Blind Into Baghdad

On a Friday afternoon last November, I met Douglas Feith in his office at the Pentagon to discuss what has happened in Iraq. Feith's title is undersecretary of defense for policy, which places him, along with several other undersecretaries, just below Secretary of Defense Donald Rumsfeld and Deputy Secretary Paul Wolfowitz in the Pentagon's hierarchy. Informally he is seen in Washington as "Wolfowitz's Wolfowitz"—that is, as a deputy who has a wide range of responsibilities but is clearly identified with one particular policy. That policy is bringing regime change to Iraq—a goal that both Wolfowitz and Feith strongly advocated through the 1990s. To opponents of the war in Iraq, Feith is one of several shadowy, Rasputinlike figures who are shaping U.S. policy. He is seen much the way enemies of the Clinton Administration saw Hillary Clinton. Others associated with the Bush Administration who are seen this way include the consultant Richard Perle; Lewis "Scooter" Libby, the chief of staff for Vice President Dick Cheney; and the Vice President himself. What these officials have in common is their presumably great private influence and—even in the case of the Vice President—their limited public visibility and accountability.

In person Douglas Feith is nothing like Rasputin. Between a Reagan-era stint in the Pentagon and his current job he was a Washington lawyer for fifteen years, and he answered my questions with a lawyer's affability in the face of presumed disagreement. I could be biased in Feith's favor, because he was the most senior Administration official who granted my request for an interview about postwar Iraq. Like Donald Rumsfeld, Feith acts and sounds younger than many others of his age (fifty). But distinctly unlike Rumsfeld at a press conference, Feith in this interview did not seem at all arrogant or testy. His replies were relatively candid and unforced, in contrast to the angry or relentlessly on-message responses that have become standard from senior Administration officials. He acknowledged what was "becoming the conventional wisdom" about the Administration's failure to plan adequately for events after the fall of Baghdad, and then explained—with animation, dramatic pauses, and gestures—why he thought it was wrong.

Feith offered a number of specific illustrations of what he considered underappreciated successes. Some were familiar —the oil wells weren't on fire, Iraqis didn't starve or flee—but others were less so. For instance, he described the Administration's careful effort to replace old Iraqi dinars, which carried Saddam Hussein's image ("It's interesting how important that is, and it ties into the whole issue of whether people think that Saddam might be coming back"), with a new form of currency, without causing a run on the currency.

But mainly he challenged the premise of most critics: that the Administration could have done a better job of preparing for the consequences of victory. When I asked what had gone better than expected, and what had gone worse, he said, "We don't exactly deal in 'expectations.' Expectations are too close to 'predictions.' We're not comfortable with predictions. It is one of the big strategic premises of the work that we do."

The limits of future knowledge, Feith said, were of special importance to Rumsfeld, "who is death to predictions." "His big strategic theme is uncertainty," Feith said. "The need to deal strategically with uncertainty. The inability to predict the future. The limits on our knowledge and the limits on our intelligence."

In practice, Feith said, this meant being ready for whatever proved to be the situation in postwar Iraq. "You will not find a single piece of paper ... If anybody ever went through all of our records—and someday some people will, presumably—nobody will find a single piece of paper that says, 'Mr. Secretary or Mr. President, let us tell you what postwar Iraq is going to look like, and here is what we need plans for.' If you tried that, you would get thrown out of Rumsfeld's office so fast—if you ever went in there and said, 'Let me tell you what something's going to look like in the future,' you wouldn't get to your next sentence!"

"This is an important point," he said, "because of this issue of What did we believe? ... The common line is, nobody planned for security because Ahmed Chalabi told us that everything was going to be swell." Chalabi, the exiled leader of the Iraqi National Congress, has often been blamed for making rosy predictions about the ease of governing postwar Iraq. "So we predicted that everything was going to be swell, and we didn't plan for things not being swell." Here Feith paused for a few seconds, raised his hands with both palms up, and put on a "Can you believe it?" expression. "I mean—one would really have to be a simpleton. And whatever people think of me, how can anybody think that Don Rumsfeld is that dumb? He's so evidently not that dumb, that how can people write things like that?" He sounded amazed rather than angry.

No one contends that Donald Rumsfeld, or Paul Wolfowitz, or Douglas Feith, or the Administration as a whole is dumb. The wisdom of their preparations for the aftermath of military victory in Iraq is the question. Feith's argument was a less defensive-sounding version of the Administration's general response to criticisms of its postwar policy: Life is uncertain, especially when the lid comes off a long-tyrannized society. American planners did about as well as anyone could in preparing for the unforeseeable. Anyone who says otherwise is indulging in lazy, unfair second-guessing. "The notion that there was a memo that was once written, that if we had only listened to that memo, all would be well in Iraq, is so preposterous," Feith told me.

The notion of a single memo's changing history is indeed farfetched. The idea that a substantial body of knowledge could have improved postwar prospects is not. The Administration could not have known everything about what it would find in Iraq. But it could have—and should have—done far more than it did.

Almost everything, good and bad, that has happened in Iraq since the fall of Saddam Hussein's regime was the subject of extensive pre-war discussion and analysis. This is particularly true of what have proved to be the harshest realities for the United States since the fall of Baghdad: that occupying the country is much more difficult than conquering it; that a breakdown in public order can jeopardize every other goal; that the ambition of patiently nurturing a new democracy is at odds with the desire to turn control over to the Iraqis quickly and get U.S. troops out; that the Sunni center of the country is the main security problem; that with each passing day Americans risk being seen less as liberators and more as occupiers, and targets.

All this, and much more, was laid out in detail and in writing long before the U.S. government made the final decision to attack. Even now the collective efforts at planning by the CIA, the State Department, the Army and the Marine Corps, the United States Agency for International Development, and a wide variety of other groups inside and outside the government are underappreciated by the public. The one pre-war effort that has received substantial recent attention, the State Department's Future of Iraq project, produced thousands of pages of findings, barely one paragraph of which has until now been quoted in the press. The Administration will be admired in retrospect for how much knowledge it created about the challenge it was taking on. U.S. government predictions about postwar Iraq's problems have proved as accurate as the assessments of pre-war Iraq's strategic threat have proved flawed.

