The Iraqal-Qa'ida controversy continued, even after Saddam was long gone from power. Once U.S. forces reached Baghdad, they discovered-stacked where they could easily find them-purported Iraqi intelligence service doc.u.ments that showed much tighter links between Saddam and Zarqawi and Saddam and al-Qa'ida. CIA a.n.a.lysts worked with the U.S. Secret Service to have the paper and ink checked and tried to verify the names and information in the doc.u.ments. Time and again, doc.u.ments that were supposedly produced in the early 1990s turned out to be forgeries. CIA officers interviewed Iraqi intelligence officers in Baghdad who also discounted the authenticity of the doc.u.ments. It was obvious that someone was trying to mislead us. But these raw, unevaluated doc.u.ments that painted a more nefarious picture of Iraq and al-Qa'ida continued to show up in the hands of senior administration officials without having gone through normal intelligence channels.

As one senior a.n.a.lyst put it to me, "The administration is relying too much on flawed information. These are doc.u.ments found on the floor of burnt-out buildings, strewn all over the floor, and taken at face value and not being looked at by trained a.n.a.lysts. Trained a.n.a.lysts would ask questions like, 'What is the source? What do I know about the source?' 'Do they have the access that they claim?' So there is absolutely no standard of a.n.a.lytic tradecraft applied to any of this. Rather, it was presented to us as proof, evidence and confirmation."

On March 13, 2003, we received for our clearance review a speech that had been drafted for the vice president to give on the eve of the war. The proposed speech was sharply at odds with our paper of January 28, 2003, going far beyond the notion of Iraq as a possible training site for al-Qa'ida operatives. The speech draft came to conclusions we could not support, suggesting Iraqi complicity in al-Qa'ida operations.

This prompted a heated conversation between John McLaughlin and Scooter Libby. John subsequently provided in writing detailed reasons why we could not support the speech. "Clearly a policy maker is free to say 'given my read of the intelligence, here is what I make of it,'" John wrote, but he went on to say that the text "goes further than most of our a.n.a.lysts would, implying that Iraq has operational direction and control over al-Qa'ida terrorists." The next morning, just before the president's intelligence briefing, I raised the issue.

"Mr. President," I said, "the vice president wants to make a speech about Iraq and al-Qa'ida that goes way beyond what the intelligence shows. We cannot support the speech and it should not be given." Although I never learned why, the vice president chose not to give his speech.

The push to make the Iraqal-Qa'ida connection didn't end with the start of the war. The November 24, 2003, issue of the Weekly Standard Weekly Standard magazine had a lengthy article called "Case Closed," which was based on a top-secret memorandum that Doug Feith had sent Senate Intelligence Committee chairman Pat Roberts and ranking member Jay Rockefeller a few weeks before. The article claimed that much of the information in the memo contained intelligence "detailed, conclusive and corroborated by multiple sources" showing an "operational relationship" between Usama bin Ladin and Saddam Hussein going back to the early 1990s. magazine had a lengthy article called "Case Closed," which was based on a top-secret memorandum that Doug Feith had sent Senate Intelligence Committee chairman Pat Roberts and ranking member Jay Rockefeller a few weeks before. The article claimed that much of the information in the memo contained intelligence "detailed, conclusive and corroborated by multiple sources" showing an "operational relationship" between Usama bin Ladin and Saddam Hussein going back to the early 1990s.

In fact, much of the material in the memo was the kind of cherry-picked, selective data that Feith, Libby, and others had been enamored of for so long. The Pentagon issued a press statement noting that the memo contained a lot of raw reports but claimed, inaccurately, that the intelligence community had cleared its submission to Congress.

Two months later, Vice President Cheney was in Denver and was asked about the Iraqal-Qa'ida connection. He cited the Weekly Standard Weekly Standard article containing the leaked Feith memo as "your best source of information" on possible ties. I disagree. The best source of information was our January 2003 paper, which said that there was no Iraqi authority, direction, or control over al-Qa'ida. article containing the leaked Feith memo as "your best source of information" on possible ties. I disagree. The best source of information was our January 2003 paper, which said that there was no Iraqi authority, direction, or control over al-Qa'ida.

