The committee is concentrating on the last ten years’ worth of reports by the C.I.A. Preliminary findings, one intelligence official told me, are disquieting. “The intelligence community made all kinds of errors and handled things sloppily,” he said. The problems range from a lack of quality control to different agencies’ reporting contradictory assessments at the same time. One finding, the official went on, was that the intelligence reports about Iraq provided by the United Nations inspection teams and the International Atomic Energy Agency, which monitored Iraq’s nuclear-weapons programs, were far more accurate than the C.I.A. estimates. “Some of the old-timers in the community are appalled by how bad the analysis was,” the official said. “If you look at them side by side, C.I.A. versus United Nations, the U.N. agencies come out ahead across the board.”

There were, of course, good reasons to worry about Saddam Hussein’s possession of W.M.D.s. He had manufactured and used chemical weapons in the past, and had experimented with biological weapons; before the first Gulf War, he maintained a multibillion-dollar nuclear-weapons program. In addition, there were widespread doubts about the efficacy of the U.N. inspection teams, whose operations in Iraq were repeatedly challenged and disrupted by Saddam Hussein. Iraq was thought to have manufactured at least six thousand more chemical weapons than the U.N. could account for. And yet, as some former U.N. inspectors often predicted, the tons of chemical and biological weapons that the American public was led to expect have thus far proved illusory. As long as that remains the case, one question will be asked more and more insistently: How did the American intelligence community get it so wrong?

Part of the answer lies in decisions made early in the Bush Administration, before the events of September 11, 2001. In interviews with present and former intelligence officials, I was told that some senior Administration people, soon after coming to power, had bypassed the government’s customary procedures for vetting intelligence.

A retired C.I.A. officer described for me some of the questions that would normally arise in vetting: “Does dramatic information turned up by an overseas spy square with his access, or does it exceed his plausible reach? How does the agent behave? Is he on time for meetings?” The vetting process is especially important when one is dealing with foreign-agent reports—sensitive intelligence that can trigger profound policy decisions. In theory, no request for action should be taken directly to higher authorities—a process known as “stovepiping”—without the information on which it is based having been subjected to rigorous scrutiny.

The point is not that the President and his senior aides were consciously lying. What was taking place was much more systematic—and potentially just as troublesome. Kenneth Pollack, a former National Security Council expert on Iraq, whose book “The Threatening Storm” generally supported the use of force to remove Saddam Hussein, told me that what the Bush people did was “dismantle the existing filtering process that for fifty years had been preventing the policymakers from getting bad information. They created stovepipes to get the information they wanted directly to the top leadership. Their position is that the professional bureaucracy is deliberately and maliciously keeping information from them.

“They always had information to back up their public claims, but it was often very bad information,” Pollack continued. “They were forcing the intelligence community to defend its good information and good analysis so aggressively that the intelligence analysts didn’t have the time or the energy to go after the bad information.”

The Administration eventually got its way, a former C.I.A. official said. “The analysts at the C.I.A. were beaten down defending their assessments. And they blame George Tenet”—the C.I.A. director—“for not protecting them. I’ve never seen a government like this.”

A few months after George Bush took office, Greg Thielmann, an expert on disarmament with the State Department’s Bureau of Intelligence and Research, or INR, was assigned to be the daily intelligence liaison to John Bolton, the Under-Secretary of State for Arms Control, who is a prominent conservative. Thielmann understood that his posting had been mandated by Secretary of State Colin Powell, who thought that every important State Department bureau should be assigned a daily intelligence officer. “Bolton was the guy with whom I had to do business,” Thielmann said. “We were going to provide him with all the information he was entitled to see. That’s what being a professional intelligence officer is all about.”

But, Thielmann told me, “Bolton seemed to be troubled because INR was not telling him what he wanted to hear.” Thielmann soon found himself shut out of Bolton’s early-morning staff meetings. “I was intercepted at the door of his office and told, ‘The Under-Secretary doesn’t need you to attend this meeting anymore.’” When Thielmann protested that he was there to provide intelligence input, the aide said, “The Under-Secretary wants to keep this in the family.”

Eventually, Thielmann said, Bolton demanded that he and his staff have direct electronic access to sensitive intelligence, such as foreign-agent reports and electronic intercepts. In previous Administrations, such data had been made available to under-secretaries only after it was analyzed, usually in the specially secured offices of INR. The whole point of the intelligence system in place, according to Thielmann, was “to prevent raw intelligence from getting to people who would be misled.” Bolton, however, wanted his aides to receive and assign intelligence analyses and assessments using the raw data. In essence, the under-secretary would be running his own intelligence operation, without any guidance or support. “He surrounded himself with a hand-chosen group of loyalists, and found a way to get C.I.A. information directly,” Thielmann said.