But the Administration will be condemned for what it did with what was known. The problems the United States has encountered are precisely the ones its own expert agencies warned against. Exactly what went wrong with the occupation will be studied for years—or should be. The missteps of the first half year in Iraq are as significant as other classic and carefully examined failures in foreign policy, including John Kennedy's handling of the Bay of Pigs invasion, in 1961, and Lyndon Johnson's decision to escalate U.S. involvement in Vietnam, in 1965. The United States withstood those previous failures, and it will withstand this one. Having taken over Iraq and captured Saddam Hussein, it has no moral or practical choice other than to see out the occupation and to help rebuild and democratize the country. But its missteps have come at a heavy cost. And the ongoing financial, diplomatic, and human cost of the Iraq occupation is the more grievous in light of advance warnings the government had.

Concern about Saddam Hussein pre-dated the 9/11 attacks and even the inauguration of George W. Bush. In 1998 Congress passed and President Bill Clinton signed the Iraq Liberation Act, which declared that "it should be the policy of the United States to support efforts to remove the regime headed by Saddam Hussein from power." During the 2000 presidential campaign Al Gore promised to support groups working to unseat Saddam Hussein. In the week before Bush took office, Nicholas Lemann reported in The New Yorker that "the idea of overthrowing Saddam is not an idle fantasy—or, if it is, it's one that has lately occupied the minds of many American officials, including people close to George W. Bush." But the intellectual case for regime change, argued during the Clinton years by some Democrats and notably by Paul Wolfowitz, then the dean of the Johns Hopkins School of Advanced International Studies, shifted clearly toward operational planning after the destruction of the World Trade Center.

For much of the public this case for war against Iraq rested on an assumed connection (though this was never demonstrated, and was officially disavowed by the President) between Saddam Hussein's regime and the terrorist hijackers. Within the government the case was equally compelling but different. September 11 had shown that the United States was newly vulnerable; to protect itself it had to fight terrorists at their source; and because Saddam Hussein's regime was the leading potential source of future "state-sponsored" terrorism, it had become an active threat, whether or not it played any role in 9/11. The very next day, September 12, 2001, James Woolsey, who had been Clinton's first CIA director, told me that no matter who proved to be responsible for this attack, the solution had to include removing Saddam Hussein, because he was so likely to be involved next time. A military planner inside the Pentagon later told me that on September 13 his group was asked to draw up scenarios for an assault on Iraq, not just Afghanistan.

Soon after becoming the Army Chief of Staff, in 1999, General Eric Shinseki had begun ordering war-game exercises to judge strategies and manpower needs for possible combat in Iraq. This was not because he assumed a war was imminent. He thought that the greater Caspian Sea region, including Iraq, would present a uniquely difficult challenge for U.S. troops, because of its geography and political tensions. After 9/11, Army war games involving Iraq began in earnest.

In his first State of the Union address, on January 29, 2002, President Bush said that Iraq, Iran, and North Korea were an "axis of evil" that threatened world peace. "By seeking weapons of mass destruction, these regimes pose a grave and growing danger. They could provide these arms to terrorists, giving them the means to match their hatred. They could attack our allies or attempt to blackmail the United States."

By the time of this speech efforts were afoot not simply to remove Saddam Hussein but also to imagine what Iraq would be like when he was gone. In late October of 2001, while the U.S. military was conducting its rout of the Taliban from Afghanistan, the State Department had quietly begun its planning for the aftermath of a "transition" in Iraq. At about the time of the "axis of evil" speech, working groups within the department were putting together a list of postwar jobs and topics to be considered, and possible groups of experts to work on them.

Thus was born the Future of Iraq project, whose existence is by now well known, but whose findings and potential impact have rarely been reported and examined. The State Department first publicly mentioned the project in March of 2002, when it quietly announced the lineup of the working groups. At the time, media attention was overwhelmingly directed toward Afghanistan, where Operation Anaconda, the half-successful effort to kill or capture al-Qaeda and Taliban fighters, was under way.

For several months before announcing the project the State Department had been attempting to coordinate the efforts of the many fractious Iraqi exile organizations. The Future of Iraq project held the potential for harnessing, and perhaps even harmonizing, the expertise available from the exile groups.

It was also in keeping with a surprisingly well established U.S. government tradition of preparing for postwar duties before there was a clear idea of when fighting would begin, let alone when it would end. Before the United States entered World War II, teams at the Army War College were studying what went right and wrong when American doughboys occupied Germany after World War I. Within months of the attack on Pearl Harbor a School of Military Government had been created, at the University of Virginia, to plan for the occupation of both Germany and Japan. In 1995, while U.S. negotiators, led by Richard Holbrooke, were still working at the Dayton peace talks to end the war in the Balkans, World Bank representatives were on hand to arrange loans for the new regimes.

Contemplating postwar plans posed a problem for those who, like many in the State Department, were skeptical of the need for war. Were they making a war more likely if they prepared for its aftermath? Thomas Warrick, the State Department official who directed the Future of Iraq project, was considered to be in the antiwar camp. But according to associates, he explained the importance of preparing for war by saying, "I'm nervous that they're actually going to do it—and the day after they'll turn to us and ask, 'Now what?'" So he pushed ahead with the project, setting up numerous conferences and drafting sessions that would bring together teams of exiles—among them Kanan Makiya, the author of the influential anti-Saddam book Republic of Fear, first published in 1989. A small number of "international advisers," mainly from the United States, were also assigned to the teams. Eventually there would be seventeen working groups, designed systematically to cover what would be needed to rebuild the political and economic infrastructure of the country. "Democratic Principles and Procedures" was the name of one of the groups, which was assigned to suggest the legal framework for a new government; Makiya would write much of its report. The "Transitional Justice" group was supposed to work on reparations, amnesty, and de-Baathification laws. Groups studying economic matters included "Public Finance," "Oil and Energy," and "Water, Agriculture and Environment."

