Iraq, and Analysis, Revisited

By Dafna Linzer
Washington Post Staff Writer
Saturday, February 3, 2007

The intelligence community's bleak set of judgments about Iraq, made public yesterday, differs significantly from the flawed assessments of that country's alleged weapons programs that the White House released more than four years ago.

Twenty-seven pages of confident assertions about a nuclear weapons program, mustard gas, nerve agents and elusive government ties to al-Qaeda have been replaced, more than four years later, with heavily qualified glimmers of hope across a mere 3 1/2 pages laying out the prospects for reversing Iraq's dive into chaos.

The intelligence community's glossy October 2002 estimate, which claimed that Saddam Hussein had weapons of mass destruction, was touted by the White House as a main reason for war with Iraq. After no such weapons were found, the intelligence community -- particularly the CIA -- significantly altered the way in which it would conduct future analyses, highlight uncertainty and acknowledge dissent.

Current and former intelligence officers said a better system for producing the National Intelligence Estimate was put in place immediately. The result, they said, is that the new estimate -- which describes an Iraq engulfed by numerous bloody wars -- is strikingly similar to a still-classified assessment written in late 2004 and widely distributed among President Bush's top advisers.

That assessment predicted that over a period of 12 to 18 months, there would be "at best, tenuous stability, at worst a slide into civil war" in Iraq, said Paul R. Pillar, a Middle East scholar and veteran intelligence officer who was one of the report's principal authors.

"If you could compare the 2004 and 2007 estimates, it would be clear they are two installments of the same story, done the same way. The only difference is that the policy debate has become more intense in the intervening two years because the situation has become worse and worse," said Pillar, who left the intelligence community in 2005 and now teaches at Georgetown University.

Pillar could not discuss details of the 2004 assessment, but a senior intelligence official agreed to discuss some of its findings on the condition of anonymity. The official said that like yesterday's estimate, the 2004 review offered three difficult scenarios for Iraq, all of which have been borne out to some degree.

"We imagined some slow and uneven progress, which has happened in parts of the country," the official said. The estimate also considered a breakup of Iraq along sectarian lines. "The preconditions for that have sort of happened, but only really up in the north with the Kurds; and the third one was chaos and instability and we've certainly seen some of that."

Based on the accuracy of the previous political estimates, Pillar said, it is very possible that the worst-case scenarios set out in yesterday's assessment could occur in the next year and a half.

In crafting key judgments of the 2007 estimate, analysts from across the intelligence community laid out three possible scenarios, each of which could have enormously negative consequences for Iraq: Chaos leading to partition, the emergence of a Shiite strongman who would seize power and rule by dictatorship, or a complete collapse of central control resulting in violent fiefdoms across the country.

The official said the new estimate "highlighted a trend of deterioration that is not irreversible. But if it's not reversed soon, then we're heading toward a worsening security situation over the next 12 to 18 months. It's more of a main judgment than we were able to give in 2004, if significantly caveated."

Robert Hutchings, who until 2004 was the chairman of the National Intelligence Council, which produced the earlier assessments, said, "The basic tack on the political, security and economic situation in Iraq has been pretty consistent, before, during and after the war."

Hutchings, who is now a diplomat-in-residence at Princeton University, said a January 2003 estimate, which was done two months before the U.S. invasion of Iraq and which also remains classified, was the first to include the word "insurgency." The 2004 estimate "was the first to use the term 'civil war,' " he said.

Pillar said, "What we saw and see today is a very unfavorable state of affairs and unfavorable trend lines."

The senior intelligence official who agreed to discuss some aspects of both reports said that while the 2004 estimate now seems prescient, analysts failed to predict "the rapid degree of intensification of sectarian mobilization and consciousness, and the speed in which that happened. I don't think we were that insightful."

Staff researcher Julie Tate contributed to this report.


© 2007 The Washington Post Company