In a subsequent interview, Bolton acknowledged that he had changed the procedures for handling intelligence, in an effort to extend the scope of the classified materials available to his office. “I found that there was lots of stuff that I wasn’t getting and that the INR analysts weren’t including,” he told me. “I didn’t want it filtered. I wanted to see everything—to be fully informed. If that puts someone’s nose out of joint, sorry about that.” Bolton told me that he wanted to reach out to the intelligence community but that Thielmann had “invited himself” to his daily staff meetings. “This was my meeting with the four assistant secretaries who report to me, in preparation for the Secretary’s 8:30 a.m. staff meeting,” Bolton said. “This was within my family of bureaus. There was no place for INR or anyone else—the Human Resources Bureau or the Office of Foreign Buildings.”

There was also a change in procedure at the Pentagon under Paul Wolfowitz, the Deputy Secretary of Defense, and Douglas Feith, the Under-Secretary for Policy. In the early summer of 2001, a career official assigned to a Pentagon planning office undertook a routine evaluation of the assumption, adopted by Wolfowitz and Feith, that the Iraqi National Congress, an exile group headed by Ahmad Chalabi, could play a major role in a coup d’état to oust Saddam Hussein. They also assumed that Chalabi, after the coup, would be welcomed by Iraqis as a hero.

An official familiar with the evaluation described how it subjected that scenario to the principle of what planners call “branches and sequels”—that is, “plan for what you expect not to happen.” The official said, “It was a ‘what could go wrong’ study. What if it turns out that Ahmad Chalabi is not so popular? What’s Plan B if you discover that Chalabi and his boys don’t have it in them to accomplish the overthrow?”

The people in the policy offices didn’t seem to care. When the official asked about the analysis, he was told by a colleague that the new Pentagon leadership wanted to focus not on what could go wrong but on what would go right. He was told that the study’s exploration of options amounted to planning for failure. “Their methodology was analogous to tossing a coin five times and assuming that it would always come up heads,” the official told me. “You need to think about what would happen if it comes up tails.”

Getting rid of Saddam Hussein and his regime had been a priority for Wolfowitz and others in and around the Administration since the end of the first Gulf War. For years, Iraq hawks had seen a coup led by Chalabi as the best means of achieving that goal. After September 11th, however, and the military’s quick victory in Afghanistan, the notion of a coup gave way to the idea of an American invasion.

In a speech on November 14, 2001, as the Taliban were being routed in Afghanistan, Richard Perle, a Pentagon consultant with long-standing ties to Wolfowitz, Feith, and Chalabi, articulated what would become the Bush Administration’s most compelling argument for going to war with Iraq: the possibility that, with enough time, Saddam Hussein would be capable of attacking the United States with a nuclear weapon. Perle cited testimony from Dr. Khidhir Hamza, an Iraqi defector, who declared that Saddam Hussein, in response to the 1981 Israeli bombing of the Osiraq nuclear reactor, near Baghdad, had ordered future nuclear facilities to be dispersed at four hundred sites across the nation. “Every day,” Perle said, these sites “turn out a little bit of nuclear materials.” He told his audience, “Do we wait for Saddam and hope for the best, do we wait and hope he doesn’t do what we know he is capable of . . . or do we take some preemptive action?”

In fact, the best case for the success of the U.N. inspection process in Iraq was in the area of nuclear arms. In October, 1997, the International Atomic Energy Agency issued a definitive report declaring Iraq to be essentially free of nuclear weapons. The I.A.E.A.’s inspectors said, “There are no indications that there remains in Iraq any physical capability for the production of amounts of weapon-usable nuclear material of any practical significance.” The report noted that Iraq’s nuclear facilities had been destroyed by American bombs in the 1991 Gulf War.

The study’s main author, Garry Dillon, a British nuclear-safety engineer who spent twenty-three years working for the I.A.E.A. and retired as its chief of inspection, told me that it was “highly unlikely” that Iraq had been able to maintain a secret or hidden program to produce significant amounts of weapons-usable material, given the enormous progress in the past decade in the technical ability of I.A.E.A. inspectors to detect radioactivity in ground locations and in waterways. “This is not kitchen chemistry,” Dillon said. “You’re talking factory scale, and in any operation there are leaks.”

The Administration could offer little or no recent firsthand intelligence to contradict the I.A.E.A.’s 1997 conclusions. During the Clinton years, there had been a constant flow of troubling intelligence reports on Iraqi weapons of mass destruction, but most were in the context of worst-case analyses—what Iraq could do without adequate United Nations inspections—and included few, if any, reliable reports from agents inside the country. The inspectors left in 1998. Many of the new reports that the Bush people were receiving came from defectors who had managed to flee Iraq with help from the Iraqi National Congress. The defectors gave dramatic accounts of Iraq’s efforts to reconstituteits nuclear-weapons program, and of its alleged production of chemical and biological weapons—but the accounts could not be corroborated by the available intelligence.

Greg Thielmann, after being turned away from Bolton’s office, worked with the INR staff on a major review of Iraq’s progress in developing W.M.D.s. The review, presented to Secretary of State Powell in December, 2001, echoed the earlier I.A.E.A. findings. According to Thielmann, “It basically said that there is no persuasive evidence that the Iraqi nuclear program is being reconstituted.”