In May of 2002 Congress authorized $5 million to fund the project's studies. In the flurry of news from Afghanistan the project went unnoticed in the press until June, when the State Department announced that the first meetings would take place in July. "The role of the U.S. government and State Department is to see what the Iraqis and Iraqi-Americans want," Warrick said at a conference on June 1, 2002. "The impetus for change comes from [Iraqis], not us. This is the job of Iraqis inside and outside."

That same day President Bush delivered a graduation speech at West Point, giving a first look at the doctrine of pre-emptive war. He told the cadets, to cheers, "Our security will require all Americans to be forward-looking and resolute, to be ready for pre-emptive action when necessary to defend our liberty and to defend our lives." Later in the summer the doctrine was elaborated in a new National Security Strategy, which explained that since "rogue states" could not be contained or deterred, they needed to be destroyed before they could attack.

Whenever National Security Adviser Condoleezza Rice was interviewed that summer, she talked mainly about the thinking behind the new policy. When Vice President Dick Cheney was interviewed, he talked mainly about Saddam Hussein's defiance of international law. But when Secretary of State Colin Powell was interviewed, he constantly stressed the value of an international approach to the problem and the need to give UN arms inspectors adequate time to do their job.

War with Iraq was not inevitable at this point, but it seemed more and more likely. Daily conversation in Washington, which usually reverts to "So, who do you think will be the next President?," switched instead to "So, when do you think we're going to war?"

It was in these circumstances that the Future of Iraq project's working groups deliberated. Most of the meetings were in Washington. Some were in London, and one session, in early September, took place in Surrey, where representatives of a dozen mutually suspicious exile groups discussed prospects for democratic coexistence when Saddam Hussein was gone. (Along with Chalabi's INC the meeting included several rival Kurdish groups, Assyrian and Turkomen organizations, the Iraqi Constitutional Monarchy Movement, and others.)

The project did not overcome all the tensions among its members, and the results of its deliberations were uneven. Three of its intended working groups never actually met—including, ominously, "Preserving Iraq's Cultural Heritage." The "Education" group finally produced a report only six pages long, in contrast to many hundreds of pages from most others. Some recommendations were quirky or reflected the tastes of the individual participants who drafted them. A report titled "Free Media" proposed that all Iraqi journalists be taken out of the country for a month-long re-education process: "Those who 'get it' go back as reporters; others would be retired or reassigned." A group that was considering ways of informing Iraq about the realities of democracy mentioned Baywatch and Leave It to Beaver as information sources that had given Iraqis an imprecise understanding of American society. It recommended that a new film, Colonial America: Life in a Theocracy, be shot, noting, "The Puritan experiments provide amazing parallels with current Moslem fundamentalism. The ultimate failures of these US experiments can also be vividly illustrated—witch trials, intolerance, etc."

But whatever may have been unrealistic or factional about these efforts, even more of what the project created was impressive. The final report consisted of thirteen volumes of recommendations on specific topics, plus a one-volume summary and overview. These I have read—and I read them several months into the occupation, when it was unfairly easy to judge how well the forecast was standing up. (Several hundred of the 2,500 pages were in Arabic, which sped up the reading process.) The report was labeled "For Official Use Only"—an administrative term that implies confidentiality but has no legal significance. The State Department held the report closely until, last fall, it agreed to congressional requests to turn over the findings.

Most of the project's judgments look good in retrospect—and virtually all reveal a touching earnestness about working out the details of reconstructing a society. For instance, one of the thickest volumes considered the corruption endemic in Iraqi life and laid out strategies for coping with it. (These included a new "Iraqi Government Code of Ethics," which began, "Honesty, integrity, and fairness are the fundamental values for the people of Iraq.") The overview volume, which appears to have been composed as a series of PowerPoint charts, said that the United States was undertaking this effort because, among other things, "detailed public planning" conveys U.S. government "seriousness" and the message that the U.S. government "wants to learn from past regime change experiences."

For their part, the Iraqi participants emphasized several points that ran through all the working groups' reports. A recurring theme was the urgency of restoring electricity and water supplies as soon as possible after regime change. The first item in the list of recommendations from the "Water, Agriculture and Environment" group read, "Fundamental importance of clean water supplies for Iraqis immediately after transition. Key to coalition/community relations." One of the groups making economic recommendations wrote, "Stressed importance of getting electrical grid up and running immediately—key to water systems, jobs. Could go a long way to determining Iraqis' attitudes toward Coalition forces."

A second theme was the need to plan carefully for the handling and demobilization of Iraq's very sizable military. On the one hand, a functioning army would be necessary for public order and, once coalition forces withdrew, for the country's defense. ("Our vision of the future is to build a democratic civil society. In order to make this vision a reality, we need to have an army that can work alongside this new society.") On the other hand, a large number of Saddam's henchmen would have to be removed. The trick would be to get rid of the leaders without needlessly alienating the ordinary troops—or leaving them without income. One group wrote, "All combatants who are included in the demobilization process must be assured by their leaders and the new government of their legal rights and that new prospects for work and education will be provided by the new system." Toward this end it laid out a series of steps the occupation authorities should take in the "disarmament, demobilization, and reintegration" process. Another group, in a paper on democratic principles, warned, "The decommissioning of hundreds of thousands of trained military personnel that [a rapid purge] implies could create social problems."

Next the working groups emphasized how disorderly Iraq would be soon after liberation, and how difficult it would be to get the country on the path to democracy—though that was where it had to go. "The removal of Saddam's regime will provide a power vacuum and create popular anxieties about the viability of all Iraqi institutions," a paper on rebuilding civil society said. "The traumatic and disruptive events attendant to the regime change will affect all Iraqis, both Saddam's conspirators and the general populace." Another report warned more explicitly that "the period immediately after regime change might offer these criminals the opportunity to engage in acts of killing, plunder and looting." In the short term the occupying forces would have to prevent disorder. In the long term, according to a report written by Kanan Makiya, they would need to recognize that "the extent of the Iraqi totalitarian state, its absolute power and control exercised from Baghdad, not to mention the terror used to enforce compliance, cannot be overestimated in their impact on the Iraqi psyche and the attendant feeling of fear, weakness, and shame." Makiya continued, "These conditions and circumstances do not provide a strong foundation on which to build new institutions and a modern nation state."

