A flawed assessment

NO ONE who reads the Senate Intelligence Committee's report on the CIA's assessment of prewar Iraq can fail to learn that there are critical problems in U.S. intelligence that go far beyond the Iraq war.

Like similar reporting by the House Intelligence Committee, it outlines critical failures in the way the intelligence community, the CIA and director of central intelligence do business.


At the same time, the Senate panel's report has serious failures. The most glaring is its inability to detect and describe the level of indirect political and policy-level pressure on the intelligence community to reach the "right" conclusions.

The report may be correct in concluding that senior officials never interfered directly or acted to politicize intelligence. But it fails to address the climate of policy-level expectations that indirectly demand one kind of answer, the impact of repeated searches for revised analysis, the questions at the staff level and higher searching for the "right" answers and the knowledge that policy-level recognition affects intelligence careers.


This is pressure of a different, but very real, kind. Intelligence officers and managers are only human, and the intelligence community almost always responds to strong user demands and perceptions when an administration decides that one view is correct.

The report also creates an artificial decoupling of intelligence from the policy-maker and user. It does not examine what was done with intelligence products or informal intelligence inputs and support.

The failure to fully examine the writings and role of lower-level political appointees in Vice President Dick Cheney's office and the Pentagon's Office of the Undersecretary for Policy, headed by Douglas Feith, means that the report blames the intelligence community without examining the full range of classified and intelligence-related data that policy-level staffs sent to policy-makers. The issue of what the staffs did with intelligence and classified information is just as important as any failings in the intelligence community.

The narrow focus on prewar assessments of weapons of mass destruction, and links between Iraq and terrorism, is even more serious. The report does not address the almost catastrophic failure to accurately assess the problems of security and nation-building in Iraq -- a failure duplicated in Afghanistan. It is difficult to believe that this omission did not have a political dimension, since the evidence of inept and destructive political interference would have been far clearer.

In fairness, the report also fails to note how many of the problems in the intelligence community's underlying assessment of Iraq developed or increased during the Clinton administration. It effectively dodges the responsibility of administrations from both political parties.

Perhaps for similar reasons, the report leaves the current and future situation of U.S. intelligence in limbo. It makes no real analysis of the corrective steps, if any, that have been taken so far. It does not analyze the community's plans to fix things. It also fails to adequately link its analysis of the problems in dealing with Iraq to a broader analysis of the general performance of the intelligence community. Presumably, the same failings affect all of our intelligence work globally in every area, but the report of some 500 pages largely brushes over this issue.

There is no sense of history or continuity in the committee's analysis. As someone who dealt with similar intelligence failures, I am struck that nearly all of the major problems and criticisms could have been written after Vietnam or in describing the CIA's assessments of Soviet-led Warsaw Pact conventional forces in the late 1980s and early 1990s.

Certainly the problems with over-compartmentalization and over-classification, failure to explicitly analyze uncertainty, failure to examine alternative approaches and conclusions, repetition and layering of past studies and assessments, and failure to share with other agencies are all problems that have been raised repeatedly since the 1970s. A failure in preparing for a single war is one thing; 20 to 30 years of similar failures is another.


The report's criticism of the human intelligence (HUMINT) -- spies -- in Iraq is devastating. But no effort is made to look at the adequacy of HUMINT in fighting the broader war on terrorism or elsewhere. There also seems to be little understanding of the severe limits on what HUMINT can do and the illusion that HUMINT can always succeed where spy satellites and electronic systems fail. The report also would be more convincing if it explained whether the current HUMINT effort has the money and people it needs; it borders on the ridiculous to simply ask that CIA agents be more daring.

Generally, these problems with the Senate Intelligence Committee's actions raise serious questions about what Congress does next. Most of the proposals that members have made are likely to do more damage than good.

Follow-up efforts to cure the problems of U.S. intelligence by changing the role of the director of central intelligence, or simply giving him more control of the intelligence budget, are counterproductive. It is the entire process of collection and analysis that needs fixing, not the organization chart, the top of the chain of command or budgeting methods.

Moreover, creating a stronger "intelligence czar" is the last way to spur debate and attract alternative views, independent efforts at analysis and honest discussions of uncertainty.

Anthony H. Cordesman holds the Arleigh A. Burke chair in strategy at the Center for Strategic and International Studies in Washington.