The defectors, however, had an audience prepared to believe the worst. Secretary of Defense Donald Rumsfeld had long complained about the limits of American intelligence. In the late nineteen-nineties, for example, he had chaired a commission on ballistic-missile programs that criticized the unwillingness of intelligence analysts “to make estimates that extended beyond the hard evidence they had in hand.” After he became Secretary of Defense, a separate intelligence unit was set up in the Pentagon’s policy office, under the control of William Luti, a senior aide to Feith. This office, which circumvented the usual procedures of vetting and transparency, stovepiped many of its findings to the highest-ranking officials.

In the fall of 2001, soon after the September 11th attacks, the C.I.A. received an intelligence report from Italy’s Military Intelligence and Security Service, or sismi, about a public visit that Wissam al-Zahawie, then the Iraqi Ambassador to the Vatican, had made to Niger and three other African nations two and a half years earlier, in February, 1999. The visit had been covered at the time by the local press in Niger and by a French press agency. The American Ambassador, Charles O. Cecil, filed a routine report to Washington on the visit, as did British intelligence. There was nothing untoward about the Zahawie visit. “We reported it because his picture appeared in the paper with the President,” Cecil, who is now retired, told me. There was no article accompanying the photograph, only the caption, and nothing significant to report. At the time, Niger, which had sent hundreds of troops in support of the American-led Gulf War in 1991, was actively seeking economic assistance from the United States.

None of the contemporaneous reports, as far as is known, made any mention of uranium. But now, apparently as part of a larger search for any pertinent information about terrorism, sismi dug the Zahawie-trip report out of its files and passed it along, with a suggestion that Zahawie’s real mission was to arrange the purchase of a form of uranium ore known as “yellowcake.” (Yellowcake, which has been a major Niger export for decades, can be used to make fuel for nuclear reactors. It can also be converted, if processed differently, into weapons-grade uranium.)

What made the two-and-a-half-year-old report stand out in Washington was its relative freshness. A 1999 attempt by Iraq to buy uranium ore, if verified, would seem to prove that Saddam had been working to reconstitute his nuclear program—and give the lie to the I.A.E.A. and to intelligence reports inside the American government that claimed otherwise.

The sismi report, however, was unpersuasive. Inside the American intelligence community, it was dismissed as amateurish and unsubstantiated. One former senior C.I.A. official told me that the initial report from Italy contained no documents but only a written summary of allegations. “I can fully believe that sismi would put out a piece of intelligence like that,” a C.I.A. consultant told me, “but why anybody would put credibility in it is beyond me.” No credible documents have emerged since to corroborate it.

The intelligence report was quickly stovepiped to those officials who had an intense interest in building the case against Iraq, including Vice-President Dick Cheney. “The Vice-President saw a piece of intelligence reporting that Niger was attempting to buy uranium,” Cathie Martin, the spokeswoman for Cheney, told me. Sometime after he first saw it, Cheney brought it up at his regularly scheduled daily briefing from the C.I.A., Martin said. “He asked the briefer a question. The briefer came back a day or two later and said, ‘We do have a report, but there’s a lack of details.’” The Vice-President was further told that it was known that Iraq had acquired uranium ore from Niger in the early nineteen-eighties but that that material had been placed in secure storage by the I.A.E.A., which was monitoring it. “End of story,” Martin added. “That’s all we know.” According to a former high-level C.I.A. official, however, Cheney was dissatisfied with the initial response, and asked the agency to review the matter once again. It was the beginning of what turned out to be a year-long tug-of-war between the C.I.A. and the Vice-President’s office.

As the campaign against Iraq intensified, a former aide to Cheney told me, the Vice-President’s office, run by his chief of staff, Lewis (Scooter) Libby, became increasingly secretive when it came to intelligence about Iraq’s W.M.D.s. As with Wolfowitz and Bolton, there was a reluctance to let the military and civilian analysts on the staff vet intelligence.

“It was an unbelievably closed and small group,” the former aide told me. Intelligence procedures were far more open during the Clinton Administration, he said, and professional staff members had been far more involved in assessing and evaluating the most sensitive data. “There’s so much intelligence out there that it’s easy to pick and choose your case,” the former aide told me. “It opens things up to cherry-picking.” (“Some reporting is sufficiently sensitive that it is restricted only to the very top officials of the government—as it should be,” Cathie Martin said. And any restrictions, she added, emanate from C.I.A. security requirements.)

By early 2002, the sismi intelligence—still unverified—had begun to play a role in the Administration’s warnings about the Iraqi nuclear threat. On January 30th, the C.I.A. published an unclassified report to Congress that stated, “Baghdad may be attempting to acquire materials that could aid in reconstituting its nuclear-weapons program.” A week later, Colin Powell told the House International Relations Committee, “With respect to the nuclear program, there is no doubt that the Iraqis are pursuing it.”

