Shortly thereafter, Kay returned to Baghdad to resume the weapons search. In his absence, the already shaky security situation there had deteriorated considerably. On October 9, a suicide bomber drove his car into a group of Baghdad policemen, killing nine and injuring forty-five. Three days later, a bombing outside a Baghdad hotel used by senior Coalition officials killed at least eight. On October 27, the first day of Ramadan, four coordinated suicide attacks against three other Baghdad police stations and the Islamic Red Crescent killed forty-three people and wounded more than two hundred. Six days later, on November 2, sixteen U.S. soldiers were killed and twenty-one injured when a helicopter was shot down. November would go on to be the bloodiest month up to that point for U.S. military personnel, with seventy-five dead.

Central Command generals were scrambling to try to find out where the attacks were coming from and to figure out ways to stop them. Not surprisingly, they looked to the Iraq Survey Group as a resource for a.n.a.lysts who might help stop the bleeding. The request wasn't large-Central Command was seeking the temporary loan of a handful of area experts, in the single digits-but Kay objected.

A senior military officer later told me of a conversation he had with Kay. The official was "flabbergasted," he said, when Kay refused to lend some of the ISG's experienced intelligence a.n.a.lysts to help him find insurgents "that are killing us." "that are killing us." Kay said he could not afford to do that because it would "destroy his operation." He didn't want any a.s.sets pulled away from the weapons hunt, despite the fact that the insurgency was making the mission of his troops nearly impossible to complete. Kay said he could not afford to do that because it would "destroy his operation." He didn't want any a.s.sets pulled away from the weapons hunt, despite the fact that the insurgency was making the mission of his troops nearly impossible to complete.

After weighing many competing demands, John McLaughlin managed to identify a few other intelligence community personnel who could be sent to Iraq to replace anyone diverted from the ISG staff. Still, Kay could not be placated. On our periodic video conferences with him, he became obstreperous, claiming that he was not getting the support he needed to do his job. In one call with McLaughlin, Kay said that he would not "stake his name and reputation" on this mission unless he got everything he wanted.

Had he been a regular CIA officer, I would have relieved Kay of his command and ordered him home. American servicemen and women were dying; Gen. John Abizaid needed help. Instead, McLaughlin made a visit to Iraq in November and met with Kay and the ISG leadership. McLaughlin expected at least small thanks for the difficult choices that were being made to divert people to Kay's mission, but instead he found Kay quite brusque, and insistent that his needs be met.

On November 19, a month and a half after Kay had told the media his mission would require six to nine more months, I learned from the rumor mill he was planning on quitting that very day. I called him, and he confirmed the rumor. He didn't provide any rationale for wanting to abandon his post other than expressing a general unhappiness that anyone in the ISG might be asked to help control the increasingly deadly insurgency. His threat to resign would occur on the day President Bush was arriving in London for a state visit to the United Kingdom. "No," I said in reply. "I won't allow you to embarra.s.s the president in that way." I reminded Kay, without much effect, of the extraordinary support and a.s.sets he had been provided and the importance of coming to some final resolution on WMD. "Look David," I said, "why don't you come home for the holidays, take some time off to think about continuing the job?" He agreed to do so, but when he left Baghdad, his colleagues couldn't help but notice that he cleaned out the trailer he was living in and took home all his personal effects.

While Kay spent much of December decompressing, Maj. Gen. Keith Dayton, the senior military officer who headed the Iraq Survey Group, led the search for WMD. Then, sometime around Christmas, Kay informed us that his mind was made up and that he was not going back. I asked that he withhold any public announcement of his departure until we could identify a suitable replacement for Kay.

John McLaughlin undertook the effort to find a replacement for Kay. We developed a list of about five candidates and began checking them out. As McLaughlin gathered recommendations from various proliferation experts, one name kept coming up-Charles Duelfer, a former UN weapons inspector with a wealth of experience on the ground in Iraq and inside the UN, where he had served as the deputy chief of the weapons inspection effort.

Duelfer had the reputation for being iron-willed and dogged about his work. One person McLaughlin talked to cautioned us that Duelfer had a strong independent streak and was no slave to bureaucracy. But that was exactly what we were looking for. Although the security situation had continued to slide downhill, we felt certain Duelfer wouldn't be intimidated. After all, the guy does free-fall skydiving for fun. And from his past lengthy time on the ground in Iraq, he knew the country, its culture, and, most important, many of its leaders-a depth of knowledge that would prove invaluable in the months ahead.

We prepared a press announcement of Kay's departure and Duelfer's hiring. As is the form in such matters, I said some nice things about the individual departing. No matter what my personal feelings about him, the man had given up six months of his life to live in Baghdad, and he deserved our thanks. Our press office coordinated the statement with Kay, including a quote from him about there being many unresolved issues for the ISG to pursue.

In a final meeting in my office, with John McLaughlin present, Kay said that he was going to leave "quietly and like a gentleman." We invited him to stay on for the swearing in of Duelfer later that morning, but Kay said he had to go. Within forty-five minutes of leaving the CIA headquarters compound, Kay was being quoted by Reuters intelligence correspondent Toby Zakaria as saying that he concluded that "there were no Iraqi (WMD) stockpiles to be found." Although this later proved to be correct, it was quite a change from his comments just weeks before that it would take another six to nine months to know for sure. As for his promise to go quietly and allow his successor to finish the job, I can only say that I greatly regret the manner of Kay's departure.