Each of the preceding themes would seem to imply a long, difficult U.S. commitment in Iraq. America should view its involvement in Iraq, the summary report said, not as it had Afghanistan, which was left to stew in lightly supervised warlordism, but as it had Germany and Japan, which were rebuilt over many years. But nearly every working group stressed one other point: the military occupation itself had to be brief. "Note: Military government idea did not go down well," one chart in the summary volume said. The "Oil and Energy" group presented a "key concept": "Iraqis do not work for American contractors; Americans are seen assisting Iraqis."

Americans are often irritated by the illogic of "resentful dependence" by weaker states. South Koreans, for example, complain bitterly about U.S. soldiers in their country but would complain all the more bitterly if the soldiers were removed. The authors of the Future of Iraq report could by those standards also be accused of illogical thinking, in wanting U.S. support but not wanting U.S. control. Moreover, many of the project's members had a bias that prefigured an important source of postwar tension: they were exiles who considered themselves the likeliest beneficiaries if the United States transferred power to Iraqis quickly—even though, precisely because of their exile, they had no obvious base of support within Iraq.

To skip ahead in the story: As chaos increased in Baghdad last summer, the chief U.S. administrator, L. Paul "Jerry" Bremer, wrestled constantly with a variant of this exile paradox. The Iraqi Governing Council, whose twenty-five members were chosen by Americans, was supposed to do only the preparatory work for an elected Iraqi government. But the greater the pressure on Bremer for "Iraqification," the more tempted he was to give in to the council's demand that he simply put it in charge without waiting for an election. More than a year earlier, long before combat began, the explicit recommendations and implicit lessons of the Future of Iraq project had given the U.S. government a very good idea of what political conflicts it could expect in Iraq.

As combat slowed in Afghanistan and the teams of the Future of Iraq project continued their deliberations, the U.S. government put itself on a wartime footing. In late May the CIA had begun what would become a long series of war-game exercises, to think through the best- and worst-case scenarios after the overthrow of Saddam Hussein. According to a person familiar with the process, one recurring theme in the exercises was the risk of civil disorder after the fall of Baghdad. The exercises explored how to find and secure the weapons of mass destruction that were then assumed to be in and around Baghdad, and indicated that the hardest task would be finding and protecting scientists who knew about the weapons before they could be killed by the regime as it was going down.

The CIA also considered whether a new Iraqi government could be put together through a process like the Bonn conference, which was then being used to devise a post-Taliban regime for Afghanistan. At the Bonn conference representatives of rival political and ethic groups agreed on the terms that established Hamid Karzai as the new Afghan President. The CIA believed that rivalries in Iraq were so deep, and the political culture so shallow, that a similarly quick transfer of sovereignty would only invite chaos.

Representatives from the Defense Department were among those who participated in the first of these CIA war-game sessions. When their Pentagon superiors at the Office of the Secretary of Defense (OSD) found out about this, in early summer, the representatives were reprimanded and told not to participate further. "OSD" is Washington shorthand, used frequently in discussions about the origins of Iraq war plans, and it usually refers to strong guidance from Rumsfeld, Wolfowitz, Feith, and one of Feith's deputies, William Luti. Their displeasure over the CIA exercise was an early illustration of a view that became stronger throughout 2002: that postwar planning was an impediment to war.

Because detailed thought about the postwar situation meant facing costs and potential problems, and thus weakened the case for launching a "war of choice" (the Washington term for a war not waged in immediate self-defense), it could be seen as an "antiwar" undertaking. The knowledge that U.S. soldiers would still be in Germany and Japan sixty-plus years after Pearl Harbor would obviously not have changed the decision to enter World War II, and in theory the Bush Administration could have presented the overthrow of Saddam Hussein in a similar way: as a job that had to be done, even though it might saddle Americans with costs and a military presence for decades to come. Everyone can think of moments when Bush or Rumsfeld has reminded the nation that this would be a long-term challenge. But during the months when the Administration was making its case for the war—successfully to Congress, less so to the United Nations—it acted as if the long run should be thought about only later on.

On July 31, 2002, the Senate Foreign Relations Committee invited a panel of experts to discuss the case for war against Iraq. On August 1 it heard from other experts about the likely "day after" consequences of military victory. Senator Joseph Biden, a Democrat from Delaware, was then the chairman of the committee. That first day Biden said that the threat of WMD might force him to vote in favor of the war (as he ultimately did). But he worried that if the United States invaded without full allied support, "we may very well radicalize the rest of the world, we may pick up a bill that's $70 billion, $80 billion, we may have to have extensive commitment of U.S. forces for an extended period of time in Iraq."

Phebe Marr, an Iraq scholar retired from the National Defense University, told the committee that the United States "should assume that it cannot get the results it wants on the cheap" from regime change. "It must be prepared to put some troops on the ground, advisers to help create new institutions, and above all, time and effort in the future to see the project through to a satisfactory end. If the United States is not willing to do so, it had best rethink the project." Rend Rahim Francke, an Iraqi exile serving on the Future of Iraq project (and now the ambassador from Iraq to the United States), said that "the system of public security will break down, because there will be no functioning police force, no civil service, and no justice system" on the first day after the fighting. "There will be a vacuum of political authority and administrative authority," she said. "The infrastructure of vital sectors will have to be restored. An adequate police force must be trained and equipped as quickly as possible. And the economy will have to be jump-started from not only stagnation but devastation." Other witnesses discussed the need to commit U.S. troops for many years—but to begin turning constitutional authority over to the Iraqis within six months. The upshot of the hearings was an emphasis on the short-term importance of security, the medium-term challenge of maintaining control while transferring sovereignty to the Iraqis, and the long-term reality of commitments and costs. All the experts agreed that what came after the fall of Baghdad would be harder for the United States than what came before.