The C.I.A. assessment reflected both deep divisions within the agency and the position of its director, George Tenet, which was far from secure. (The agency had been sharply criticized, after all, for failing to provide any effective warning of the September 11th attacks.) In the view of many C.I.A. analysts and operatives, the director was too eager to endear himself to the Administration hawks and improve his standing with the President and the Vice-President. Senior C.I.A. analysts dealing with Iraq were constantly being urged by the Vice-President’s office to provide worst-case assessments on Iraqi weapons issues. “They got pounded on, day after day,” one senior Bush Administration official told me, and received no consistent backup from Tenet and his senior staff. “Pretty soon you say ‘Fuck it.’” And they began to provide the intelligence that was wanted.

In late February, the C.I.A. persuaded retired Ambassador Joseph Wilson to fly to Niger to discreetly check out the story of the uranium sale. Wilson, who is now a business consultant, had excellent credentials: he had been deputy chief of mission in Baghdad, had served as a diplomat in Africa, and had worked in the White House for the National Security Council. He was known as an independent diplomat who had put himself in harm’s way to help American citizens abroad.

Wilson told me he was informed at the time that the mission had come about because the Vice-President’s office was interested in the Italian intelligence report. Before his departure, he was summoned to a meeting at the C.I.A. with a group of government experts on Iraq, Niger, and uranium. He was shown no documents but was told, he said, that the C.I.A. “was responding to a report that was recently received of a purported memorandum of agreement”—between Iraq and Niger—“that our boys had gotten.” He added, “It was never clear to me, or to the people who were briefing me, whether our guys had actually seen the agreement, or the purported text of an agreement.” Wilson’s trip to Niger, which lasted eight days, produced nothing. He learned that any memorandum of understanding to sell yellowcake would have required the signatures of Niger’s Prime Minister, Foreign Minister, and Minister of Mines. “I saw everybody out there,” Wilson said, and no one had signed such a document. “If a document purporting to be about the sale contained those signatures, it would not be authentic.” Wilson also learned that there was no uranium available to sell: it had all been pre-sold to Niger’s Japanese and European consortium partners.

Wilson returned to Washington and made his report. It was circulated, he said, but “I heard nothing about what the Vice-President’s office thought about it.” (In response, Cathie Martin said, “The Vice-President doesn’t know Joe Wilson and did not know about his trip until he read about it in the press.” The first press accounts appeared fifteen months after Wilson’s trip.)

By early March, 2002, a former White House official told me, it was understood by many in the White House that the President had decided, in his own mind, to go to war. The undeclared decision had a devastating impact on the continuing struggle against terrorism. The Bush Administration took many intelligence operations that had been aimed at Al Qaeda and other terrorist groups around the world and redirected them to the Persian Gulf. Linguists and special operatives were abruptly reassigned, and several ongoing anti-terrorism intelligence programs were curtailed.

Chalabi’s defector reports were now flowing from the Pentagon directly to the Vice-President’s office, and then on to the President, with little prior evaluation by intelligence professionals. When INR analysts did get a look at the reports, they were troubled by what they found. “They’d pick apart a report and find out that the source had been wrong before, or had no access to the information provided,” Greg Thielmann told me. “There was considerable skepticism throughout the intelligence community about the reliability of Chalabi’s sources, but the defector reports were coming all the time. Knock one down and another comes along. Meanwhile, the garbage was being shoved straight to the President.”

A routine settled in: the Pentagon’s defector reports, classified “secret,” would be funnelled to newspapers, but subsequent C.I.A. and INR analyses of the reports—invariably scathing but also classified—would remain secret.

“It became a personality issue,” a Pentagon consultant said of the Bush Administration’s handling of intelligence. “My fact is better than your fact. The whole thing is a failure of process. Nobody goes to primary sources.” The intelligence community was in full retreat.

In the spring of 2002, the former White House official told me, Rumsfeld and Wolfowitz began urging the President to release more than ninety million dollars in federal funds to Chalabi. The 1998 Iraq Liberation Act had authorized ninety-seven million dollars for the Iraqi opposition, but most of the funds had not been expended. The State Department opposed releasing the rest of the money, arguing that Chalabi had failed to account properly for the funds he had already received. “The Vice-President came into a meeting furious that we hadn’t given the money to Chalabi,” the former official recalled. Cheney said, “Here we are, denying him money, when they”—the Iraqi National Congress—“are providing us with unique intelligence on Iraqi W.M.D.s.”

In late summer, the White House sharply escalated the nuclear rhetoric. There were at least two immediate targets: the midterm congressional elections and the pending vote on a congressional resolution authorizing the President to take any action he deemed necessary in Iraq, to protect America’s national security.