Five days later, on January 28, 2004, Kay testified before the Senate Armed Services Committee, carried live on all the cable networks. He started by saying that "we were almost all wrong." Kay also inserted the familiar theme that the ISG needed more resources, ones that would be devoted entirely to the WMD hunt. Why we would need more resources to hunt for weapons that he had concluded were not present went unexplained.

Kay proceeded to describe the global significance of what he hadn't found in Iraq and even added some commentary about how wrong the United States had been about Libya and North Korea, two accounts about which he was not briefed and about which he was spectacularly misinformed. Kay ended his opening statement by saying,

And let me just conclude by my own personal tribute, both to the president and to George Tenet, for having the courage to select me to do this, and my successor, Charlie Duelfer, as well. Both of us are known for probably, at times, a regrettable streak of independence. I came not from within the administration, and it was clear-and clear in our discussions, and no one asked otherwise-that I would lead this the way I thought best, and I would speak the truth as we found it. I have had absolutely no pressure, prior, during the course of the work at the ISG or after I left, to do anything otherwise.

I mention the above quote not for the supposed "tribute" he gave me, but because several months later Kay would start telling people that he had concluded Iraq had no WMD before he left his post but had not been allowed to say what he thought. Apparently, this was part of an ongoing revision of his own recent performance because after he returned to the private sector, Kay stopped giving me any tributes altogether and became instead my long-distance psychoa.n.a.lyst.

In a widely reported interview taped for PBS's Frontline Frontline, Kay said that "George Tenet wanted to be a player...and if you didn't give the policy makers what they wanted...your views wouldn't be taken and you wouldn't be invited into the closed meetings." He concluded that I had "traded integrity for access, and that's a bad bargain any time in life. It's particularly a bad bargain if you're running an intelligence agency."

Ringing allegations. Great TV drama. And as wrong as any words can be. Never did I give policy makers information that I knew to be bad. We said what we said about WMD because we believed it.

In October 2004, I ran into David at a conference hosted by Ted Forstmann in Aspen, Colorado. Sir Richard Dearlove and I had appeared on a panel together moderated by Charlie Rose. One of the topics we discussed was how our respective intelligence communities had reached their judgments regarding Iraq and WMD. David approached me afterward and said, "You know, we are not much in disagreement on the substance." I looked at him and said, "There is one big difference: you have made this personal." Appearing on PBS, he had talked about my meetings and interactions with senior policy makers that he had never attended. He did not have a shred of evidence to back up his allegation.

Although Kay expressed the view that the WMD job was almost over, n.o.body in Baghdad believed it. He didn't deliver the evidence needed to make that case persuasively and in a definitive way that would put the issue to rest. It was never enough merely to cite Kay's opinion opinion that there were no WMD and that the job was done. Why? Because to close this chapter of history in a responsible way, we needed hard data, lots of it, organized and presented in a manner that would give future policy makers and historians confidence that we had gone about this thoroughly and professionally. We also wanted our own a.n.a.lysts to have the data necessary to understand what went wrong and what lessons should be drawn from it. that there were no WMD and that the job was done. Why? Because to close this chapter of history in a responsible way, we needed hard data, lots of it, organized and presented in a manner that would give future policy makers and historians confidence that we had gone about this thoroughly and professionally. We also wanted our own a.n.a.lysts to have the data necessary to understand what went wrong and what lessons should be drawn from it.

That is what Charles Duelfer delivered. When he arrived in Baghdad early in 2004, Duelfer installed himself at the ISG's airport headquarters, not downtown in the relative safety of the Green Zone, and then set about putting his own stamp on the WMD search. A number of the a.n.a.lysts had been working on draft chapters for a possible next report, but Duelfer put that effort on hold. He told the staff he didn't want to buy into any further interim conclusions unless he personally had had an opportunity to understand the underlying information. In particular, he wasn't going to make incremental decisions on important issues such as mobile biological weapons trailers.

We resumed with Duelfer the weekly secure video conferences that we had held with Kay. Duelfer would bring a varying set of partic.i.p.ants to the meetings on the Baghdad end, and he kept us carefully apprised of what he was and was not finding. It was apparent even via a long-distance video hookup that he was exercising hands-on leadership and restoring momentum to the effort.

As it turned out, Duelfer arrived in Baghdad at about the same time that I was making my second visit to that country, in February 2004. Shortly after arriving, I asked for an all-hands meeting of the ISG at the airport. I used the occasion to tell Duelfer's troops that although he was obviously more than a little crazy for jumping out of perfectly good airplanes, he was going to be a great leader.

I also wanted to let them know we appreciated their heroic work in what had become a very, very tough environment. I gave them a pep talk about the importance of their mission and how much they were appreciated. My remarks seemed well received at the time, but a couple of years later some of the foreigners present complained anonymously to the media that by ending my remarks with something like "Now, go out and find WMD," I was subtly suggesting that there was only one permissible result of their mission. That, of course, is nonsense. My guidance to Duelfer-just like my guidance to Kay-and to everyone in the ISG was simply to go out and find the truth.

