, a grave mistake. In the information-technology world, we are on the verge of dramatic new breakthroughs in data-mining capabilities that are giving ordinary analysts an extraordinary ability not just to search but to analyze and understand enormous quantities of data from a vast array of different data sources. The cutting edge of intelligence analysis, in other words, is likely to be in "crunching" massive amounts of data on a genuinely all-source basis, drawing upon multiple data-streams in ways never before possible.
However, as long as we have no one in a position to see all the many data-streams that exist within the federal government - much less those that may also exist in the state and local arena, and in the thriving information economy of the private sector - all of these rapidly-advancing analytical tools will be of little use. Already, it has been one of our frustrations at the SSCI to see the degree to which even agencies that acknowledge the importance of inter-agency electronic information-sharing are each independently pursuing separate "answers" to this problem. Even their responses to the problem of agency-specific "stovepipes" are too often themselves "stovepiped" responses. The DCI's own initiative to create an Intelligence Community-wide "Intelligence Community System for Information Sharing" (ICSIS) depends wholly upon agencies deciding what information they think other agencies' analysts need to know. Every agency will be charged with populating its own "shared space" that will be searchable by cleared and accredited on-line users. No outsider, it seems, would ever have access to an agency's real databases.
Without some modification to the President's Homeland Security proposal, and to the DCI's refusal to consider providing "raw" information to the new Department, this initiative runs the risk of replicating and institutionalizing these limitations. The exciting part about the new Department is precisely that it offers the prospect of getting beyond - or above - bureaucratic stovepipes in the ways we imagined for the anti-terrorist "Manhattan Project" we discussed with Governor Ridge last November. Rather than having every agency decide on its own what every other agency needs to know about its own information holdings, we need instead to create an institution that finally has real visibility into all government information on terrorist threats. The President's proposal for a Homeland Security information analysis office has the potential to be that organization and to rise above bureaucratic "business as usual." But its access cannot be limited just to what the agency chieftains decide it should have.
In my view, the President's proposal can and should be improved by giving the Secretary of Homeland Security access to essentially all information related to terrorist threats - including "raw" data - that is in the possession of any government agency. Homeland Security intelligence analysts should be free to data-mine agency holdings in order to undertake true all-source intelligence fusion.
Senator Specter has offered an amendment that would help fill this hole in the President's otherwise very promising proposal by creating a National Terrorism Assessment Center (NTAC) with the authority to "direct" the CIA, FBI, and other federal agencies to provide it with "all intelligence and information relating to threats of terrorism." As I see it, Senator Specter is clearly thinking the right thoughts, although I believe it would be a mistake to duplicate analytical functions by creating a new Center within or parallel to the Homeland Security information analysis office. Personally, I think the soundest step would be to apply the concept of unfettered information access to the Department of Homeland Security itself. Section 203 of the President's proposal should be modified to allow for the creation of an information architecture that will enable Department analysts to seek and obtain whatever information they deem necessary to understand and thwart terrorist threats against the United States.
The only qualifier on this authority would be to provide that such transmittals must occur pursuant to some kind of agreement or memorandum of understanding (MOU) with the DCI regarding security procedures for handling classified information and with the Attorney General with respect to handling "U.S. person" information and protected law enforcement information pursuant to applicable law. Provided that the new Department's intelligence functions were also subjected to appropriate intelligence oversight by Congress, the United States would then be well on the way to creating, for the first time, a genuinely "all-source" national analysis organization devoted to combating the threat of terrorism in the United States.
B. Other Issues
Naturally, the Department of Homeland Security - including its intelligence functions - will require close Congressional scrutiny and oversight as it is created. Whatever the final information-access rules end up providing, it will be necessary to ensure that appropriate agreements are worked out between the agencies involved, and that personnel are properly trained and equipped to implement them. In hidebound bureaucracies such as our Intelligence Community, this can be no small task. As you may recall, we put mandatory sharing provisions in Title 9 of the USA PATRIOT Act, but today - eight months after the President signed that Act into law - procedures for implementing such sharing are still being negotiated between the Attorney General and the DCI. The detailed procedures for information-sharing with the new Department of Homeland Security will likely require close Congressional attention.
Another of my concerns relates to the importance of ensuring that the new Department's Information Analysis and Infrastructure Protection office maintains an appropriate balance within its own ranks. Under the President's proposal, that office will acquire infrastructure-protection organizations from a number of existing federal agencies, these entities being transferred en masse to the new Department. The "information analysis" side of the office, however, will apparently have to be built up largely from scratch. It will not acquire specific analytical offices from other agencies within the federal system, but will rather have to be "grown" within the Department. If done right, this could be a great strength, allowing the Department of Homeland Security to build its own elite analytical cadre largely independent of the institutional biases and bureaucratic mindsets of the existing Intelligence Community. Careful attention over time - not to mention close Congressional oversight - will be needed in order to ensure that the new Department's information analysis functions become large and robust enough to prove useful. This process may involve "growing pains," and the fledgling organization may also need to be nurtured and protected against its bureaucratic rivals and others who may not wish it to succeed.
