We can thank the well-publicized recent intelligence failures for one thing: the brand-new genre of the beautifully written intelligence critique. First came the report of the Sept. 11 commission, with its riveting narrative of the events surrounding the attacks. Now comes the report of the Commission on the Intelligence Capabilities of the United States Regarding Weapons of Mass Destruction, co-chaired by retired judge Laurence Silberman and former senator Charles Robb. The heart of this report is another brilliant narrative, that of the mistakes -- notably deception by the well-code-named spy "Curveball" -- that convinced the intelligence community that Saddam Hussein possessed weapons of mass destruction.
The Sept. 11 commission, perhaps exhausted by its narrative efforts, failed to defend its radical recommendations for reforming intelligence -- which nevertheless captivated Congress. The WMD commission devotes 300 pages to defending its lengthy menu of reform proposals, many sensible -- especially those the intelligence community had adopted in advance of the report.
President Bush, surrounded by Charles Robb, left, and Laurence Silberman.
(Ron Edmonds -- AP)
For all of its genuine distinction, the report has weaknesses. Foremost among them -- a product of the blinding clarity of hindsight -- is a misplaced perfectionism that feeds the dangerous fallacy that all intelligence failures are the product of culpable, and therefore remediable, blunders. Actually, most such failures are the inevitable result of the inherent limitations of intelligence. Before the invasion of Iraq, nearly every competent observer, including the intelligence services of foreign nations opposed to the invasion, believed that Saddam Hussein had a stockpile of chemical and biological weapons and was trying to build nuclear bombs as well. Hussein's history, and above all the logic of the situation -- surely he wouldn't risk his regime by failing to come clean if did not have such weapons -- created a presumption that he had them. The commission criticizes the intelligence agencies for embracing the presumption. But no inquiry operates without preconceptions that shift the burden of proof to the doubters -- of whom there were, in the case of the Iraqi weapons, precious few.
The commission also neglected the elementary statistical principle that weak data, all pointing in the same direction, can add up to strong evidence. If two independent observations each have a 40 percent probability of being false, the probability that both are false is only 16 percent. (If there are three, it falls to 6 percent.) Lots of apparently independent data all pointed to the alarming conclusion that Hussein had weapons of mass destruction. The costs of disregarding this evidence might have been horrendous, Sept. 11 having underscored the danger of insisting that certitude precede action.
The report neglects sound organizational principles. It suggests that the office of the director of national intelligence (the principal legacy of the Sept. 11 commission) be organized along functional rather than substantive lines, with a budget division, an information technology division, a human resources division and so forth. But then all the 15 agencies that make up the U.S. intelligence system would be fighting each other in each division, and coordination would occur only at the very top. Actually the agencies sort nicely into four groups -- military intelligence agencies, technical agencies (mainly utilizing spy satellites), foreign intelligence agencies and domestic intelligence agencies. Each grouping could be organized as a semi-autonomous service, with its own budget, information technology, legal staff, etc., and then the director of national intelligence would have to coordinate only four divisions, not 15.
That is roughly the structure of Britain's well-regarded intelligence community -- which points up another deficiency of the WMD commission's report: indifference to theoretical, historical (intelligence failures go back to the Trojan War) and comparative perspectives. Every nation that we take seriously -- even Canada -- has a domestic intelligence agency separate from its national police force, the best known being Britain's MI5. Only the United States buries its principal domestic intelligence service in a police force (the FBI). Police hunt criminals, and criminal law enforcement will not defeat terrorism. An agency 100 percent dedicated to domestic intelligence is more likely to do a good job than the FBI, which is 10 percent intelligence and 90 percent criminal investigation.
The commission's report blasts the FBI's ineptitude in domestic intelligence but flinches from concluding that we should learn from nations with a longer experience of fighting terrorism than our own and create a U.S. counterpart to MI5. The FBI's attitude, reflecting the domination of the bureau by its 56 scattered field offices, is epitomized by a remark by the head of one of those offices: "[Osama] bin Laden is never going to Des Moines." (So if bin Laden is smart, he'll attack Des Moines; it's unprotected.) The commission tells us the FBI hopes to get its act together -- by 2010 -- and meanwhile the bureau has managed to move 96 percent of its intelligence budget into divisions not subject to the budgetary authority of the director of national intelligence. I hope bin Laden isn't reading the commission's report.
The writer is a judge on the U.S. Court of Appeals for the 7th Circuit and a senior lecturer in law at the University of Chicago. He will be available to answer questions at 3 p.m. today on www.washingtonpost.com.