Duelfer turned out to be a remarkably good choice for the job. He had a wealth of experience in Iraq and knew senior bureaucrats in almost every one of Saddam's key government ministries. In a large room at ISG headquarters they turned one entire wall, about twenty feet long, to a time line plotting anything to do with Iraq and WMD. The timeline covered the period from 1980 to 2003. At any point in that time span, they could draw a line down and say, this was Saddam's worldview at this point. The time line also gave context to the data and the interviews that the a.n.a.lysts were acc.u.mulating. On another wall a second time line plotted when Iraq made funds available for weapons programs. The ISG was thus able to track the relationship between funding and WMD activity. Duelfer was convinced that the answer to the questions "Did Saddam have WMD, and if not, why not?" would come not from doc.u.ments or scavenger hunts but from talking to the right people.

Getting to the right people was hard to do. The security situation in Iraq made the ISG's job nearly impossible. Large parts of the country were simply inaccessible for search without a huge military contingent to provide protection.

Because of the increasingly dangerous environment in Baghdad, to protect our personnel we purchased armored sedans wherever we could find them on the open market. One day an ISG team en route to a suspect site found themselves riding in an armored BMW that we had just had flown into Iraq. Originally intended for some European industrialist, the BMW came equipped with a DVD player in the backseat. One of the team members accidentally hit the DVD's Play b.u.t.ton, not knowing that there was a copy of the movie Saving Private Ryan Saving Private Ryan already in the machine, and the volume was on high. Seconds later the sound of gunfire and explosions came blasting through the car's speakers. It was the opening scene of the movie. For a few seconds, the vehicle's driver and security team thought the gunfire was live. While that might have been a humorous incident, most of the travel around Iraq was no joking matter. The threats were real and considerable. already in the machine, and the volume was on high. Seconds later the sound of gunfire and explosions came blasting through the car's speakers. It was the opening scene of the movie. For a few seconds, the vehicle's driver and security team thought the gunfire was live. While that might have been a humorous incident, most of the travel around Iraq was no joking matter. The threats were real and considerable.

On April 26, 2004, the ISG conducted a well-planned and well-rehea.r.s.ed inspection of an area of Baghdad known as the "Chemical Souk," looking for people and materials that might have been involved in chemical weapons production. Teams of armored military vehicles with .50-caliber gun turrets escorted the ISG team to the scene. Overhead a UAV provided surveillance video. The inspectors, wearing full body armor, which adds up to forty pounds to a person, inspected a building full of leaking barrels of mysterious chemicals. Suddenly, a huge explosion erupted, nearly trapping in the bas.e.m.e.nt an Australian scientist from the team. She narrowly escaped as the building collapsed above her. The fireball blew outward from the building to the rear, where soldiers were providing perimeter security. Two sergeants were killed, and five other soldiers were very badly burned.

On November 8, 2004, Charles Duelfer was traveling along the airport highway toward downtown Baghdad with three or four security vehicles. A civilian car loaded with explosives, known as a mobile improvised explosive device (IED), tried to insert itself in the middle of the convoy. Before it could get close enough, one of the security vehicles cut it off. The car detonated, killing two soldiers from the Kansas National Guard and seriously wounding another. Duelfer's car was severely damaged but he was unhurt. After he returned to the United States, Duelfer traveled to visit the families of the soldiers to thank them personally for their sacrifice. Throughout its existence, the ISG worked heroically to find the truth.

Duelfer told me much later that when he watched the Powell UN speech, he had the gut feeling that half of the information in it was wrong. "I just didn't know which half," he said. "With the Iraqis there was often some wacky, implausible, but true explanation for the way things seemed," he said.

From his subsequent conversations with Iraqis, Duelfer said that they were convinced that no matter what they did in the period prior to the war, it was not going to be good enough to satisfy us. Therefore, why try? Given our deep suspicions of Iraq, their track record of deception, and Saddam's desire to restart his weapons programs as soon as possible, Duelfer's contacts might have been right.

A number of the people Duelfer had known during his previous visits to Saddam-controlled Iraq were now in detention. So he spent a lot of time talking to these officials, trying to get to ground truth. He explained that any ability he had to influence the treatment the detainees received would go away when sovereignty was turned over to a new Iraqi government, around June 30. If they had useful information to share, now was the time to share it. Among those he talked to was Saddam himself.

According to Duelfer, "Saddam Husayn so dominated the Iraqi regime that its strategic intent was his alone. He wanted to end sanctions while preserving the capability to reconst.i.tute his weapons of ma.s.s destruction (WMD) when sanctions were lifted." Duelfer wrote that Saddam wanted WMD to deter Iran, in his view Iraq's princ.i.p.al enemy. The belief that he had such weapons would also, Saddam thought, deter hostile groups inside Iraq. Maintaining a calculated position of ambiguity on whether he had WMD was, in Saddam's view, essential to deterring these external and internal threats. The Oil-for-Food (OFF) program (a UN program that allowed Iraq to sell oil on the world market and use the proceeds for food and medicines but not to rebuild its military), and the a.s.sociated corruption, had terminally undermined the effect of sanctions on Iraq. Saddam believed he could simply wait out the sanctions and then begin re-creating Iraq's WMD capabilities.

In April of 2004, Duelfer met me in Amman, Jordan, where he asked me to support the release of an entirely decla.s.sified final report. I quickly agreed, viewing the report as a way to renew some faith in the intelligence community. I knew Charlie Duelfer would be thorough and fair and that he wouldn't pull any punches or spare anyone's feelings, including CIA's. In the end, that's just what happened. I had been gone from office for three months before Duelfer delivered his roughly thousand-page report to the new DCI, Porter Goss. As with Kay, Duelfer was given complete independence in putting the report together. He had the final word on what it said, but CIA did give a last-minute heads-up to a few key policy makers on what Duelfer had discovered about corruption in the UN Oil-for-Food program, because the doc.u.ments he uncovered would prove embarra.s.sing to several of their foreign counterparts.