Stretching the case continues to this day. On the eve of the fifth anniversary of September 11, the vice president appeared on NBC's Meet the Press. Meet the Press. Asked about previous administration comments seeming to link Iraq to 9/11, the vice president ducked the question but referred to testimony I had given a few years before, about contacts between Iraq and al-Qa'ida. "The basis for that is probably best captured in George Tenet's testimony before the Senate Intel Committee, in open session, where he said specifically that there was a pattern of relationship that went back at least a decade between Iraq and al-Qa'ida." On Asked about previous administration comments seeming to link Iraq to 9/11, the vice president ducked the question but referred to testimony I had given a few years before, about contacts between Iraq and al-Qa'ida. "The basis for that is probably best captured in George Tenet's testimony before the Senate Intel Committee, in open session, where he said specifically that there was a pattern of relationship that went back at least a decade between Iraq and al-Qa'ida." On Fox News Sunday Fox News Sunday, Condi Rice was asked a similar question and gave a similar answer. "What the president and I and other administration officials relied on-and you simply rely on the central intelligence. The Director of Central Intelligence, George Tenet, gave that very testimony, that, in fact, there were ties going on between al-Qa'ida and Saddam Hussein's regime going back for a decade. Indeed, the 9/11 Commission talked about contacts between the two."

They quoted my testimony accurately, as far as they went, but both failed to mention that, at the same time, I told them and Congress that our intelligence did not show Iraq and al-Qa'ida had ever moved beyond seeking ways to take advantage of each other. We were aware of no evidence of Baghdad's having "authority, direction and control" of al-Qa'ida operations. In other words, they told only half the story.

CHAPTER 19

Slam Dunk

Many people today believe that my use of the phrase "slam dunk" was the seminal moment for steeling the president's determination to remove Saddam Hussein and to launch the Iraq war. It certainly makes for a memorable sound bite, but it is belied by the facts. Those two words and a meeting that took place in the Oval Office in December 2002 had nothing to do with the president's decision to send American troops into Iraq. That decision had already been made. In fact, the Oval Office meeting came:

* ten months after the president saw the first workable war plan for Iraq;* four months after the vice president's Veterans of Foreign Wars speech in which he said there was "no doubt" that Saddam had weapons of ma.s.s destruction;* three months after the president told the United Nations that the Iraqi regime should "immediately and unconditionally forswear, disclose, and remove or destroy all weapons of ma.s.s destruction, long-range missiles, and all related material";* several months after the U.S. military began repositioning a.s.sets to be used in war to facilities throughout the Middle East;* two months after Congress had authorized the use of force in Iraq; and* two weeks after the Pentagon had issued the first military deployment order sending U.S. troops to the region.

As so often happens with these matters, the context has disappeared, and all that is left are the words themselves, two words that have taken on a significance that far exceeds their import at the time. Let me set the scene.

On Sat.u.r.day, December 21, 2002, I went to the White House for the usual briefing that we delivered to the president six days a week. But that day an additional meeting had also been scheduled after the morning briefing. About two and a half weeks earlier, NSC officials had asked us to start a.s.sembling a public case that might be made against Saddam regarding his possession and possible use of WMD. Although this presentation by CIA would eventually evolve into the speech that Secretary of State Colin Powell delivered to the United Nations, at the time it was not clear who the ultimate audience would be-or even who would present the case. That morning, our charge was simply to a.s.semble materials for a briefing that might someday go public. White House staffers had made it clear that they were looking for an "Adlai Stevenson moment," a reference to Stevenson's famous UN presentation during the Cuban missile crisis, but Bob Walpole had told them that our collected intelligence was nowhere near that categorical.

In the intervening few weeks, a small team of senior a.n.a.lysts had pulled together the requested material. Now it was our turn to deliver it to the president, vice president, Andy Card, Condi Rice, and a few others. The presentation itself fell to John McLaughlin. A champion debater in college, John is not one to go beyond the facts or to stray into bombast. Within and beyond the Agency, John's briefings were well known for being precise, measured, and low key. He had brought some charts with him to ill.u.s.trate his points and an executive a.s.sistant to help with the visual presentation.

It's important to remember both what John was doing that morning and what our charge had been. This meeting was not called for us to mull over the entire issue of Iraq and WMD. Everybody in the room-as well as the most credible intelligence services in the world-already believed that Saddam had chemical and biological weapons and was working on a nuclear program. The incomplete data declarations Saddam was giving to the UN and a stream of information from well-placed intelligence sources served only to b.u.t.tress our confidence. Our job that day wasn't to prove the WMD case, or validate the claim. Our job was to lay out the information relevant to WMD that we believed (a) to be true, and (b) could be cleared for public release without doing damage to intelligence sources and methods. We weren't going to put anything in a public presentation that would jeopardize the lives or continued productivity of precious intelligence sources. Nothing John said in his briefing should have been new to anyone in the room.