On August 7th, Vice-President Cheney, speaking in California, said of Saddam Hussein, “What we know now, from various sources, is that he . . . continues to pursue a nuclear weapon.” On August 26th, Cheney suggested that Saddam had a nuclear capability that could directly threaten “anyone he chooses, in his own region or beyond.” He added that the Iraqis were continuing “to pursue the nuclear program they began so many years ago.” On September 8th, he told a television interviewer, “We do know, with absolute certainty, that he is using his procurement system to acquire the equipment he needs in order to enrich uranium to build a nuclear weapon.” The President himself, in his weekly radio address on September 14th, stated, “Saddam Hussein has the scientists and infrastructure for a nuclear-weapons program, and has illicitly sought to purchase the equipment needed to enrich uranium for a nuclear weapon.” There was no confirmed intelligence for the President’s assertion.

The government of the British Prime Minister, Tony Blair, President Bush’s closest ally, was also brought in. As Blair later told a British government inquiry, he and Bush had talked by telephone that summer about the need “to disclose what we knew or as much as we could of what we knew.” Blair loyally took the lead: on September 24th, the British government issued a dossier dramatizing the W.M.D. threat posed by Iraq. In a foreword, Blair proclaimed that “the assessed intelligence has established beyond doubt that Saddam . . . continues in his efforts to develop nuclear weapons.” The dossier noted that intelligence—based, again, largely on the sismi report—showed that Iraq had “sought significant quantities of uranium from Africa.” A subsequent parliamentary inquiry determined that the published statement had been significantly toned down after the C.I.A. warned its British counterpart not to include the claim in the dossier, and in the final version Niger was not named, nor was sismi.

The White House, meanwhile, had been escalating its rhetoric. In a television interview on September 8th, Condoleezza Rice, the national-security adviser, addressing questions about the strength of the Administration’s case against Iraq, said, “We don’t want the smoking gun to be a mushroom cloud”—a formulation that was taken up by hawks in the Administration. And, in a speech on October 7th, President Bush said, “Facing clear evidence of peril, we cannot wait for the final proof—the smoking gun—that could come in the form of a mushroom cloud.”

At that moment, in early October, 2002, a set of documents suddenly appeared that promised to provide solid evidence that Iraq was attempting to reconstitute its nuclear program. The first notice of the documents’ existence came when Elisabetta Burba, a reporter for Panorama, a glossy Italian weekly owned by the publishing empire of Prime Minister Silvio Berlusconi, received a telephone call from an Italian businessman and security consultant whom she believed to have once been connected to Italian intelligence. He told her that he had information connecting Saddam Hussein to the purchase of uranium in Africa. She considered the informant credible. In 1995, when she worked for the magazine Epoca, he had provided her with detailed information, apparently from Western intelligence sources, for articles she published dealing with the peace process in Bosnia and with an Islamic charity that was linked to international terrorism. The information, some of it in English, proved to be accurate. Epoca had authorized her to pay around four thousand dollars for the documents—a common journalistic practice in Italy.

Now, years later, “he comes to me again,” Burba told me. “I knew he was an informed person, and that he had contacts all over the world, including in the Middle East. He deals with investment and security issues.” When Burba met with the man, he showed her the Niger documents and offered to sell them to her for about ten thousand dollars.

The documents he gave her were photocopies. There were twenty-two pages, mostly in French, some with the letterhead of the Niger government or Embassy, and two on the stationery of the Iraqi Embassy to the Holy See. There were also telexes. When Burba asked how the documents could be authenticated, the man produced what appeared to be a photocopy of the codebook from the Niger Embassy, along with other items. “What I was sure of was that he had access,” Burba said. “He didn’t receive the documents from the moon.”

The documents dealt primarily with the alleged sale of uranium, Burba said. She informed her editors, and shared the photocopies with them. She wanted to arrange a visit to Niger to verify what seemed to be an astonishing story. At that point, however, Panorama’s editor-in-chief, Carlo Rossella, who is known for his ties to the Berlusconi government, told Burba to turn the documents over to the American Embassy for authentication. Burba dutifully took a copy of the papers to the Embassy on October 9th.

A week later, Burba travelled to Niger. She visited mines and the ports that any exports would pass through, spoke to European businessmen and officials informed about Niger’s uranium industry, and found no trace of a sale. She also learned that the transport company and the bank mentioned in the papers were too small and too ill-equipped to handle such a transaction. As Ambassador Wilson had done eight months earlier, she concluded that there was no evidence of a recent sale of yellowcake to Iraq. The Panorama story was dead, and Burba and her editors said that no money was paid. The documents, however, were now in American hands.

Two former C.I.A. officials provided slightly different accounts of what happened next. “The Embassy was alerted that the papers were coming,” the first former official told me, “and it passed them directly to Washington without even vetting them inside the Embassy.” Once the documents were in Washington, they were forwarded by the C.I.A. to the Pentagon, he said. “Everybody knew at every step of the way that they were false—until they got to the Pentagon, where they were believed.”

The documents were just what Administration hawks had been waiting for. The second former official, Vincent Cannistraro, who served as chief of counter-terrorism operations and analysis, told me that copies of the Burba documents were given to the American Embassy, which passed them on to the C.I.A.’s chief of station in Rome, who forwarded them to Washington. Months later, he said, he telephoned a contact at C.I.A. headquarters and was told that “the jury was still out on this”—that is, on the authenticity of the documents.