In Duelfer's report, the ISG noted, as had just about everyone else, that Saddam had cheated consistently on United Nations sanctions, but on the critical issue that had been used as justification for the war, the report concluded that Saddam did not possess stockpiles of biological, chemical, and nuclear weapons at the time of the U.S.-led invasion of Iraq in March 2003, and that he had no active program to produce them. Asked in testimony by Senator Edward Kennedy what the chances were that WMD might still be uncovered, Duelfer replied, "The chance of finding a significant stockpile is less than five percent." That still sounds right to me.

Throughout this process, CIA and the intelligence community were committed to finding the truth and learning lessons from it. This is quite remarkable, and not very typical of what normally goes on in Washington. We oversaw a process that independently and unflinchingly drew unflattering conclusions about our work. Duelfer's report, produced solely under our guidance, was then used as the basis for many of the harsh judgments of the intelligence community rendered by the Silberman-Robb Commission. This willingness to look at itself critically is one of the strengths of the community I was privileged to lead and one of the few points of pride to come out of the whole WMD episode.

CHAPTER 23

Mission Not Accomplished

I first flew into Iraq just about the time Jerry Bremer took over as head of the Coalition Provisional Authority, or CPA, during the third week of May 2003. I took a helicopter ride with Jerry right over Baghdad. It was daylight. The helicopter door was wide open, and I was looking out as we flew. I remember thinking, as we scudded along, how precise the U.S. military action had been. There had been no ma.s.sive carpet bombing; whatever they intended to get, they'd hit. first flew into Iraq just about the time Jerry Bremer took over as head of the Coalition Provisional Authority, or CPA, during the third week of May 2003. I took a helicopter ride with Jerry right over Baghdad. It was daylight. The helicopter door was wide open, and I was looking out as we flew. I remember thinking, as we scudded along, how precise the U.S. military action had been. There had been no ma.s.sive carpet bombing; whatever they intended to get, they'd hit.

On the ground, the environment was strikingly permissive, considering that a foreign army had just invaded the capital and deposed the country's long-term dictator. People were going out, eating in restaurants. You half expected to see double-decker buses rolling down the main streets, with curious tourists gaping out the windows.

That same sense of optimism pervaded our station in Baghdad. Half the people there were young men and women who had just finished up their training. Mixed in with them were seasoned older pros and retired guys who had come back to work as contractors. I knew a lot of the veterans from odd spots all around the globe. Now they were in Baghdad, to help finish up the job of launching a new and democratic nation.

When I returned to Iraq in February 2004, the environment had changed dramatically. We flew into Baghdad at night, because you couldn't come in during the day. The C-17 bringing us there made a full-combat landing-a steep dive, quick on the ground. I was seated far forward, wearing flak jacket and helmet. There was no sightseeing this time. We flew into the Green Zone at treetop level and landed in the dark, on an unlit tarmac. I never felt in anything other than competent hands, but when you are flying black and wearing Kevlar, the pucker factor is hard to ignore.

By this time, CIA's presence in Iraq had grown quite large. Many of our officers showed up for a get-together that our senior man in Baghdad had arranged. Just about everyone arrived in body armor. I'd never seen so many stressed-out young people in one place in my life. I stayed three or four hours, talking with them. Then I was off again. I had to be somewhere else the next day, and in Baghdad in early 2004, you could fly out only at night.

In those intervening ten months, Iraq had become a very different place, but not at all in the way that the U.S. government had intended. How did it get that way? Through a series of decisions that, in retrospect, look like a slow-motion car crash.

In fact, the problems started well before the war. There was little planning before the invasion concerning the physical reconstruction that would follow. But regarding the political reconstruction of Iraq-how the country was to be administered and what role, if any, Iraqis would play in determining their political future-there was a great deal of spirited interagency discussion, often at the highest levels. Condi Rice and the vice president took an intense interest and often partic.i.p.ated directly. The usual deputy-and undersecretary-level officials represented their respective agencies. John McLaughlin and Bob Grenier, a senior CIA operations officer who was our "mission manager" for Iraq, split the duty from our side.

The debates generally broke down along familiar lines: State, CIA, and NSC favored a more inclusive and transparent approach, in which Iraqis representing the many tribes, sects, and interest groups in the country would be brought together to consult and put together some sort of rough const.i.tuent a.s.sembly that might then select an advisory council and a group of ministers to govern the country. No one advocated immediate introduction of Jeffersonian democracy, but many believed that the Iraqis should be encouraged to partic.i.p.ate in a process that would quickly help identify-and legitimize-genuine leaders of a future democratic Iraq.

The vice president and Pentagon civilians, however, advocated a very different approach. Rather than risking an open-ended political process that Americans could influence but not control, they wanted to be able to limit the Iraqis' power and handpick those Iraqis who would partic.i.p.ate. In practice, that meant Ahmed Chalabi and a handful of other well-known, longtime exiled oppositionists, along with the leaders of the essentially autonomous Kurdish areas. The differences in approach were clear and starkly articulated. The vice president himself summed up the dilemma: The choice, he said, was between "control and legitimacy." Doug Feith clearly stated his belief that it would not be necessary for the Iraqi exiles to legitimize themselves: "We can legitimize them," he said, through our economic a.s.sistance and the good governance the U.S. would provide. They never understood that, fundamentally, political control depends on the consent of the governed.