Inevitably perhaps, given the high expectations, the substance of John's presentation underwhelmed the audience. This was a first cut, and as most first drafts go, it was very rough. Clearly this didn't compare to the Stevenson moment the White House was searching for. I was disappointed, too. I was sure there was more supporting data in the recently produced NIE, and I felt certain we could find a way to release some of it to the public. Worse, I felt that we had wasted the president's time by giving him an inferior briefing.

George Bush was gracious. "Nice try," he told John, but he quickly added that what he had just heard was not likely to convince "Joe Public." The president suggested that maybe we could add punch to the presentation by bringing in some lawyers who were accustomed to arguing cases before juries. At no time did he or anyone else in the room suggest that we collect more intelligence to find out if the WMD were there or not. As I said, everyone in the room already believed Saddam possessed WMD. The focus was simply on sharpening the arguments. Some might criticize us for partic.i.p.ating in what was essentially a marketing meeting, but intelligence was going to be used in a public presentation and it was our responsibility to ensure that the script was faithful to what we believed to be true and that it placed no sensitive intelligence sources or methods at risk.

To that end, I was asked if we didn't have better information to add to the debate, and I said I was sure we did. I wanted to convey that I thought it would be possible to decla.s.sify enough additional information-communications intercepts, satellite reconnaissance photos, sanitized human intelligence reports, and so forth-to help the public understand what we believed to be true. If I had simply said, "I'm sure we can do better," I wouldn't be writing this chapter-or maybe even this book. Instead, I told the president that strengthening the public presentation was a "slam dunk," a phrase that was later taken completely out of context and has haunted me ever since it first appeared in Bob Woodward's book Plan of Attack Plan of Attack.

Whoever later described the scene to Bob Woodward painted a caricature of me leaping into the air and simulating a slam dunk, not once but twice, with my arms flailing. Credit Woodward's source with a fine sense of the ridiculous, or at least a fine sense of how to make me look ridiculous, but don't credit him or her with a deep sense of obligation to the truth. Even though I am often blunt and p.r.o.ne to talk with my hands, both McLaughlin and I know that this basketball pantomime never happened. In fact, neither John nor his executive a.s.sistant even remember my statement. I certainly don't deny using the term "slam dunk" or strongly believing that Saddam had WMD. But the phrase has, in my view, been intentionally misused and thus completely misunderstood by the public at large.

To double-check John's recall and mine, I asked another CIA officer who was sitting next to me in the Oval Office that morning and who had accompanied me to daily presidential briefings for nearly three years what the officer remembered about the incident. "I am sure you said 'slam dunk,' but it was no more than a pa.s.sing comment. I have been with you when you are really trying to make a point, so I have a basis for comparison. The picture that has been incorrectly portrayed is: You said, 'slam dunk' and they all went, 'Well, we're done. Let's go to war.' But that's not the way it was."

In thinking about all this, I have a few tips for future CIA directors, and for anyone who aspires to partic.i.p.ate in government at a similar level. First, you are never offstage. Anything you say can be used down the road to make someone else's point. That's the way Washington has evolved-there are no private conversations, even in the Oval Office.

Second, in a position such as mine, you owe the president exactness in language. I didn't give him that, and as a result I ended up writing the talking points for those anxious to shift the blame for Iraq away from them and onto CIA in general and me in particular.

Third, I advise future directors of the Agency to be wary of the pitfalls when engaging with policy makers on intelligence related to their policies. On the one hand, if you keep hands off, chances are the intelligence may be misused. On the other, if you engage, you run the risk of seeming to support policy even when you are striving for neutrality.

I can honestly say that we always sought to give the president our best judgments. We did not go beyond our conclusions to justify a policy. Those who feel that we were stretching the case or telling the president what he wanted to hear are simply wrong.

That said, how influential was my comment to the president's thinking? In a way, President Bush and I are much alike. We sometimes say things from our gut, whether it's his "bring 'em on" or my "slam dunk." I think he gets that about me, just as I get that about him. What's more, I think each of us regularly factored that into his understanding of what the other was saying. Other than that, I don't pretend to know what was going through his mind that Sat.u.r.day morning or in the weeks afterward, but there are some hints.

