March 2004
Serious Intelligence Reform
The controversy over prewar intelligence on Iraq's weapons programs has raised-once again-larger questions concerning U.S. intelligence gathering and analysis and the nature of intelligence provided to policymakers. Many good ideas for intelligence reform have been proposed and debated over the past decades, but despite a succession of notorious intelligence failures, no reform has been seriously attempted. On March 5, 2004, leading experts discussed proposals for serious intelligence reform and considered whether the latest controversy might lead to long-term institutional improvement. Congresswoman Jane Harman, ranking Democrat on the House Permanent Select Committee on Intelligence, gave opening remarks.
The Honorable Jane Harman
U.S. House of Representatives
Transcending party lines is a vital step toward achieving meaningful intelligence reform, and there are many opportunities for bipartisan cooperation. No single administration or party can be faulted for intelligence failures; notable mishaps have occurred under presidents of both parties. The war on terror and the creation of the Department of Homeland Security are helping American lawmakers re-evaluate how intelligence gathering must occur in order to most effectively counter new global threats. Real reform is impossible, however, if senior officials remain in a state of denial. The Bush administration has failed to acknowledge the need to correct obvious flaws in the intelligence systems; the White House has instead indicated that it will wait until after the election to hear recommendations from a new WMD commission. But we must act now to make our country safer.
Five Steps for Better Intelligence: First, global WMD estimates should be reevaluated entirely, independent of previous analysis. This is particularly true of Iran, Libya, North Korea, Syria, and Pakistan. Libya's newfound cooperation provides an important opportunity to compare intelligence estimates with actual weapons programs. Second, intelligence gathering and verification procedures must be improved. Dramatic change is needed in the way we find and assess potential spies; there should be an emphasis on recruiting people who understand local cultures and speak local languages. An effective alternative to the pre-existing Cold War intelligence apparatus has yet to be developed.
Third, the method through which intelligence is analyzed and disseminated to policymakers must be reassessed. Former Director of Central Intelligence Bob Gates wrote in 1992 that analysts should focus more on what they do not know, refrain from exaggerating certainty, and seek to challenge the fundamental assumptions of a culture that fosters "group-think." For example, the analysts who wrote the National Intelligence Estimate (NIE) in October 2002 did not work to challenge assumptions at all, but rather believed that the decision to go to war had already been made and that their duty was to advise military commanders on whether troops would face biological and chemical weapons as they invaded Iraq. Better forums to raise doubts must be provided, regardless of inconvenience to policymakers. But policymakers must be able to rely on the intelligence community for "the hard truth;" there must be a distinction between those who provide intelligence and those who promote policy.
Fourth, the president should direct a review of the activities of the various offices in the Department of Defense (DoD), particularly those that report to undersecretary of defense Doug Feith and the Office of Special Plans.
Finally, international inspections regimes must be supported and strengthened, and the various agencies and groups conducting inspections should cooperate with each other and share information.
Long-term Changes: The importance of strong leadership in the intelligence community and the creation of a new Director of National Intelligence (DNI) separate from the Director of Central Intelligence would help facilitate interagency cooperation. The DNI would not necessarily preside over a federal department but could instead ensure that information would flow more freely between intelligence agencies.
There is a need for competition and risk-taking, but both can exist in an integrated system. This does not mean putting fifteen agencies under one roof with a clear, formal structure, but instead creating a way for each piece of information to be integrated in order to allow analysts with good judgment to connect the dots. September 11 showed the disastrous consequences of a failure to share intelligence. To prevent future intelligence failures, a way to integrate information without integrating institutions must be found.
R. James Woolsey
Booz Allen Hamilton
An assessment of areas of vulnerability in our intelligence and security infrastructures must be made. This may be the key to post-Cold War intelligence, where specific tips from informers are less important than understanding networks and searching for the weaknesses. Using conventional intelligence techniques in the war on terror will not provide us with the kind of assessments that are needed to confront the new threat. It is essential to review WMD estimates worldwide; it is important to consider why intelligence estimates have been off the mark (in both directions).
Combined operations, where spies tip off satellites and satellites tip off spies, will remain important, but even the best intelligence can sometimes miss the mark. It appears that Iraqi generals, and even Saddam Hussein himself, may have been misled about Iraq's weapons capabilities. Captured Iraqi generals were usually convinced that, although his own unit had no chemical weapons, the units on his right and left certainly did.
There is nothing inherently wrong with having institutions in place to advise the secretary of defense on intelligence matters; there are similar institutions at the State Department. The most important part of being an intelligence officer is getting information, even if that is not always the easiest thing to do. Intelligence officers should be willing to talk to anyone who wants to talk to them- liars, cheats, defectors, and princes. Risk-taking is crucial to collecting good intelligence.
Reuel Marc Gerecht
AEI
Iraq has become "too charged" an issue to explore much further; it might be more beneficial to look at Iran's WMD programs instead. The CIA and the clandestine service's methods for dealing with Iran's clerical regimes since 1979 should be examined; serious intelligence lapses will be evident, and the agency should symbolically and retrospectively fire agents who failed to do their jobs. No reform is possible if agents are not held accountable for their intelligence.
The suggestion of an integrated intelligence community should be carefully considered. This could create inter-office disagreement and a competitive ethic, or it could lead to a neutralized, complacent intelligence community with an ethic that is neither curious nor provocative. Human intelligence will continue to be necessary, but it cannot be our only means of intelligence gathering.
Christopher DeMuth
AEI
Increasing competition between agencies is essential to meaningful reform, but centralization is rarely the best remedy for government problems and should not be attempted here. The CIA, State Department, and DoD are bound to disagree on intelligence estimates because they have different primary missions and employ different collection methods. In almost every administration, the degree to which the president and senior staff rely on their own contacts for information is remarkable. This shows a healthy interest in obtaining information from varied sources.
AEI executive assistant Courtney Richard prepared this summary.