No consensus was ever reached, and no clear plan ever devised. In early January 2003, however, President Bush signed National Security Presidential Directive Number 24, giving the Department of Defense total and complete ownership of postwar Iraq. We didn't fully realize it at the time, but in the end, NSPD 24 would determine who made the final decisions on these momentous questions, and set the direction of the postwar reconstruction.

Hovering over this entire process was the figure-seldom acknowledged, almost never mentioned-of Ahmed Chalabi. Time and again, during the months leading up to the invasion and for months thereafter, the representatives of the vice president and Pentagon officials would introduce ideas that were thinly veiled efforts to put Chalabi in charge of post-invasion Iraq. Immediately before the invasion, the effort took the form of a proposal, put forward insistently and repeatedly, to form an Iraqi "government in exile," comprised of the exiles and the Kurdish leaders. These exiles would then be installed as a new government once Baghdad fell. My CIA colleagues were aghast. As Grenier later recalled, it was as though Defense and the vice president's staff wanted to invite comparison with the Soviet invasion of Afghanistan, when Russian troops deposed the existing government and installed Babrak Karmal, whom they had brought with them from Moscow.

At an NSC meeting about three months before the war got under way, President Bush asked Gen. Tommy Franks what he was going to do about security and law and order in the rear areas. Franks told the president, "It's all taken care of, sir. I have an American officer who will be lord mayor of every city, town and hamlet." That simply did not turn out to be the case. Whether that was part of CENTCOM's planning early on or not, I cannot say. In practice, though, the U.S. troop strength was sufficient to defeat the Iraqi army, but woefully inadequate to maintain the peace-just as Gen. Rick Shinseki, the former army chief of staff, had predicted.

Before the Iraq war began, an NSC staffer prepared an estimate of the troop strength necessary to stabilize postwar Iraq. The answer: 139,000 if the model was Afghanistan; more than 360,000 if the model was Bosnia; and a little shy of 500,000 if it was Kosovo. Which one was Iraq? Well, the war strategists erred on the side of Afghanistan when they went into Iraq, and we've been paying ever since.

The Pentagon's first man in charge of "postmajor conflict" Iraq was retired Lt. Gen. Jay Garner. Named to his position some months before the invasion, Garner was then sent forward to Kuwait to a.s.semble and prepare his team. When he and his team arrived in Iraq on April 18 to take responsibility for the newly created Office of Reconstruction and Humanitarian a.s.sistance (ORHA), it quickly became apparent that the task before Garner was monumental and the advance planning woefully insufficient. ORHA was set up in one of Saddam's abandoned palaces, but found itself without adequate communications, short of sufficient Arabic speakers, and lacking in contacts and understanding of the Iraqi people. Garner was a good man with an impossible mission. He had responsibility without authority, and a bad situation immediately got worse.

The CIA tried to help. They set up meetings with a cross section of important Iraqi technocrats-people who could help make the country work-and brought them together to meet with senior U.S. military. Right off the bat, however, they ran into difficulty. Did the groups we were a.s.sembling include members of the Ba'ath Party? they were asked. Of course they did. You couldn't advance in Saddam's Iraq without joining the Ba'ath Party. Just as the governments in newly democratic Eastern Europe would inevitably include former members of the Communist Party, any group of skilled bureaucrats in Baghdad would have to include people who once held Ba'ath Party membership. n.o.body questioned this initially, but the understanding that was obvious to us was less so to the new ORHA, a portent of far more serious problems to follow.

Similar problems arose when the United States started looking for candidates to populate an Iraqi provisional government. U.S. officials kept searching for, as one Agency officer put it, "Mohammed Jefferson," to launch Jeffersonian democracy in Iraq. The problem was that anyone who neatly fit that description would have long before been killed off by Saddam.

In the spring of 2003, Jay Garner, with NSC senior director Zal Khalilizad's a.s.sistance, began the process of holding regional conferences in Iraq in the hope of recognizing and taking advantage of different centers of power. According to CIA officers with him, Khalilizad believed that it was essential that Iraqis legitimize themselves. There were inherent risks in this. You can guide such a process, but you cannot control it. This was, after all, the essence of the democracy we had been preaching. It was important for the future stability of the country that Iraqis see people whom they recognized as having specific gravity being involved in the political process. This did not happen. The messy process of Iraqis legitimizing themselves came to a screeching halt. And Zal and Garner were out.

The a.s.sumption the U.S. government was working under was that this was going to be like the occupation of Germany, a supine country at our feet that we could remake in essentially whatever way we chose. The United States was going to completely demolish the Ba'ath Party. In the view of Paul Wolfowitz and others, you could replace "Ba'athist" with the word "n.a.z.i." It soon became clear to us and very clear to the Iraqis that the purpose of the U.S. invasion was fundamentally to remake their society.

In early May 2003, I got a call from Colin Powell asking what I knew about Jerry Bremer. "I don't really know him," I said. From what I'd heard, Bremer was a tough-minded former amba.s.sador who for a while had been head of the State Department's office of counterterrorism. "I certainly haven't heard anything bad about him."

