Secrets, lies and yellow cake

Whatever is now discovered in Iraq, it is unlikely to include 30,000 warheads and four chemical muni

The evidence for a sweeping failure of American intelligence in the months leading up to war with Iraq can be found laid out most clearly in the speech delivered by the US secretary of state, Colin Powell, to the UN Security Council on 5 February. There, citing "facts" based on "sources, solid sources" obtained by American and British intelligence organisations, Powell made 34 claims about Saddam Hussein's possession of weapons of mass destruction, ongoing Iraqi efforts to build more, and determined efforts by the Iraqi government to hide what it was doing.

Powell supported his claims with an unprecedented array of declassified intelligence information: ten photographs from spy satellites, "intercepted communications" including three recordings of telephone conversations between officers in Saddam's elite Republican Guard, the interrogation of defectors and "detainees", and numerous other unspecified items of secret information. "These are not assertions," Powell said, with the CIA director, George Tenet, sitting behind him in the Sec-urity Council. "What we are giving you are facts and conclusions based on solid intelligence."

But in the eight weeks since the war ended, American military and intelligence officials in Iraq have failed to confirm the "solid intelligence" provided mainly by Tenet and the CIA. Saddam's deadly germs, chemical weapons and secret programme to build nuclear weapons seem to have vanished with the dictator himself - with the possible, hotly disputed exception of two tractor-trailer beds, discovered by America's Kurdish allies in northern Iraq, which allegedly had been outfitted for the production of biological weapons.

The missing weapons are no minor embarrassment. Baghdad's weapons of mass destruction (WMDs) were repeatedly cited by President Bush and other officials of his administration as justification for a pre-emptive war against Iraq. Last 18 October, Donald Rumsfeld's principal deputy in the defence department, Paul Wolfowitz, described the administration's worst nightmare: chemical, biological or even nuclear weapons provided to al-Qaeda terrorists by Saddam. "Remember now," Wolfowitz told a Washington think-tank, "the dangers we're talking about are not 3,000 dead Americans in a day, but 30,000, or 300,000, or even - God forbid - three million . . ."

The failure to find Saddam's weapons, or the large laboratories and factories necessary to develop and make them, raises two awkward questions: how could America's $30bn-a-year intelligence industry get things so wrong, and why did the White House persuade itself to go to war on the basis of evidence so flimsy?

The House intelligence committee has already asked Tenet to report on the first question after the sort of inquiry that intelligence organisations hate above all others. Post-mortems seek yes-or-no, right-or-wrong answers on matters that by their nature are elusive, hidden and subject to argument. Summary sheets at the top of intelligence estimates may claim that an enemy such as Iraq "has" this or "plans" that, but the analysis that follows will more likely say that "credible" evidence "suggests" X, or that an opponent "might" do Y. The difficulty of the job is compounded when the White House takes an interest in the outcome, and for months stories have been circulating in Washington of CIA analysts complaining bitterly of pressure to find evidence of Iraqi WMDs.

The CIA's line of defence is already apparent: it wasn't the analysts or authors of intelligence estimates who made the flat, alarmist claims, but officials at the policy-making level, especially in the White House and the office of the secretary of defence. They were the ones who squashed doubts, whipped up fears, and pressed the argument for a war to oust Saddam.

But the second question - why the United States went to war on evidence so flimsy - is the one that will cause the most trouble, probably for years to come. It is essentially political in nature, goes to the heart of administration policy and in the absence of a reassuring answer, will threaten the trust traditionally granted to presidents by people and Congress alike. The White House defended its policy of pre-emptive war in the National Security Strategy released last September, but it was Colin Powell on 5 February who made the factual case that Iraq was a rogue state, that it had deadly weapons and was building more of them, and that it might give WMDs to al-Qaeda terrorists. "I went out to the CIA and I spent four days and four nights going over everything they had," Powell told the New York Times at the end of May.