George Tenet clearly was ambivalent about the information: in early October, he intervened to prevent the President from referring to Niger in a speech in Cincinnati. But Tenet then seemed to give up the fight, and Saddam’s desire for uranium from Niger soon became part of the Administration’s public case for going to war.

On December 7th, the Iraqi regime provided the U.N. Security Council with a twelve-thousand-page series of documents in which it denied having a W.M.D. arsenal. Very few in the press, the public, or the White House believed it, and a State Department rebuttal, on December 19th, asked, “Why is the Iraqi regime hiding their Niger procurement?” It was the first time that Niger had been publicly identified. In a January 23rd Op-Ed column in the Times, entitled “Why We Know Iraq Is Lying,” Condoleezza Rice wrote that the “false declaration . . . fails to account for or explain Iraq’s efforts to get uranium from abroad.” On January 26th, Secretary Powell, speaking at the World Economic Forum in Davos, Switzerland, asked, “Why is Iraq still trying to procure uranium?” Two days later, President Bush described the alleged sale in his State of the Union address, saying, “The British government has learned that Saddam Hussein recently sought significant quantities of uranium from Africa.”

Who produced the fake Niger papers? There is nothing approaching a consensus on this question within the intelligence community. There has been published speculation about the intelligence services of several different countries. One theory, favored by some journalists in Rome, is that sismi produced the false documents and passed them to Panorama for publication.

Another explanation was provided by a former senior C.I.A. officer. He had begun talking to me about the Niger papers in March, when I first wrote about the forgery, and said, “Somebody deliberately let something false get in there.” He became more forthcoming in subsequent months, eventually saying that a small group of disgruntled retired C.I.A. clandestine operators had banded together in the late summer of last year and drafted the fraudulent documents themselves.

“The agency guys were so pissed at Cheney,” the former officer said. “They said, ‘O.K, we’re going to put the bite on these guys.’” My source said that he was first told of the fabrication late last year, at one of the many holiday gatherings in the Washington area of past and present C.I.A. officials. “Everyone was bragging about it—‘Here’s what we did. It was cool, cool, cool.’” These retirees, he said, had superb contacts among current officers in the agency and were informed in detail of the sismi intelligence.

“They thought that, with this crowd, it was the only way to go—to nail these guys who were not practicing good tradecraft and vetting intelligence,” my source said. “They thought it’d be bought at lower levels—a big bluff.” The thinking, he said, was that the documents would be endorsed by Iraq hawks at the top of the Bush Administration, who would be unable to resist flaunting them at a press conference or an interagency government meeting. They would then look foolish when intelligence officials pointed out that they were obvious fakes. But the tactic backfired, he said, when the papers won widespread acceptance within the Administration. “It got out of control.”

Like all large institutions, C.I.A. headquarters, in Langley, Virginia, is full of water-cooler gossip, and a retired clandestine officer told me this summer that the story about a former operations officer faking the documents is making the rounds. “What’s telling,” he added, “is that the story, whether it’s true or not, is believed”—an extraordinary commentary on the level of mistrust, bitterness, and demoralization within the C.I.A. under the Bush Administration. (William Harlow, the C.I.A. spokesman, said that the agency had no more evidence that former members of the C.I.A. had forged the documents “than we have that they were forged by Mr. Hersh.”)

The F.B.I. has been investigating the forgery at the request of the Senate Intelligence Committee. A senior F.B.I. official told me that the possibility that the documents were falsified by someone inside the American intelligence community had not been ruled out. “This story could go several directions,” he said. “We haven’t gotten anything solid, and we’ve looked.” He said that the F.B.I. agents assigned to the case are putting a great deal of effort into the investigation. But “somebody’s hiding something, and they’re hiding it pretty well.”

President Bush’s State of the Union speech had startled Elisabetta Burba, the Italian reporter. She had been handed documents and had personally taken them to the American Embassy, and she now knew from her trip to Niger that they were false. Later, Burba revisited her source. “I wanted to know what happened,” she said. “He told me that he didn’t know the documents were false, and said he’d also been fooled. ”

Burba, convinced that she had the story of the year, wanted to publish her account immediately after the President’s speech, but Carlo Rossella, Panorama’s editor-in-chief, decided against it. Rossella explained to me, “When I heard the State of the Union statement, I thought to myself that perhaps the United States government has other information. I didn’t think the documents were that important—they weren’t trustable.” Eventually, in July, after her name appeared in the press, Burba published an account of her role. She told me that she was interviewed at the American consulate in Milan by three agents for the F.B.I. in early September.

The State of the Union speech was confounding to many members of the intelligence community, who could not understand how such intelligence could have got to the President without vetting. The former intelligence official who gave me the account of the forging of the documents told me that his colleagues were also startled by the speech. “They said, ‘Holy shit, all of a sudden the President is talking about it in the State of the Union address!’ They began to panic. Who the hell was going to expose it? They had to build a backfire. The solution was to leak the documents to the I.A.E.A.”