Colin went on to say that the administration was considering Bremer as a replacement for Jay Garner. A few days later, on May 6, the White House made it official: Bremer had been selected to lead the effort to rebuild Iraq's infrastructure and help set up a new government. Although he was a presidential envoy, Bremer would report directly to the secretary of defense. His organization was given the t.i.tle Coalition Provisional Authority. Once CPA had been established, Condi Rice ordered the interagency committee that had been const.i.tuted to deal with postwar planning issues to fold its tent. It was only a short while later, however, that, as one White House official told me, "The s.h.i.t hit the fan and we had to rely on the British to tell us what was going on because we were getting no political reporting out of CPA." Rice then ordered the NSC process to start up again. But by then, fundamental decisions on disbanding the army and de-Ba'athification had already been made. The early returns filtering back to me on CPA indicated that it was not running smoothly.

The news was disquieting. It was just as worrisome that CPA was not being staffed with people with the requisite skills to enable our success. Many possessed the right political credentials but were unschooled in the complicated ways of the Middle East. What Iraq needed were Arabists and Foreign Service officers who understood the country's tribal allegiances, or who at least knew a Sunni from a Shia. What CPA seemed to be getting were people anxious to set up a Baghdad stock exchange, try out a flat-tax system, and impose other elements of a lab-school democratic-capitalist social structure. One of my officers returned from a trip to Iraq a month or two after CPA had taken over and told me, "Boss, that place runs like a graduate school seminar, none of them speaks Arabic, almost n.o.body's ever been to an Arab country, and no one makes a decision but Bremer."

The State Department had earlier a.s.sembled a team of experts to plan for a postwar Iraq, and Rich Armitage had 737s all lined up to fly them and their computers and some eighty Arabic linguists with regional knowledge out to Baghdad to begin setting up an emba.s.sy-in-waiting. The Pentagon, though, had other plans, and they certainly didn't include the Department of State, which many in Rumsfeld's circle thought had performed poorly in Afghanistan. Time and again, Marc Grossman, the undersecretary of state for political affairs, would raise the matter with Doug Feith, and time and again, Feith would say he was going to look into it. Before long it became apparent that, from the Pentagon's point of view, the State Department team of experts could sit on the runway at Dulles or Andrews Air Force Base, waiting for a lift to Baghdad, until h.e.l.l froze over.

The security situation in Iraq started heading south remarkably soon after Saddam's statues fell. A reasonable question is: Did the U.S. intelligence community fail to predict the possibility of civil strife? Did we buy into the notion that Americans would be "greeted as liberators"? The answer, as so often is the case, is not black or white.

Although CIA was not among those who confidently expected Coalition forces to be greeted as liberators, we did expect the Shia in the south, long oppressed by Saddam, to open their arms to anyone who removed him. And, initially, Coalition troops were well received in the south.

Our expectation, though, wasn't open-ended, and it wasn't blind to other possibilities. Simultaneously, we produced a doc.u.ment that we t.i.tled, prophetically as things turned out, "The Consequences of Catastrophic Success." Our a.n.a.lysis said that there would be a feeling of relief among the Iraqi people that Saddam was gone but that this would last for only a short time before old rivalries and ancient ethnic tensions resurfaced. During this critical period, we needed to demonstrate an ability to provide the services that a country demands-food, water, electricity, jobs-while creating also a sense of safety and security that was absent under Saddam.

That, to me, is where plans went awry. Our a.n.a.lysis a.s.sumed there was a plan for ensuring the peace. In fact, there was no strategy for when U.S. forces. .h.i.t the ground. This playbook wasn't written until long after kickoff.

In a January 2003 CIA paper, we said:

Iraq would be unlikely to split apart, but a post-Saddam authority would face a deeply divided society with a significant chance that domestic groups would engage in violent conflict with each other unless an occupying force prevented them from doing so. Rogue ex-regime elements could forge an alliance with existing terrorist organizations or act independently to wage guerilla warfare against the new government. In the early months after the forceful ouster of Saddam, stability in Iraq would depend partly on the perspectives of Iraqis towards whatever interim authority, military or civilian, foreign or indigenous was in control, as well as the ability of the authority to perform the administrative and security tasks of governing the country. The top priorities of most Iraqis would be to obtain peace, order, stability and such basic needs as food and shelter.... US-led defeat and occupation of Arab Iraq probably would boost proponents of political Islam. Calls by Islamists for the people of the region to unite probably would resonate widely. Fear of US domination and a widespread belief probably would attract many angry young recruits to extremists' ranks.

The same paper said, "Iraq's history of foreign occupation, first the Ottomans then the British, has left Iraqis with a deep dislike of occupiers. An indefinite military occupation with ultimate power in the hands of a non-Iraqi officer would be widely unacceptable. Iraqi military officers who oppose Saddam find the idea of a Western power conquering and governing Iraq anathema and a motivation to fight with Saddam where they otherwise would not."

In another paper we cautioned that the demobilization process would be full of pitfalls and suggested that "Baghdad's immediate post-war security needs may require that demobilization be delayed until Iraq [is] ready to begin building the armed forces."

We warned that, "Regardless of US postwar policy for Iraq, Iraqis would become alienated if not persuaded that their national and religious sensitivities, particularly their desire for self governance were part of the foundation for reconstruction. Iraqis would likely resort to obstruction, resistance and armed opposition if they perceive attempts to keep them dependent on the US and the West."

A National Intelligence Council paper in January 2003 t.i.tled "Can Iraq Ever Become a Democracy?" said that "Iraqi political culture is so imbued with norms alien to the democratic experience...that it may resist the most vigorous and prolonged democratic treatments."