One week before Labor Day, while President Bush was at his ranch in Texas, Vice President Cheney gave a speech at a Veterans of Foreign Wars convention in Nashville. "There is no doubt that Saddam Hussein now has weapons of mass destruction [and that he will use them] against our friends, against our allies, and against us," Cheney said. Time was running out, he concluded, for America to remove this threat. A few days later CNN quoted a source "intimately familiar with [Colin] Powell's thinking" as saying that Powell was still insistent on the need for allied support and would oppose any war in which the United States would "go it alone ... as if it doesn't give a damn" about other nations' views. Just after Labor Day, Powell apparently won a battle inside the Administration and persuaded Bush to take the U.S. case to the United Nations. On September 12 Bush addressed the UN General Assembly and urged it to insist on Iraqi compliance with its previous resolutions concerning disarmament.

Before the war the Administration exercised remarkable "message discipline" about financial projections. When asked how much the war might cost, officials said that so many things were uncertain, starting with whether there would even be a war, that there was no responsible way to make an estimate. In part this reflected Rumsfeld's emphasis on the unknowability of the future. It was also politically essential, in delaying the time when the Administration had to argue that regime change in Iraq was worth a specific number of billions of dollars.

In September, Lawrence Lindsay, then the chief White House economic adviser, broke discipline. He was asked by The Wall Street Journal how much a war and its aftermath might cost. He replied that it might end up at one to two percent of the gross domestic product, which would mean $100 billion to $200 billion. Lindsay added that he thought the cost of not going to war could conceivably be greater—but that didn't placate his critics within the Administration. The Administration was further annoyed by a report a few days later from Democrats on the House Budget Committee, which estimated the cost of the war at $48 billion to $93 billion. Lindsay was widely criticized in "background" comments from Administration officials, and by the end of the year he had been forced to resign. His comment "made it clear Larry just didn't get it," an unnamed Administration official told The Washington Post when Lindsay left. Lindsay's example could hardly have encouraged others in the Administration to be forthcoming with financial projections. Indeed, no one who remained in the Administration offered a plausible cost estimate until months after the war began.

In September the United States Agency for International Development began to think in earnest about its postwar responsibilities in Iraq. It was the natural contact for nongovernmental organizations, or NGOs, from the United States and other countries that were concerned with relief efforts in Iraq.

USAID's administrator, Andrew Natsios, came to the assignment with a complex set of experiences and instincts. He started his career, in the 1970s, as a Republican state legislator in Massachusetts, and before the Bush Administration he had been the administrator of the state's "Big Dig," the largest public-works effort ever in the country. Before the Big Dig, Natsios spent five years as an executive at a major humanitarian NGO called World Vision. He also served in the Persian Gulf during the 1991 Gulf War, as an Army Reserve officer. By background he was the Administration official best prepared to anticipate the combination of wartime and postwar obligations in Iraq.

At any given moment USAID is drawing up contingency plans for countries that might soon need help. "I actually have a list, which I will not show you," Natsios told me in the fall, "of countries where there may not be American troops soon, but they could fall apart—and if they do, what we could do for them." By mid-September of 2002, six months before the official beginning of Operation Iraqi Freedom, Natsios had additional teams working on plans for Iraq. Representatives of about a dozen relief organizations and NGOs were gathering each week at USAID headquarters for routine coordination meetings. Iraq occupied more and more of their time through 2002. On October 10, one day before Congress voted to authorize the war, the meetings were recast as the Iraq Working Group.

The weekly meetings at USAID quickly settled into a pattern. The representatives of the NGOs would say, "We've dealt with situations like this before, and we know what to expect." The U.S. government representatives would either say nothing or else reply, No, this time it will be different.

The NGOs had experience dealing with a reality that has not fully sunk in for most of the American public. In the nearly three decades since U.S. troops left Vietnam, the American military has fought only two wars as most people understand the term: the two against Saddam Hussein's Iraq. But through the past thirty years U.S. troops have almost continuously been involved in combat somewhere. Because those engagements—in Grenada, Lebanon, Panama, Haiti, Somalia, Bosnia, Kosovo, Afghanistan, and elsewhere—have no obvious connection with one another, politicians and the public usually discuss them as stand-alone cases. Each one seems an aberration from the "real" wars the military is set up to fight.

To the NGO world, these and other modern wars (like the ones in Africa) are not the exception but the new norm: brutal localized encounters that destroy the existing political order and create a need for long-term international supervision and support. Within the U.S. military almost no one welcomes this reality, but many recognize that peacekeeping, policing, and, yes, nation-building are now the expected military tasks. The military has gotten used to working alongside the NGOs—and the NGOs were ready with a checklist of things to worry about once the regime had fallen.

An even larger question about historical precedent began to surface. When Administration officials talked about models for what would happen in Iraq, they almost always referred to the lasting success in Japan and Germany—or else to countries of the former Soviet bloc in Eastern Europe. (A civilian adviser who went to Baghdad early in the occupation recalls looking at his fellow passengers on the military transport plane. The ones who weren't asleep or flipping through magazines were reading books about Japan or Germany, not about the Arab world. "That was not a good sign," he told me.) If one thought of Iraq as Poland, or as the former East Germany, or as the former Czechoslovakia, or as almost any part of the onetime Soviet empire in Eastern Europe other than Romania, one would naturally conclude that regime change in itself would set the country well along the path toward recovery. These countries were fine once their repressive leaders were removed; so might Iraq well be. And if the former Yugoslavia indicated darker possibilities, that could be explained as yet another failure of Clinton-era foreign policy.

Many NGO representatives assumed that postwar recovery would not be so automatic, and that they should begin working on preparations before the combat began. "At the beginning our main message was the need for access," I was told by Sandra Mitchell, the vice-president of the International Rescue Committee, who attended the USAID meetings. Because of U.S. sanctions against Iraq, it was illegal for American humanitarian organizations to operate there. (Journalists were about the only category of Americans who would not get in trouble with their own government by traveling to and spending money in Iraq.) "Our initial messages were like those in any potential crisis situation," Mitchell said, "but the reason we were so insistent in this case was the precarious situation that already existed in Iraq. The internal infrastructure was shot, and you couldn't easily swing in resources from neighboring countries, like in the Balkans." The NGOs therefore asked, as a first step, for a presidential directive exempting them from the sanctions. They were told to expect an answer to this request by December. That deadline passed with no ruling. By early last year the NGOs felt that it was too dangerous to go to Iraq, and the Administration feared that if they went they might be used as hostages. No directive was ever issued.