I subsequently met with a group of senior I.A.E.A. officials in Vienna, where the organization has its headquarters. In an interview over dinner, they told me that they did not even know the papers existed until early February of this year, a few days after the President’s speech. The I.A.E.A. had been asking Washington and London for their evidence of Iraq’s pursuit of African uranium, without receiving any response, ever since the previous September, when word of it turned up in the British dossier. After Niger was specified in the State Department’s fact sheet of December 19, 2002, the I.A.E.A. became more insistent. “I started to harass the United States,” recalled Jacques Baute, a Frenchman who, as director of the I.A.E.A.’s Iraq Nuclear Verification Office, often harassed Washington. Mark Gwozdecky, the I.A.E.A.’s spokesman, added, “We were asking for actionable evidence, and Jacques was getting almost nothing. ”

On February 4, 2003, while Baute was on a plane bound for New York to attend a United Nations Security Council meeting on the Iraqi weapons dispute, the U.S. Mission in Vienna suddenly briefed members of Baute’s team on the Niger papers, but still declined to hand over the documents. “I insisted on seeing the documents myself,” Baute said, “and was provided with them upon my arrival in New York.” The next day, Secretary Powell made his case for going to war against Iraq before the U.N. Security Council. The presentation did not mention Niger—a fact that did not escape Baute. I.A.E.A. officials told me that they were puzzled by the timing of the American decision to provide the documents. Baute quickly concluded that they were fake.

Over the next few weeks, I.A.E.A. officials conducted further investigations, which confirmed the fraud. They also got in touch with American and British officials to inform them of the findings, and give them a chance to respond. Nothing was forthcoming, and so the I.A.E.A.’s director-general, Mohamed ElBaradei, publicly described the fraud at his next scheduled briefing to the U.N. Security Council, in New York on March 7th. The story slowly began to unravel.

Vice-President Cheney responded to ElBaradei’s report mainly by attacking the messenger. On March 16th, Cheney, appearing on “Meet the Press,” stated emphatically that the United States had reason to believe that Saddam Hussein had reconstituted his nuclear-weapons program. He went on, “I think Mr. ElBaradei frankly is wrong. And I think if you look at the track record of the International Atomic Energy Agency on this kind of issue, especially where Iraq’s concerned, they have consistently underestimated or missed what it was Saddam Hussein was doing. I don’t have any reason to believe they’re any more valid this time than they’ve been in the past.” Three days later, the war in Iraq got under way, and the tale of the African-uranium-connection forgery sank from view.

Joseph Wilson, the diplomat who had travelled to Africa to investigate the allegation more than a year earlier, revived the Niger story. He was angered by what he saw as the White House’s dishonesty about Niger, and in early May he casually mentioned his mission to Niger, and his findings, during a brief talk about Iraq at a political conference in suburban Washington sponsored by the Senate Democratic Policy Committee (Wilson is a Democrat). Another speaker at the conference was the Times columnist Nicholas Kristof, who got Wilson’s permission to mention the Niger trip in a column. A few months later, on July 6th, Wilson wrote about the trip himself on the Times Op-Ed page. “I gave them months to correct the record,” he told me, speaking of the White House, “but they kept on lying.”

The White House responded by blaming the intelligence community for the Niger reference in the State of the Union address. Condoleezza Rice, the national-security adviser, told a television interviewer on July 13th, “Had there been even a peep that the agency did not want that sentence in or that George Tenet did not want that sentence . . . it would have been gone.” Five days later, a senior White House official went a step further, telling reporters at a background briefing that they had the wrong impression about Joseph Wilson’s trip to Niger and the information it had yielded. “You can’t draw a conclusion that we were warned by Ambassador Wilson that this was all dubious,” the unnamed official said, according to a White House transcript. “It’s just not accurate.”

But Wilson’s account of his trip forced a rattled White House to acknowledge, for the first time, that “this information should not have risen to the level of a Presidential speech.” It also triggered retaliatory leaks to the press by White House officials that exposed Wilson’s wife as a C.I.A. operative—and led to an F.B.I. investigation.

Among the best potential witnesses on the subject of Iraq’s actual nuclear capabilities are the men and women who worked in the Iraqi weapons industry and for the National Monitoring Directorate, the agency set up by Saddam to work with the United Nations and I.A.E.A. inspectors. Many of the most senior weapons-industry officials, even those who voluntarily surrendered to U.S. forces, are being held in captivity at the Baghdad airport and other places, away from reporters. Their families have been told little by American authorities. Desperate for information, they have been calling friends and other contacts in America for help.

One Iraqi émigré who has heard from the scientists’ families is Shakir al Kha Fagi, who left Iraq as a young man and runs a successful business in the Detroit area. “The people in intelligence and in the W.M.D. business are in jail,” he said. “The Americans are hunting them down one by one. Nobody speaks for them, and there’s no American lawyer who will take the case.”