In March 2003 we warned that "Iraqi patience with an extended US presence after an overwhelming victory would be short," and said that "humanitarian conditions in many parts of Iraq could rapidly deteriorate in a matter of days, and many Iraqis would probably not understand that the Coalition wartime logistic pipeline would require time to reorient its mission to humanitarian aid."

Our prewar a.n.a.lysis of postwar Iraq was prescient. The challenge for CIA a.n.a.lysts was not so much in predicting what the Iraqis would do. Where we ran into trouble was in our inability to foresee some of the actions of our own own government. If you don't know the game plan, it is tough to do good a.n.a.lysis. As a result, did we exactly predict everything that would unfold? No. government. If you don't know the game plan, it is tough to do good a.n.a.lysis. As a result, did we exactly predict everything that would unfold? No.

Bremer would later write that three days after the White House announced his appointment, and shortly before going to Baghdad, he met with Doug Feith in the Pentagon. Feith, he says, urged him to issue an order as soon as possible upon arriving in Iraq that would prevent former Ba'ath Party members from having a role in the new government. Bremer did just that, on May 16, just four days after landing in Iraq. That morning's New York Times New York Times carried a hint of what was to come: "Shortly I will issue an order on measures to extirpate Baathists and Baathism in Iraq forever," Bremer was quoted as saying. "We have and will aggressively move to seek to identify these people and remove them from office." carried a hint of what was to come: "Shortly I will issue an order on measures to extirpate Baathists and Baathism in Iraq forever," Bremer was quoted as saying. "We have and will aggressively move to seek to identify these people and remove them from office."

Just a few weeks before the war started, senior U.S. officials were saying publicly that the conflict might be avoided if Saddam and a few dozen of his top henchmen simply left. This concept was never embedded in our war goals. Now, the war having been waged, the United States apparently was saying that thousands of officials around the country would be aggressively removed.

Bremer writes in his memoir that the intelligence community estimated that this order would affect only about 1 percent of the Iraqi population. That could be taken to imply that we supported the move and thought it was a good idea, but that was definitely not the case. In fact, we knew nothing about it until de-Ba'athification was a fait accompli. Clearly, this was a critical policy decision, yet there was no NSC Princ.i.p.als meeting to debate the move. As for the 1 percent number Bremer cites, he didn't ask for that estimate until the day after he issued the order, and once he got it he ignored the twofold context: first, that many of those Ba'athists were technocrats of exactly the sort Iraq would soon need if it were to again resume responsibility for its governance, and, second, that every Ba'athist "extirpated" from Iraq, to use Bremer's word, had brothers and sisters and aunts, uncles, and cousins with whom to share his anger.

Privately, in fact, the senior CIA officer in Iraq and others strongly advised against this step when they were finally informed of it, and they continued to argue after the decision was made. A senior NSC staffer told me that when he briefed the president on de-Ba'athification, the staffer talked about South Africa's Truth and Reconciliation program. Just as South Africans had done, Iraqis themselves should determine who had too much blood on his or her hands to be permitted to take part in a new government. Bremer's plan put the process in the hands of an Iraqi all right. Ahmed Chalabi was named to head the de-Ba'athification Council, and as a result the implementation of the order was even more draconian.

We soon began hearing stories about how Iraqis could not send their kids to school because all the teachers had been dismissed for being members of the Ba'ath Party. In the context of a country armed to the teeth, this was not a good thing. If the kids and teachers were not in school, they were on the streets. I went to see Condi Rice and complained that the indiscriminate nature of the de-Ba'athification order had swept away not just Saddam's thugs but also, for example, something like forty thousand schoolteachers, who had joined the Ba'ath Party simply to keep their jobs. This order wasn't protecting Iraqis; it was destroying what little inst.i.tutional foundations were left in the country. The net effect was to persuade many ex-Ba'athists to join the insurgency. Condi said she was very frustrated by the situation, but nothing ever happened. Several months later, with a full-blown insurgency under way, an interagency group headed by Deputy National Security Advisor Bob Blackwill desperately looked for ways to reach out to dissident Sunni Arabs. We again raised the subject of rolling back the de-Ba'athification order. Doug Feith retorted that doing so would "undermine the entire moral justification for the war."

Bremer's de-Ba'athification order became known as CPA Proclamation Number One. As bad as that was, CPA Proclamation Number Two was worse. Again, without any formal discussion or debate back in Washington-at least any that included me or my top deputies-Bremer, on May 23, ordered the dissolution of the Iraqi army.

To be sure, elements of the Iraqi army, especially the Special Republican Guards (SRG) and the Special Security Organization (SSO), did have much blood on their hands. However, we viewed many Iraqi military officers as professionals, driven by national Iraqi values rather than loyalty to Saddam, who could form the core of a new Iraqi military, but the order struck a broad blow at the Sunnis, who comprise 20 percent of the national population and who occupied virtually all of the top ranks in the army. Granted, they were never going to be completely satisfied, short of having Iraq handed back to their control, but along with the de-Ba'athification order, this second order had effectively alienated one fifth of the population and much of the center of the country.

NSC officials were expecting Proclamation Number Two to include some language about how Iraqi military members below the rank of lieutenant colonel could apply for reinstatement. After all, the majority of army members were conscripts just trying to feed their families. CPA Proclamation Number Two appeared to be punishing them-and even the Shia who made up the bottom rung of the military-equally with those who had ruled the roost. When the p.r.o.nouncement was issued, however, that provision was not mentioned. So, as far as the rank-and-file members were concerned, Bremer had just announced that they were all unemployed.