Through the fall and winter of 2002 the International Rescue Committee, Refugees International, InterAction, and other groups that met with USAID kept warning about one likely postwar problem that, as it turned out, Iraq avoided—a mass flow of refugees—and another that was exactly as bad as everyone warned: the lawlessness and looting of the "day after" in Baghdad. The Bush Administration would later point to the absence of refugees as a sign of the occupation's underreported success. This achievement was, indeed, due in part to a success: the speed and precision of the military campaign itself. But the absence of refugees was also a sign of a profound failure: the mistaken estimates of Iraq's WMD threat. All pre-war scenarios involving huge movements of refugees began with the assumption that Saddam Hussein would use chemical or biological weapons against U.S. troops or his own Kurdish or Shiite populations—and that either the fact or the fear of such assaults would force terrified Iraqis to evacuate.

The power vacuum that led to looting was disastrous. "The looting was not a surprise," Sandra Mitchell told me. "It should not have come as a surprise. Anyone who has witnessed the fall of a regime while another force is coming in on a temporary basis knows that looting is standard procedure. In Iraq there were very strong signals that this could be the period of greatest concern for humanitarian response." One lesson of postwar reconstruction through the 1990s was that even a short period of disorder could have long-lasting effects.

The meetings at USAID gave the veterans of international relief operations a way to register their concerns. The problem was that they heard so little back. "The people in front of us were very well-meaning," says Joel Charny, who represented Refugees International at the meetings. "And in fairness, they were on such a short leash. But the dialogue was one-way. We would tell them stuff, and they would nod and say, Everything's under control. To me it was like the old four-corners offense in basketball. They were there to just dribble out the clock but be able to say they'd consulted with us."

And again the question arose of whether what lay ahead in Iraq would be similar to the other "small wars" of the previous decade-plus or something new. If it was similar, the NGOs had their checklists ready. These included, significantly, the obligations placed on any "occupying power" by the Fourth Geneva Convention, which was signed in 1949 and is mainly a commonsense list of duties—from protecting hospitals to minimizing postwar reprisals—that a victorious army must carry out. "But we were corrected when we raised this point," Sandra Mitchell says. "The American troops would be 'liberators' rather than 'occupiers,' so the obligations did not apply. Our point was not to pass judgment on the military action but to describe the responsibilities."

In the same mid-October week that the Senate approved the war resolution, a team from the Strategic Studies Institute at the Army War College, in Carlisle Barracks, Pennsylvania, began a postwar-planning exercise. Even more explicitly than the NGOs, the Army team insisted that America's military past, reaching back to its conquest of the Philippines, in 1898, would be a useful guide to its future duties in Iraq. As a rule, professional soldiers spend more time thinking and talking about history than other people do; past battles are the only real evidence about doctrine and equipment. The institute—in essence, the War College's think tank—was charged with reviewing recent occupations to help the Army "best address the requirements that will necessarily follow operational victory in a war with Iraq," as the institute's director later said in a foreword to the team's report. "As the possibility of war with Iraq looms on the horizon, it is important to look beyond the conflict to the challenges of occupying the country."

The study's principal authors were Conrad Crane, who graduated from West Point in the early 1970s and taught there as a history professor through the 1990s, and Andrew Terrill, an Army Reserve officer and a strategic-studies professor. With a team of other researchers, which included representatives from the Army and the joint staff as well as other government agencies and think tanks, they began high-speed work on a set of detailed recommendations about postwar priorities. The Army War College report was also connected to a pre-war struggle with yet another profound postwar consequence: the fight within the Pentagon, between the civilian leadership in OSD and the generals running the Army, over the size and composition of the force that would conquer Iraq.

On November 5, 2002, the Republicans regained control of the Senate and increased their majority in the House in national midterm elections. On November 8 the UN Security Council voted 15-0 in favor of Resolution 1441, threatening Iraq with "serious consequences" if it could not prove that it had abandoned its weapons programs.

Just before 9/11 Donald Rumsfeld had been thought of as standing on a banana peel. The newspapers were full of leaked anonymous complaints from military officials who thought that his efforts to streamline and "transform" the Pentagon were unrealistic and damaging. But with his dramatic metamorphosis from embattled Secretary of Defense to triumphant Secretary of War, Rumsfeld's reputation outside the Administration and his influence within it rose. He was operating from a position of great power when, in November, he decided to "cut the TPFDD."

"Tipfid" is how people in the military pronounce the acronym for "time-phased force and deployment data," but what it really means to the armed forces, in particular the Army, is a way of doing business that is methodical, careful, and sure. The TPFDD for Iraq was an unbelievably complex master plan governing which forces would go where, when, and with what equipment, on which planes or ships, so that everything would be coordinated and ready at the time of attack. One reason it took the military six months to get set for each of its wars against Iraq, a comparatively pitiful foe, was the thoroughness of TPFDD planning. To its supporters, this approach is old-school in the best sense: if you fight, you really fight. To its detractors, this approach is simply old—ponderous, inefficient, and, although they don't dare call it cowardly, risk-averse at the least.

A streamlined approach had proved successful in Afghanistan, at least for a while, as a relatively small U.S. force left much of the ground fighting to the Northern Alliance. In the longer run the American strategy created complications for Afghanistan, because the victorious Northern Alliance leaders were newly legitimized as warlords. Donald Rumsfeld was one member of the Administration who seemed still to share the pre-9/11 suspicion about the risks of nation-building, and so didn't much care about the postwar consequences of a relatively small invasion force. (His deputy, Paul Wolfowitz, was more open to the challenge of rebuilding Iraq, but he would never undercut or disobey Rumsfeld.) In November, Rumsfeld began working through the TPFDD, with the goal of paring the force planned for Iraq to its leanest, lightest acceptable level.

The war games run by the Army and the Pentagon's joint staff had led to very high projected troop levels. The Army's recommendation was for an invasion force 400,000 strong, made up of as many Americans as necessary and as many allied troops as possible. "All the numbers we were coming up with were quite large," Thomas White, a retired general (and former Enron executive) who was the Secretary of the Army during the war, told me recently. But Rumsfeld's idea of the right force size was more like 75,000. The Army and the military's joint leadership moderated their requests in putting together the TPFDD, but Rumsfeld began challenging the force numbers in detail. When combat began, slightly more than 200,000 U.S. soldiers were massed around Iraq.