That Christmas Eve, three days after our Oval Office meeting, Jami Miscik was up at Camp David providing the PDB for the president. One of my predecessors was also there, the president's father.

The first President Bush mentioned to Jami that he had heard that there had been an Iraq WMD briefing a few days earlier and that it "hadn't gone well." She later told me that she informed both presidents Bush, father and son, that while there was no "smoking gun" on Iraq WMD, she offered to review the data that had been presented a few days before. In discussing the matter further, she said she was troubled by the lack of intercepted communications one would expect to find with an active WMD program. Human intelligence in a place like Iraq is hard to get, but why there wasn't more signals intelligence was a mystery, she told them.

The second President Bush responded to Jami that there had to be better information that could be presented, but in doing so he made no mention of the "slam dunk" incident. Jami says she never heard that phrase until she read about my purported performance in Bob Woodward's book. That certainly doesn't sound to me like a seminal moment in the decision to go to war.

How is it, then, that an offhand comment made in a closed-door meeting on a Sat.u.r.day morning has come to symbolize so much? I don't think it was an accident. Back in early 2001, when my old mentor Senator David Boren advised me to a.s.sist the new administration for six months before resigning, he added a cautionary note: "Be careful, you are not one of the inner circle going back to the campaign. It doesn't matter how the president may feel; if it suits that group, they will throw you overboard."

If I had cared less about carrying out the Agency's mission in a time of war, I would have heeded the caution.

From the fall of 2003 onward, the security situation in Iraq continued to deteriorate. Rather than acknowledge responsibility, the administration's message was: Don't blame us. George Tenet and the CIA got us into this mess. To this day, certain administration officials continue to use the phrase "slam dunk" as a talking point. In his September 10, 2006, appearance on Meet the Press Meet the Press, in response to a question from Tim Russert, d.i.c.k Cheney referenced me and cited "slam dunk" not once but twice. I remember watching and thinking, "As if you needed me me to say 'slam dunk' to convince to say 'slam dunk' to convince you you to go to war with Iraq." to go to war with Iraq."

Like the vice president and many others, Bob Woodward has not been above using this phrase for his own ends. Shortly after the start of the Iraq war in 2003, White House communications officials had strongly urged CIA cooperation with Woodward on his latest book. We had provided some background, again at White House request, for Woodward's previous book, Bush at War Bush at War, and the administration wanted to replicate what they saw as the PR success of that effort.

I was not at all certain that cooperating this time was a good idea. While the Afghan campaign was then a clear success, the war in Iraq was still unresolved, the hunt for WMD was ongoing, and the rising insurgency in Iraq was proving problematic. Nonetheless, we kept getting calls from the White House saying, "We're cooperating fully with Woodward, and we would like CIA to do so, too."

Accordingly, we provided some senior officials to give Woodward background information, describing our role in the preparation for and conduct of the war. We believed that there was a way, without giving away any secrets, to talk about, for example, the dangerous and vital work done by our case officers who had spent months in northern Iraq gathering intelligence prior to the war.

Woodward was in frequent contact with my spokesman, Bill Harlow, chasing down things he had heard elsewhere and trying to set up interviews. In one background session with a senior CIA official in early 2004, at which I was not present, Woodward offhandedly raised the subject of the December 21, 2002, meeting and the phrase "slam dunk." He made no special issue of it then. Nor did he request that Harlow ask me about the meeting or the context in which the words had been used.

After his ma.n.u.script had gone to print, Woodward mentioned to Harlow that there was going to be something in it that we might find a bit dicey, and he described in greater detail the supposed "slam dunk" scene. Still, he downplayed it and said it was not that big a deal. Maybe that really is how he felt, but when the book came out, following extensive excerpting in the Washington Post Washington Post, "slam dunk" seemed to be all anyone talked about.

Reporters later told Harlow that when they called the White House for reaction to the Woodward book, administration spokesmen were quick to point out the quote. It was, after all, the perfect public-relations deflection. In a situation as complex as the war in Iraq, the public yearns for a simple explanation. Now they had one.