Jay Garner, who was still in Iraq at the time, went to see Bremer along with our senior CIA officer in the country. They both told him that the demobilization order was madness. Garner had been counting on using some of the former Iraqi military for stabilization and security. Our officer told Bremer that the action would only "give oxygen to the rejectionists."

The argument from some supporters of CPA Proclamation Two was that the army had essentially dissolved itself anyway, so what was the big deal? Our officer on the ground at the time, however, estimated that the majority of the army could have been recalled within a two-week period and put to useful work.

Bremer was unmoved. He reportedly told Garner that he could raise the issue with the secretary of defense if he wanted to, but that this was a done deal and a decision made at a level "above Rumsfeld's pay grade."

Whoever had made the decision, the reaction from former Iraqi army members was swift. A New York Times New York Times report on a May 25 demonstration in Basra by dismissed Iraqi soldiers quoted one former Iraqi tank driver as saying, "The U.S. planes dropped the papers telling us to stay in our homes...They said our families would be fine," he said. More ominously, a lieutenant colonel told the reporter, "We have guns at home. If they don't pay us, if they make our children suffer, they'll hear from us." report on a May 25 demonstration in Basra by dismissed Iraqi soldiers quoted one former Iraqi tank driver as saying, "The U.S. planes dropped the papers telling us to stay in our homes...They said our families would be fine," he said. More ominously, a lieutenant colonel told the reporter, "We have guns at home. If they don't pay us, if they make our children suffer, they'll hear from us."

Eventually some army members were paid and allowed to apply to rejoin the new Iraqi army, but all officers with ranks of lieutenant colonel and above were permanently banned-despite the fact that, like many non-Western armies, Iraq had a disproportionate number of army members with high ranks. A typical Iraqi lieutenant colonel did not have the same level of authority or influence wielded by his U.S. Army counterpart.

At meetings in the White House and in Baghdad after the two proclamations were issued, we argued that the orders were having unintended negative consequences. The actions had taken large numbers of common Iraqis and given them few prospects beyond being paupers, criminals, or insurgents. One of our senior officers tallied the numbers, including affected family members and the like, and came up with a pool of a hundred thousand Iraqis who had been driven toward the brink by the de-Ba'athification order alone. In the end, too many of them chose insurgency.

For some officials in the Pentagon, the accelerating violence simply proved the wisdom of excluding these Ba'athists and exarmy members from the future of Iraq. As late as the spring of 2004, at a meeting in the White House, one of our officers was asked for "out-of-the-box" ideas to stem the violence. He suggested rescinding CPA Proclamation Two and mounting an aggressive campaign to round up former army members and enlist them to help secure Iraq's borders and maintain internal security. As later described to me, a U.S. Army colonel present, who had been DIA's liaison to Ahmed Chalabi and the Iraqi National Congress, said, "I agree. We should round them all up and shoot them."

The moves the U.S. government was making were driving a wedge between the various factions in Iraq. Charles Duelfer was told by an Iraqi friend that, in the past, Iraqis were not accustomed to thinking of themselves primarily as Shia or Sunni. But the way we implemented democracy had led people to believe that they deserved a piece of the pie based on their membership in a certain group. So the whole dynamic was to pull away from the center. The decisions we made tended to fracture Iraq, not to bring it together.

On one of his trips to Iraq, Wolfowitz told our senior man there, "You don't understand the policy of the U.S. government, and if you don't understand the policy, you are hardly in a position to collect the intelligence to help that policy succeed." It was an arrogant statement that masked a larger reality. In many cases we were not not aware of what our own government was trying to do. The one thing we were certain of was that our warnings were falling on deaf ears. aware of what our own government was trying to do. The one thing we were certain of was that our warnings were falling on deaf ears.

In the midst of all this, we started pushing for the establishment of a new Iraqi intelligence service. Any government intent on protecting people needs an organization to acquire information regarding internal security and external threats. That much seems obvious, but we ran into strong and immediate resistance to our suggestions on building such a service.

John McLaughlin tried to get authorization through the Deputies Committee to help set up such a capability, only to be thwarted. In all the years that I have known John, I don't think I have ever seen him more exasperated. "The only country in the world where the U.S. intelligence community doesn't have a counterpart is Iraq," he remembers saying at one of the deputies meetings. "The best way to get a handle on who is causing the violence in Iraq is to have Iraqis figure it out." That message, too, never seemed to get heard.

On another occasion, Steve Kappes, the then second-ranking operations officer at CIA, was pushing the same theme at a meeting where Condi Rice was present. "How do I know you guys aren't going to create another KGB?" Condi asked. "We didn't create the first one," Steve reminded her. Condi's comment was emblematic of the mind-set we were up against. Policy makers didn't seem to want us dealing with anyone who wasn't "politically acceptable" to them on some firm but unannounced scale. Our point was that Americans were dying, jihadists were running all over the country, and it was time to figure out how to vet Iraqis who had the capabilities to do something about it.

We'd been through this before. When the Soviet Union fell and the West inherited Eastern Europe, we set about building intelligence services there out of what was already on hand to work with. Was there a high probability that Soviet agents still peopled those services? Sure. Is there a high probability that over the course of time, they'll be weeded out? Sure, again. The point was, you have to take some risk if you want to make the government work.