For the most part, I have no other serious concerns about the President's proposal to create a new Department of Homeland Security. I would only note that under § 710 of the President's bill, the Secretary would have the power to terminate any Inspector General investigation that he felt to be inappropriate, provided only that he provides notice of this termination to the Speaker of the House and to the President of the Senate. Given the important role that Inspectors General play in our system of legal and policy oversight, I would think this provision to be too limiting. Even if the Secretary could derail investigations, I would think it imperative that notice of such a decision be given to the Congressional committees of jurisdiction - which would presumably vary depending upon the subject matter.
IV. Do not Forget to Reform the Rest of the Intelligence Community
I would like to emphasize, that while I believe the President's proposal for a terrorism-focused information analysis function within the Department of Homeland Security is a vital step forward, its creation alone will not solve the intelligence problems affecting our country - and which we and our House counterparts are currently studying as part of our inquiry into the intelligence failures that led up to September 11th. We must not forget that we will still have a large intelligence bureaucracy that will not be part of the new department, and that the department's important analytical functions will have no chance of succeeding if the information-collection system that feeds it remains broken. Furthermore, the new Department's system will focus upon domestic terrorist threats, leaving the whole universe of foreign intelligence unreformed.
The President has noted that his proposals for the Department will "complement the reforms on intelligence-gathering and information-sharing already underway at the FBI and the CIA." While the FBI is doing a commendable job trying to reform itself, the CIA has yet to even consider significant changes. Indeed, as its leadership has repeatedly indicated in testimony before the Senate Select Committee on Intelligence, the CIA's response to September 11th has mostly been to insist that it is on the right track and that Congress should simply give it more money and personnel with which to continue doing more of the same thing. As I have said elsewhere, I think that response is inadequate, and that we can do much better.
Finally, I would like to make a brief comment about the analysis of information that already exists in the private sector. This is another area that our Technical Advisory Group has emphasized in our internal discussions of intelligence reform at the SSCI. The private sector collects and maintains vast amounts of information that would be of enormous use to intelligence analysts seeking to track terrorists. To the extent that the government can obtain access to such information on the same commercial terms as any member of the public - e.g., through fee-based access to various databases or services - such data presumably constitute "publicly available" information which our intelligence analysts can already freely collect and analyze under current intelligence collection guidelines.
To the extent that current policy decisions within the Intelligence Community restrict analysts' ability to use such information for "U.S. person" or other reasons, the new Department of Homeland Security should not tie its own hands in this way. September 11th illustrated that the distinction between "foreign" and "domestic" intelligence is not as bright a line as we once thought. A government agency devoted to the all-source analysis of domestic terrorist threats cannot afford to forswear such private-sector information. I would expect, therefore, that the intelligence oversight rules for the domestic-focused Department of Homeland Security intelligence functions would have to be different in some key respects than those that today govern intelligence work by agencies focused upon threats abroad.
In summation, I strongly support President Bush's plans for the creation of a Department of Homeland Security, and I urge my colleagues to lend it their support. The terrible events of September 11th have graphically illustrated the need for such an organization, and I am particularly pleased that the President has incorporated in his proposal many of the ideas we at the SSCI discussed with Governor Ridge last November. The terrorism-focused all-source intelligence functions of the new Department's Information Analysis and Infrastructure Protection office will be a major step forward, and ought greatly to increase the federal government's ability to combat terrorist threats in the United States.
I urge my colleagues, however, not to travel this road only half way.
We should modify the President's proposal in order to ensure that the new Department has full and unfettered access to terrorist threat information in the possession of other federal agencies - whether or not those agencies wish to share it. The Department of Homeland Security cannot provide real homeland security if it remains at the mercy of agency bureaucrats who reserve themselves the right to decide what information the new Department's intelligence analysts need to know in order to protect the United States.
We should support the development of the Department of Homeland Security's information analysis office into a robust and effective organization capable of holding its own against the bureaucratic rivals it is sure to acquire in the Intelligence Community.
Finally, we should not forget that intelligence analysis relating to homeland security threats is only part of the picture. Our homeland will not be secure if the intelligence collection machinery that provides information to the new Department is dysfunctional. And our national security interests will suffer if our foreign intelligence and counterintelligence organizations remain unreformed. The FBI is taking steps to bring about meaningful internal change, but we should not let other organizations within the Intelligence and Law Enforcement communities escape without similar scrutiny.
I applaud the President for presenting his proposal to Congress, and I thank you, Mr. Chairman, for holding a hearing on this important topic. Together, I hope and expect that we will be able to create a new and highly effective Department of Homeland Security that will better secure the safety and livelihood of all Americans in the years to come.