The Red Pill

By David Ignatius

Tuesday, October 21, 2003; Page A25


If the CIA's predictions about Iraqi weapons of mass destruction prove to be wrong, how can Americans have confidence in CIA intelligence warnings in the future?

That question shadows the agency as its operatives sift through files and munitions depots in postwar Iraq, trying to square the surviving evidence about Iraqi chemical, biological and nuclear weapons with the prewar estimates.

President Bush has a political problem because of the missing WMD. Already, Democrats are focusing on the issue as evidence that the president misled the country in his arguments for going to war.

A sign of how sharp the debate will be came last week from Sen. Edward M. Kennedy: "All the administration's rationalizations as we prepared to go to war now stand revealed as double talk," he said. "The American people were told Saddam Hussein was building nuclear weapons. He was not. We were told he had stockpiles of other weapons of mass destruction. He did not. . . . We were told lie after lie after lie after lie."

The CIA's problem is as serious, at least in terms of the nation's security. For it goes to what might be called the "epistemology" of intelligence: How do we know what we think we know?

Before the war the agency was as sure as intelligence professionals can be that Saddam Hussein possessed weapons of mass destruction and was prepared to use them. That wasn't a political judgment or even an American one. It was shared by the intelligence services of Britain, France and other nations. And it dated back to the 1990s, long before George W. Bush came to Washington.

Yet the weapons of mass destruction haven't been found. Sen. Dianne Feinstein, a California Democrat who serves on the intelligence committee, probably spoke for many colleagues Sunday when she told CNN's Wolf Blitzer: "One of the things that's come through to me, based on reading and re-reading the National Intelligence Estimates and the daily intelligence, is that a lot of the key judgments that were made were not correct judgments."

These political criticisms upset CIA officials. But what seems to trouble them more is the prospect that the American public will lose confidence in the agency's work. In an era when the nation's margin of security against terrorist attack depends largely on intelligence, that loss of trust could have devastating consequences.

CIA officials defend the agency's performance in several ways. First, they contend that it's too early to conclude that they were wrong about Iraqi WMD. The process of inspection and analysis is just beginning. The agency's chief inspector, David Kay, has just filed a report on his first three months of investigation, which turned up none of the banned weapons. But Kay has another nine months of work, and will probably file two more reports.

Second, agency officials argue, Kay's initial findings have actually supported the case that Iraq was continuing to develop banned weapons. He found strong evidence that the regime was seeking to build or buy ballistic missiles with ranges of 600 miles or more, far beyond the U.N. limit of 90 miles. He also found new evidence that the Iraqis concealed WMD programs before the war, and systematically destroyed evidence after the war.

The epistemological problem -- of confirming what the CIA thought it knew -- is complicated by the war itself, which contaminated the crime scene, so to speak. "The Iraq that existed March 18 is gone," says one CIA official.

Despite the CIA's caveats and counterarguments, the agency is vexed by the failure to find a smoking gun. Officials have reverse-engineered every statement in the prewar intelligence estimates, reviewing the evidence they used to support the arguments that were made. They describe it as a cascade of reports from agents inside Iraq, intercepted communications, analysis of material recovered in U.N. inspections and other information collected from clandestine sources.

"With all the information we had, if we hadn't issued our warnings about Iraqi WMD, we should have been shot," says one CIA official.

The CIA's credibility is a national asset. Restoring it may be as important to the United States as anything that happens in Iraq.

To guarantee that credibility, the CIA should take some unconventional steps: It should disclose more of its prewar intelligence about Iraqi weapons of mass destruction so people can understand why it reached its judgments; and it should convene an outside board to analyze that evidence and assess whether the CIA's warnings were sound.

The worst thing that could happen for the CIA would be for it to become a political football in the 2004 campaign. The only way the agency can avoid that is to level with the public about how it knew what it thought it knew.

davidignatius@washpost.com