During those four days, Powell, sometimes aided by the national security adviser, Condoleezza Rice, rejected many doubtful claims that had strong support in the White House or the Pentagon - that Saddam had played a role in the terrorist attacks of 11 September 2001, for example, or had tried to acquire Australian software which could map US terrain for a possible attack. But the most important claim that Powell refused to include in his Security Council speech concerned alleged Iraqi efforts to buy from the West African nation of Niger 500 tons of a kind of uranium ore known as "yellow cake" - a sure sign, if true, that Saddam was defying UN resolutions and trying to revive his nuclear weapons programme.

The origins of the yellow cake story are murky, but it appears that the CIA obtained documents about the purchase late last summer - possibly from Ahmad Chalabi's Iraqi National Congress, which had many supporters in the Pentagon. The yellow cake documents were shared with the British, who announced their discovery publicly last 24 September - the same day, according to Seymour Hersh in the New Yorker, that Tenet briefed the Senate foreign relations committee on the yellow cake episode in a closed hearing. Two days later, Hersh reported, Powell also discussed the yellow cake story with the Senate foreign relations committee.

But in the months that followed, doubts were raised about the documents by analysts inside the CIA and the State Department's bureau of intelligence and research. Despite these doubts, Bush, in his State of the Union address of 28 January, cited the yellow cake story as evidence that Saddam "clearly has much to hide".

The story was also included in a suggested draft for Powell's Security Council speech prepared by Vice-President Dick Cheney's chief of staff, Lewis Libby. But when Powell arrived at CIA headquarters in Langley, Virginia, on 1 February, he made no reference to the yellow cake story. About a month later, the yellow cake documents were finally given to the Iraq Nuclear Verification Office of the International Atomic Energy Agency, which had been seeking them ever since their first mention by the British. In less than a day, the IAEA determined that the documents had been fabricated, and crudely at that - one of them, for example, had been allegedly signed by an official who left office more than ten years earlier.

But if Powell successfully spotted one fabrication, he was gulled by another - a 19-page report, Iraq: its infrastructure of concealment, deception and intimidation, released by the British government the day before Powell's speech. "From our sources," Powell told the Security Council on 5 February, "we know that [UN] inspectors are under constant surveillance by an army of Iraqi intelligence operatives . . . I would call my colleagues' attention to the fine paper that the United Kingdom distributed yesterday which describes in exquisite detail Iraqi deception activities."

The "exquisite detail", British journalists soon revealed, came not from British intelligence services, as the release of the document implied, but in largely verbatim chunks from an article in the Middle East Review of International Affairs by a doctoral student in California, Ibrahim al-Marashi, with additional information from Jane's Intelligence Review. Neither was cited.

None of Powell's other claims has been revealed as positively bogus, but with the possible exception of the truck labs, already mentioned, none has been confirmed, either. Some would require Iraqis to step forward and say it's so - for example, that the "weapons experts at one facility" were, in fact, replaced by intelligence officers to deceive UN inspectors, as Powell claimed. This has not occurred, nor have operational files been discovered of Saddam's secret higher committee for monitoring inspection teams, allegedly run by Saddam's son Qusay.

Much of what Powell claimed involved large material things, impossible to misplace - a factory for making poisons and explosives near the northern Iraqi town of Khurmal, four chemical munitions bunkers at Taji, "rocket launchers and warheads containing biological warfare agent . . .", a centre for testing biological and chemical weapons on "1,600 death-row prisoners . . . transferred in 1995 to a special unit" where an Iraqi source saw "blood oozing from the victims' mouths", "a few dozen Scud-variant ballistic missiles". The US president, in his State of the Union address, cited things bigger still - 30,000 Iraqi warheads, 500 tons of chemical weapons, 25,000 litres of anthrax, 38,000 litres of botulinum toxin. None has been found.

President Bush has repeatedly insisted he is "absolutely convinced" the WMDs will turn up, while Rumsfeld has wondered aloud if perhaps Saddam ordered it all destroyed on the eve of war. But no matter what may be discovered in the coming months by the Pentagon's large new Iraqi Survey Group, just now taking to the field to scour the country top to bottom, it is apparent that Bush persuaded himself to go to war, and Congress to vote for war, with intelligence claims that were exaggerated, manipulated, in one case fabricated and in others something worse - plain wrong.

Thomas Powers's Intelligence Wars: American secret history from Hitler to al-Qaeda is published by the New York Review of Books at £16.99