"In what I came to think of as Secretary Rumsfeld's style," an Army official who was involved in the process told me recently, "he didn't directly say no but asked a lot of hard questions about the plan and sent us away without approval. He would ask questions that delayed the activation of units, because he didn't think the planned flow was right. Our people came back with the understanding that their numbers were far too big and they should be thinking more along the lines of Afghanistan"—that is, plan for a light, mobile attack featuring Special Forces soldiers. Another participant described Rumsfeld as looking line by line at the deployments proposed in the TPFDD and saying, "Can't we do this with one company?" or "Shouldn't we get rid of this unit?" Making detailed, last-minute adjustments to the TPFDD was, in the Army's view, like pulling cogs at random out of a machine. According to an observer, "The generals would say, Sir, these changes will ripple back to every railhead and every company."

The longer-term problem involved what would happen after Baghdad fell, as it inevitably would. This was distinctly an Army rather than a general military concern. "Where's the Air Force now?" an Army officer asked rhetorically last fall. "They're back on their bases—and they're better off, since they don't need to patrol the 'no-fly' zones [in northern and southern Iraq, which U.S. warplanes had patrolled since the end of the Gulf War]. The Navy's gone, and most of the Marines have been pulled back. It's the Army holding the sack of shit." A related concern involved what a long-term commitment to Iraq would do to the Army's "ops tempo," or pace of operations—especially if Reserve and National Guard members, who had no expectations of long-term foreign service when they signed up, were posted in Iraq for months or even years.

The military's fundamental argument for building up what Rumsfeld considered a wastefully large force is that it would be even more useful after Baghdad fell than during actual combat. The first few days or weeks after the fighting, in this view, were crucial in setting long-term expectations. Civilians would see that they could expect a rapid return to order, and would behave accordingly—or they would see the opposite. This was the "shock and awe" that really mattered, in the Army's view: the ability to make clear who was in charge. "Insights from successful occupations suggest that it is best to go in real heavy and then draw down fast," Conrad Crane, of the Army War College, told me. That is, a larger force would be necessary during and immediately after the war, but might mean a much smaller occupation presence six months later.

"We're in Baghdad, the regime is toppled—what's next?" Thomas White told me, recounting discussions before the war. One of the strongest advocates of a larger force was General Eric Shinseki, the Army Chief of Staff. White said, "Guys like Shinseki, who had been in Bosnia [where he supervised the NATO force], been in Kosovo, started running the numbers and said, 'Let's assume the world is linear.' For five million Bosnians we had two hundred thousand people to watch over them. Now we have twenty-five million Iraqis to worry about, spread out over a state the size of California. How many people is this going to take?" The heart of the Army's argument was that with too few soldiers, the United States would win the war only to be trapped in an untenable position during the occupation.

A note of personal rancor complicated these discussions, as it did many disagreements over postwar plans. In our interview Douglas Feith played this down—maintaining that press reports had exaggerated the degree of quarreling and division inside the Administration. These reports, he said, mainly reflected the experience of lower-level officials, who were embroiled in one specific policy area and "might find themselves pretty much always at odds with their counterparts from another agency." Higher up, where one might be "fighting with someone on one issue but allied with them on something else," relations were more collegial. Perhaps so. But there was no concealing the hostility within the Pentagon between most uniformed leaders, especially in the Army, and the civilians in OSD.

Donald Rumsfeld viewed Shinseki as a symbol of uncooperative, old-style thinking, and had in the past gone out of his way to humiliate him. In the spring of 2002, fourteen months before the scheduled end of Shinseki's term, Rumsfeld announced who his successor would be; such an announcement, which converts the incumbent into a lame duck, usually comes at the last minute. The action was one of several calculated insults.

From OSD's point of view, Shinseki and many of his colleagues were dragging their feet. From the Army's point of view, OSD was being reckless about the way it was committing troops and high-handed in disregarding the military's professional advice. One man who was then working in the Pentagon told me of walking down a hallway a few months before the war and seeing Army General John Abizaid standing outside a door. Abizaid, who after the war succeeded Tommy Franks as commander of the Central Command, or CENTCOM, was then the director of the Joint Staff—the highest uniformed position in the Pentagon apart from the Joint Chiefs. A planning meeting for Iraq operations was under way. OSD officials told him he could not take part.

The military-civilian difference finally turned on the question of which would be harder: winning the war or maintaining the peace. According to Thomas White and several others, OSD acted as if the war itself would pose the real challenge. As White put it, "The planning assumptions were that the people would realize they were liberated, they would be happy that we were there, so it would take a much smaller force to secure the peace than it did to win the war. The resistance would principally be the remnants of the Baath Party, but they would go away fairly rapidly. And, critically, if we didn't damage the infrastructure in our military operation, as we didn't, the restart of the country could be done fairly rapidly." The first assumption was clearly expressed by Cheney three days before the war began, in an exchange with Tim Russert on Meet the Press:

RUSSERT: If your analysis is not correct, and we're not treated as liberators but as conquerors, and the Iraqis begin to resist, particularly in Baghdad, do you think the American people are prepared for a long, costly, and bloody battle with significant American casualties?

CHENEY: Well, I don't think it's likely to unfold that way, Tim, because I really do believe that we will be greeted as liberators ... The read we get on the people of Iraq is there is no question but what they want to get rid of Saddam Hussein and they will welcome as liberators the United States when we come to do that.

Through the 1990s Marine General Anthony Zinni, who preceded Tommy Franks as CENTCOM commander, had done war-gaming for a possible invasion of Iraq. His exercises involved a much larger U.S. force than the one that actually attacked last year. "They were very proud that they didn't have the kind of numbers my plan had called for," Zinni told me, referring to Rumsfeld and Cheney. "The reason we had those two extra divisions was the security situation. Revenge killings, crime, chaos—this was all foreseeable."