Woodward quotes the president in his book as saying that my "slam dunk" comment was a very important moment. I truly doubt President Bush had any better recollection of the comment than I did. Nor will I ever believe it shaped his view about either the legitimacy or timing of waging war. Far more likely, the president's staff brought up the "slam dunk" scene in the course of prepping him for the Woodward interview-quite possibly the same staff member or members who originally fed the scene to Woodward. They might even have suggested that the president work "slam dunk" into one of his answers if the question was never directly asked. Then, with all the prep work done, the memories "refreshed," Woodward was ushered into the Oval Office, the tape recorder was turned on, and the rest is now history.

I've spoken to Woodward several times since his book came out, and he, of course, doesn't think that he was used or was unfair. He believed the phrase wasn't as big a deal as some might make it. But when he was on television in 2005, defending himself over not originally reporting what he knew about the Valerie Plame and Joe Wilson incident, Woodward said he was too busy in 2003, working on his book and learning important stuff like "slam dunk."

"Slam dunk," he said, "was the basis of this incredibly critical decision the President and his war cabinet were making on, do we invade Iraq?"

I have another two-word reaction to that statement. The first word is "bull."

CHAPTER 20

Taking "the Case" Public

The last thing I ever expected was to be a member of a Greek chorus. But there I was, on international TV, a prop on the set, sitting behind Colin Powell as he spoke to the UN General a.s.sembly on February 5, 2003. Little did I know that as Powell ran through chapter and verse of what we thought thought we knew about Saddam's WMD programs, this drama would later turn out to be a tragedy. we knew about Saddam's WMD programs, this drama would later turn out to be a tragedy.

The speech was the end result of several months of planning, extrapolating, and negotiating. If the United States and our allies were going to win international support for an invasion of Iraq, it was going to take a compelling argument that would turn the legions of skeptics into the "coalition of the willing." The administration debated who could make such a presentation, to whom it would be given and, most important, what would be said.

On a Sat.u.r.day morning shortly after Christmas 2002, John McLaughlin and Bob Walpole were attending yet another meeting at the White House. The subject turned to trying to improve upon the unsatisfactory presentation we had given a week or so before, during the "slam dunk" meeting, and how we could improve on it. The NSC staff suggested drawing from the NIE to bolster the public argument for toppling Saddam. Condi asked Walpole to summarize the Estimate's key judgments. He began doing so from memory, citing all the "we a.s.sess" and "we judge" language that appears in the doc.u.ment.

"Wait a minute," Condi interrupted. "Bob, if you are saying these are a.s.sertions a.s.sertions, we need to know this now." That was the word she used. "We can't send troops to war based on a.s.sertions."

Walpole calmly said that the NIE was an "a.s.sessment" and that these were a.n.a.lytical judgments. He explained that the agencies attached certain levels of confidence to the various judgments-some matters we had high confidence in, others moderate or low-but there was a reason the doc.u.ment's t.i.tle contained the word "estimate."

Condi asked what he meant about confidence levels. Walpole said that, for example, the a.n.a.lysts had "high confidence" that Saddam had chemical weapons.

"What's high confidence, ninety percent?" she asked.

"Yeah, that's about right," Bob replied.

Condi said, "That's a heck of a lot lower than we're getting from reading the PDB." After the war, as part of our lessons-learned efforts, we went back and had a.n.a.lysts review everything the Agency had written regarding Iraq and WMD. We had in fact been much more a.s.sertive in what we were writing for the president on some issues, such as aluminum tubes, than we had been in some of our other publications, including the NIE. Walpole told her that the strongest case for Saddam having weapons of concern was missiles. Walpole was aware that the Iraqis had recently made a declaration to the UN about their Al-Samoud missile. Our experts studied the data and had just concluded that the missile was badly designed and would not reach as far as previously feared. "But you cannot go to war over missiles that exceed authorized ranges by just a few tens of kilometers," he said.

Relying on the information that we would later learn was wrong, Walpole a.s.sured her that the next strongest case was biological weapons. While we had confidence about chemical weapons, Walpole said, that case was largely built on a.n.a.lytical inference. "The weakest case," he explained, "was nuclear." There were alternative views, and the agencies had only moderate confidence in the views that they expressed.

Turning to John McLaughlin, the national security advisor said, "You [the intelligence community] have gotten the president way out on a limb on this."

McLaughlin was stunned and not at all happy about being chastised. He later came back to Langley and told me about the conversation. "We've got them them out on a limb?" he said. It wasn't, after all, the intelligence community that had been clamoring to go to war in Iraq. We had had our hands full with the war on terrorism. out on a limb?" he said. It wasn't, after all, the intelligence community that had been clamoring to go to war in Iraq. We had had our hands full with the war on terrorism.