Not all the senior scientists are in captivity, however. Jafar Dhia Jafar, a British-educated physicist who coördinated Iraq’s efforts to make the bomb in the nineteen-eighties, and who had direct access to Saddam Hussein, fled Iraq in early April, before Baghdad fell, and, with the help of his brother, Hamid, the managing director of a large energy company, made his way to the United Arab Emirates. Jafar has refused to return to Baghdad, but he agreed to be debriefed by C.I.A. and British intelligence agents. There were some twenty meetings, involving as many as fifteen American and British experts. The first meeting, on April 11th, began with an urgent question from a C.I.A. officer: “Does Iraq have a nuclear device? The military really want to know. They are extremely worried.” Jafar’s response, according to the notes of an eyewitness, was to laugh. The notes continued:

Jafar insisted that there was not only no bomb, but no W.M.D., period. “The answer was none.” . . . Jafar explained that the Iraqi leadership had set up a new committee after the 91 Gulf war, and after the unscom [United Nations] inspection process was set up. . . and the following instructions [were sent] from the Top Man [Saddam]—“give them everything.”

The notes said that Jafar was then asked, “But this doesn’t mean all W.M.D.? How can you be certain?” His answer was clear: “I know all the scientists involved, and they chat. There is no W.M.D.”

Jafar explained why Saddam had decided to give up his valued weapons:

Up until the 91 Gulf war, our adversaries were regional. . . . But after the war, when it was clear that we were up against the United States, Saddam understood that these weapons were redundant. “No way we could escape the United States.” Therefore, the W.M.D. warheads did Iraq little strategic good.

Jafar had his own explanation, according to the notes, for one of the enduring mysteries of the U.N. inspection process—the six-thousand-warhead discrepancy between the number of chemical weapons thought to have been manufactured by Iraq before 1991 and the number that were accounted for by the U.N. inspection teams. It was this discrepancy which led Western intelligence officials and military planners to make the worst-case assumptions. Jafar told his interrogators that the Iraqi government had simply lied to the United Nations about the number of chemical weapons used against Iran during the brutal Iran-Iraq war in the nineteen-eighties. Iraq, he said, dropped thousands more warheads on the Iranians than it acknowledged. For that reason, Saddam preferred not to account for the weapons at all.

There are always credibility problems with witnesses from a defeated regime, and anyone involved in the creation or concealment of W.M.D.s. would have a motive to deny it. But a strong endorsement of Jafar’s integrity came from an unusual source—Jacques Baute, of the I.A.E.A., who spent much of the past decade locked in a struggle with Jafar and the other W.M.D. scientists and technicians of Iraq. “I don’t believe anybody,” Baute told me, “but, by and large, what he told us after 1995 was pretty accurate.”

In early October, David Kay, the former U.N. inspector who is the head of the Administration’s Iraq Survey Group, made his interim report to Congress on the status of the search for Iraq’s W.M.D.s. “We have not yet found stocks of weapons,” Kay reported, “but we are not yet at the point where we can say definitively either that such weapon stocks do not exist or that they existed before the war.” In the area of nuclear weapons, Kay said, “Despite evidence of Saddam’s continued ambition to acquire nuclear weapons, to date we have not uncovered evidence that Iraq undertook significant post-1998 steps to actually build nuclear weapons or produce fissile material.” Kay was widely seen as having made the best case possible for President Bush’s prewar claims of an imminent W.M.D. threat. But what he found fell far short of those claims, and the report was regarded as a blow to the Administration. President Bush, however, saw it differently. He told reporters that he felt vindicated by the report, in that it showed that “Saddam Hussein was a threat, a serious danger.”

The President’s response raises the question of what, if anything, the Administration learned from the failure, so far, to find significant quantities of W.M.D.s in Iraq. Any President depends heavily on his staff for the vetting of intelligence and a reasonable summary and analysis of the world’s day-to-day events. The ultimate authority in the White House for such issues lies with the President’s national-security adviser—in this case,Condoleezza Rice. The former White House official told me, “Maybe the Secretary of Defense and his people are short-circuiting the process, and creating a separate channel to the Vice-President. Still, at the end of the day all the policies have to be hashed out in the interagency process, led by the national-security adviser.” What happened instead, he said, “was a real abdication of responsibility by Condi.”

Vice-President Cheney remains unabashed about the Administration’s reliance on the Niger documents, despite the revelation of their forgery. In a September interview on “Meet the Press,” Cheney claimed that the British dossier’s charge that “Saddam was, in fact, trying to acquire uranium in Africa” had been “revalidated.” Cheney went on, “So there may be a difference of opinion there. I don’t know what the truth is on the ground. . . . I don’t know Mr. Wilson. I probably shouldn’t judge him.”

The Vice-President also defended the way in which he had involved himself in intelligence matters: “This is a very important area. It’s one that the President has asked me to work on. . . . In terms of asking questions, I plead guilty. I ask a hell of a lot of questions. That’s my job.”