On January 6, 2003, I attended another meeting in Condi's office along with McLaughlin, Walpole, and Steve Hadley. Hadley noted that the Iraq nuclear case in the proposed speech, a presentation that did not yet have an audience, was weak and needed to be "beefed up." Walpole replied that the draft was weak because the case case was weak. That was why there were alternative views expressed on the issue in the NIE. was weak. That was why there were alternative views expressed on the issue in the NIE.

On January 24, 2003, at still another meeting, Hadley asked Walpole to provide him information on what Saddam needed if he were to obtain nuclear weapons. Walpole replied that that information was contained in the NIE published three months previously.

"Humor me," Hadley said. "The NIE is ninety pages. Can you just excerpt that part and send it to me?"

Walpole subsequently faxed twenty-four pages of material to Hadley for background purposes. Out of that, and out of context, White House officials much later seized on one paragraph from page twenty-four of the NIE to justify including Niger yellowcake and Saddam's nuclear weapons ambitions in the president's State of the Union speech, delivered only days later. Not only did doing so completely ignore the tenor of what we had been telling Rice, Hadley, and others in these meetings, but it also ignited the "sixteen words" flap that would come back to bite us a half year later.

By late January, Colin Powell was picked to make the case for going to war before the United Nations. His mandate was to give a speech that would tell the world why time was running out for Iraq. At one point, Condi Rice and Karen Hughes had urged Powell to speak on three consecutive days. Their vision was that he would speak one day only about Iraq and terrorism. The next day he would address Iraq and human rights. Then he would finish with a lengthy speech about Iraq and WMD. Colin wisely nixed that notion, but it was clear to everyone that this was going to be a speech of extraordinary importance.

Colin asked to come out to CIA headquarters along with several of his speechwriters and senior aides to work through the speech and make sure it was as solid as possible. Although he didn't say so explicitly, I believe one of the reasons he wanted to have the speech worked on at the Agency was the sense that, within our barbed wireencircled headquarters compound, we were relatively free from interference from downtown.

This was an unusual role for us. We had two undesirable options from which to pick. We could let the administration write its own script, knowing that they might easily mischaracterize complex intelligence information, or we could jump in and help craft the speech ourselves. We chose the latter.

We believed Colin would use as a template for his speech a doc.u.ment that grew out of John McLaughlin's infamous presentation in the "slam dunk" session. Bob Walpole had sent the NSC a revised draft weeks earlier based on the NIE, as they requested. When Colin's team first arrived at CIA, they had in their hands a fifty-nine-page doc.u.ment on WMD with which they presumed we were familiar. Powell a.s.sumed that the White House had pulled the doc.u.ment together in coordination with the intelligence community. But what the White House handed him was something very different, something that we had never seen before and that had not been cleared by CIA. Powell's team kept asking us about intelligence underlying elements in the draft, and my staff found themselves repeatedly saying, "We don't know what you are talking about." Colin later told me he saw Scooter Libby at one point and asked, "What are you guys thinking, giving me a draft like that?" Libby reportedly gave him a sheepish look and said, "I wrote it as a lawyer presenting a brief." Powell said the draft looked like it was "a lawyer's brief, not an a.n.a.lytical product."

Eventually, those working on the speech figured out that John Hannah of the vice president's staff was quite familiar with the WMD brief. So, despite the desire to shield the information vetting from kibitzers, they had to ask Hannah to come out to Langley to explain the origins of the material in the speech draft.

Hannah arrived with a stack of raw intelligence, and each time he was asked about some item that had mysteriously appeared in the speech draft, he cited a fragment of information. Time and again, CIA a.n.a.lysts would explain that the information being relied on was fragmentary, unsubstantiated, or had previously been proved wrong. In the end, line after line of the speech draft was thrown out. At one point Hannah asked Mike Morell, who was coordinating the review of the speech for CIA, why the Niger uranium story wasn't in the latest draft. "Because we don't believe it," Mike told him. "I thought you did," Hannah said. After much wrangling and precious time lost in explaining our doubts, Hannah understood why we believed it was inappropriate for Colin to use the Niger material in his speech.

Some members of Secretary Powell's team who partic.i.p.ated in a.s.sembling the speech have subsequently spoken out about the ordeal and given the impression that they were standing alone on the bulwark, keeping out the bad intelligence. That is not how CIA partic.i.p.ants remember it. We had a number of senior intelligence professionals a.s.signed to check the accuracy of what was being said against the intelligence reporting, and others charged with examining the reliability of the sources. Our memory is that CIA and State Department officials worked side by side to rid the draft of material that would not stand up. Our goal from beginning to end was to come up with rhetoric that was both supported by underlying intelligence and worthy of what we all hoped would be a defining moment. Despite our efforts, a lot of flawed information still made its way into the speech. No one involved regrets that more than I do. But I have often wondered whether we might have uncovered more of those flaws if our people had not had to spend two days getting the garbage out of a White House draft that we had never seen before.

The UN speech was supposed to focus mostly on WMD. Weapons of ma.s.s destruction programs were, as Colin once put it, in the UN's "in-box"-in other words, something it was concerned about and responsible for-because Saddam had so consistently ignored United Nations sanctions. The White House staff, however, seemed especially keen on including material about terrorism. In addition to their own piece on WMD, Scooter Libby had provided Powell with a forty-page paper of unknown origin ent.i.tled "Iraq's Dangerous Support for Terror," which the secretary promptly dismissed. They kept suggesting language so far over the top (for example, suggesting possible Iraqi-9/11 connections) that I finally pulled aside Phil Mudd, the then deputy chief of our Counterterrorism Center, and told him to write the terrorism piece of the speech himself.

"It is highly unusual, h.e.l.l, it is practically inappropriate," I told him, for us to write a speech for policy makers. "But if we don't do it, the White House will cram some c.r.a.p in here that we will never live down." Mudd wrote the terrorism portion of the speech, and he did a d.a.m.n good job of it. Despite some problems, that piece of Powell's remarks stands up much better today than does the larger portion on Iraq and WMD.

The process of working on the speech was difficult right up to the end. A handful of senior CIA a.n.a.lysts and I went to New York on February 4 along with Powell and his staff and joined them as he continued to refine and rehea.r.s.e the remarks he planned to give the next day. The one fax machine capable of sending and receiving cla.s.sified material broke down, and we struggled to get last-minute information from Washington and from Powell's staff across town. I stayed up until about two o'clock the night before-actually the morning of-the presentation, working on the terrorism portion of the speech. At last, though, we were all able to agree on a text. After all the back-and-forth, we believed we had produced a solid product.

If Colin had any reservations about giving the speech, he did not tell me. Once he had agreed to undertake the mission, he was going to give it his best shot. Late in the process, Colin asked me to sit behind him at the UN. That was about the last place I wanted to be-I had been scheduled to make an overseas trip to the Middle East at the time-but Powell and his deputy, Rich Armitage, were two of my closest colleagues in the administration. If he wanted me there, I was going to be there, even if my presence was more than a little odd for a serving DCI.

Walking into the UN General a.s.sembly on the morning of February 5 was a surreal moment for me. I sat next to John Negroponte, who at the time was the U.S. amba.s.sador to the UN. After Colin finished what I thought was an extraordinary performance, and other council members began to speak, I left the chamber mentally and physically exhausted.

It was a great presentation, but unfortunately the substance didn't hold up. One by one, the various pillars of the speech, particularly on Iraq's biological and chemical weapons programs, began to buckle. The secretary of state was subsequently hung out to dry in front of the world, and our nation's credibility plummeted.

One particularly d.a.m.ning part of the speech is now so notorious that it deserves special attention. The story begins in 1998, when an Iraqi chemical engineer wandered into a German refugee camp. Within a year or so, he had earned his German immigration card by agreeing to cooperate and provide information to the German Federal Intelligence Service, or BND. The Germans gave the man his perversely prescient code name: Curve Ball.

As intelligence services generally do with their spies, the BND kept its engineer under tight wraps, but eventually shared with the U.S. Defense Intelligence Agency some of the information he was providing them. Curve Ball alleged that Iraqi scientists had a biological weapons program located in mobile laboratories that could be moved to evade UN weapons inspectors.

Because BND controlled the a.s.set tightly and because DIA had responsibility for intelligence from Iraqi refugees in Germany, CIA was twice removed from the source. It was a situation far from ideal. The Germans would not permit either DIA or CIA to have direct access to Curve Ball. They told us that he did not speak English and that he disliked Americans. (It later turned out that his English was pretty good.) We did have one opportunity to observe him when a German-speaking U.S. doctor evaluated him during a physical. The doctor noted that the man appeared hungover and he expressed doubts about his reliability. Those doubts seem prophetic now, but I must say that if we dismissed everything we heard from sources with drinking problems, some accurate intelligence would be thrown out the window.

I've since learned that there were debates between our a.n.a.lysts and our intelligence collectors about the case. Some of the collectors from our Directorate of Operations didn't like the way the case "felt"-they had a gut instinct that there was something wrong with Curve Ball, but little more to go on. The a.n.a.lysts believed pa.s.sionately that the science Curve Ball was describing was accurate-too accurate to be dismissed. There was the fine detail of Curve Ball's reporting-he clearly knew what a mobile lethal-germ lab looked like-and the ever-increasing value of his information as the search for Saddam's WMD mounted.

On balance, and in the absence of any other red flags from the Germans or DIA, Curve Ball appeared to be an invaluable a.s.set. He wasn't. As the Silberman-Robb Commission, a presidential panel looking into Iraq intelligence shortcomings, would report in March 2005, sirens should have been going off all over the place. Whether they were or not is a matter of fierce debate.

Jim Pavitt, the then deputy director of operations and head of the clandestine service, instructed Tyler Drumh.e.l.ler, head of the European Division, to ask for a CIA officer to be allowed to have a face-to-face meeting with the engineer. In late September or early October 2002, Drumh.e.l.ler met with his German counterpart over lunch at a Washington restaurant to convey the request, but got nowhere.

Drumh.e.l.ler, whom I always considered to be a capable officer, now says the German told him, "You do not want to see him [Curve Ball] because he's crazy. Speaking to him would be 'a waste of time.'" The German reportedly went on to say that his service was not sure whether Curve Ball was telling the truth, that he had serious doubts about Curve Ball's mental stability and reliability. Curve Ball, he said, may have had a nervous breakdown. Further, the BND representative worried that Curve Ball was "a fabricator." According to Drumh.e.l.ler's account, the German cautioned, however, that the BND would publicly and officially deny these views if pressed, because they did not wish to be embarra.s.sed.

If that is true, this is how it should have played out: What the German had to say at that lunch in late September or early October 2002 should have been immediately and formally disseminated as a matter of record in a report that would have alerted intelligence and policy officials to the potential problem with Curve Ball. A second, corresponding formal report also should have been instantly sent across the intelligence and policy communities to a.n.a.lysts and policy makers who had received previous Curve Ball reporting. The transmittal of these two reports would have immediately alerted experts doing the work on Iraq WMD issues across the intelligence community to a problem requiring resolution. No such report was disseminated, nor was the issue ever brought to my attention. In fact, I've been told that subsequent investigations have produced not a single piece of paper anywhere at CIA doc.u.menting Drumh.e.l.ler's meeting with the German. The lead a.n.a.lyst on this case in our Weapons Intelligence, Nonproliferation, and Arms Control Center (WINPAC) insists she was never told about the meeting.

Issuing "burn notices," as they are called, on questionable sources is how the system is supposed to work. Because this didn't happen in this instance, we're forced to rely now on the recollection of individuals as to what may or may not have been said or what did or did not occur.

In his testimony before the Silberman-Robb Commission and in interviews subsequent to publication of the commission's findings in early April 2005, Drumh.e.l.ler insisted that the news of the German lunch hit Langley like a small bombsh.e.l.l.

In an April 26, 2005, L.A. Times L.A. Times story, he was even more insistent that word of his meeting with the German had spread broadly through the Agency. He admitted not telling me personally, but he said, "Everyone in the chain of command knew exactly what was happening.... Literally inches and inches of doc.u.mentation," including "dozens and dozens of e-mails and memos," would show that warnings had been sent to John McLaughlin's office and to WINPAC, and that Curve Ball's credibility had been seriously questioned in numerous meetings. story, he was even more insistent that word of his meeting with the German had spread broadly through the Agency. He admitted not telling me personally, but he said, "Everyone in the chain of command knew exactly what was happening.... Literally inches and inches of doc.u.mentation," including "dozens and dozens of e-mails and memos," would show that warnings had been sent to John McLaughlin's office and to WINPAC, and that Curve Ball's credibility had been seriously questioned in numerous meetings.