Thomas White agrees. Because of reasoning like Cheney's, "we went in with the minimum force to accomplish the military objectives, which was a straightforward task, never really in question," he told me. "And then we immediately found ourselves shorthanded in the aftermath. We sat there and watched people dismantle and run off with the country, basically."

In the beginning of December, Iraq submitted its 12,000-page declaration to the UN Security Council contending that it had no remaining WMD stores. Near the end of December, President Bush authorized the dispatch of more than 200,000 U.S. soldiers to the Persian Gulf.

There had still been few or no estimates of the war's cost from the Administration—only contentions that projections like Lawrence Lindsay's were too high. When pressed on this point, Administration officials repeatedly said that with so many uncertainties, they could not possibly estimate the cost. But early in December, just before Lindsay was forced out, The New York Review of Books published an article by William Nordhaus titled "Iraq: The Economic Consequences of War," which included carefully considered estimates. Nordhaus, an economist at Yale, had served on Jimmy Carter's Council of Economic Advisers; the article was excerpted from a much longer economic paper he had prepared. His range of estimates was enormous, depending on how long the war lasted and what its impact on the world economy proved to be. Nordhaus calculated that over the course of a decade the direct and indirect costs of the war to the United States could be as low as $121 billion or as high as $1.6 trillion. This was a more thoroughgoing approach than the congressional budget committees had taken, but it was similar in its overall outlook. Nordhaus told me recently that he thinks he should have increased all his estimates to account for the "opportunity costs" of stationing soldiers in Iraq—that is, if they are assigned to Iraq, they're not available for deployment somewhere else.

On the last day of December, Mitch Daniels, the director of the Office of Management and Budget, told The New York Times that the war might cost $50 billion to $60 billion. He had to backtrack immediately, his spokesman stressing that "it is impossible to know what any military campaign would ultimately cost." The spokesman explained Daniels's mistake by saying, "The only cost estimate we know of in this arena is the Persian Gulf War, and that was a sixty-billion-dollar event." Daniels would leave the Administration, of his own volition, five months later.

In the immediate run-up to the war the Administration still insisted that the costs were unforeseeable. "Fundamentally, we have no idea what is needed unless and until we get there on the ground," Paul Wolfowitz told the House Budget Committee on February 27, with combat less than three weeks away. "This delicate moment—when we are assembling a coalition, when we are mobilizing people inside Iraq and throughout the region to help us in the event of war, and when we are still trying, through the United Nations and by other means, to achieve a peaceful solution without war—is not a good time to publish highly suspect numerical estimates and have them drive our declaratory policy."

Wolfowitz's stonewalling that day was in keeping with the policy of all senior Administration officials. Until many months after combat had begun, they refused to hazard even the vaguest approximation of what financial costs it might involve. Shinseki, so often at odds with OSD, contemplated taking a different course. He was scheduled to testify, with Thomas White, before the Senate Appropriations Committee on March 19, which turned out to be the first day of actual combat. In a routine prep session before the hearing he asked his assistants what he should say about how much the operations in Iraq were going to cost. "Well, it's impossible to predict," a briefer began, reminding him of the official line.

Shinseki cut him off. "We don't know everything," he said, and then he went through a list of the many things the military already did know. "We know how many troops are there now, and the projected numbers. We know how much it costs to feed them every day. We know how much it cost to send the force there. We know what we have spent already to prepare the force and how much it would cost to bring them back. We have estimates of how much fuel and ammunition we would use per day of operations." In short, anyone who actually wanted to make an estimate had plenty of information on hand.

At this point Jerry Sinn, a three-star general in charge of the Army's budget, said that in fact he had worked up some numbers—and he named a figure, for the Army's likely costs, in the tens of billions of dollars. But when Senator Byron Dorgan, of North Dakota, asked Shinseki at hearings on March 19 how much the war just beginning would cost, Shinseki was loyally vague ("Any potential discussion about what an operation in Iraq or any follow-on probably is undefined at this point").

When Administration officials stopped being vague, they started being unrealistic. On March 27, eight days into combat, members of the House Appropriations Committee asked Paul Wolfowitz for a figure. He told them that whatever it was, Iraq's oil supplies would keep it low. "There's a lot of money to pay for this," he said. "It doesn't have to be U.S. taxpayer money. We are dealing with a country that can really finance its own reconstruction, and relatively soon." On April 23 Andrew Natsios, of USAID, told an incredulous Ted Koppel, on Nightline, that the total cost to America of reconstructing Iraq would be $1.7 billion. Koppel shot back, "I mean, when you talk about one-point-seven, you're not suggesting that the rebuilding of Iraq is gonna be done for one-point-seven billion dollars?" Natsios was clear: "Well, in terms of the American taxpayers' contribution, I do; this is it for the U.S. The rest of the rebuilding of Iraq will be done by other countries who have already made pledges ... But the American part of this will be one-point-seven billion dollars. We have no plans for any further-on funding for this." Only in September did President Bush make his request for a supplemental appropriation of $87 billion for operations in Iraq.

Planning for the postwar period intensified in December. The Council on Foreign Relations, working with the Baker Institute for Public Policy, at Rice University, convened a working group on "guiding principles for U.S. post-war conflict policy in Iraq." Leslie Gelb, then the president of the Council on Foreign Relations, said that the group would take no position for or against the war. But its report, which was prepared late in January of last year, said that "U.S. and coalition military units will need to pivot quickly from combat to peacekeeping operations in order to prevent post-conflict Iraq from descending into anarchy." The report continued, "Without an initial and broad-based commitment to law and order, the logic of score-settling and revenge-taking will reduce Iraq to chaos."

The momentum toward war put officials at the United Nations and other international organizations in a difficult position. On the one hand, they had to be ready for what was coming; on the other, it was awkward to be seen discussing the impending takeover of one of their member states by another. "Off-the-record meetings were happening in every bar in New York," one senior UN official told me in the fall. An American delegation that included Pentagon representatives went to Rome in December for a confidential meeting with officials of the UN's World Food Programme, to discuss possible food needs after combat in Iraq. As The Wall Street Journal later reported, the meeting was uncomfortable for both sides: the Americans had to tell the WFP officials, as one of them recalled, "It is looking most probable you are going to witness one of the largest military engagements since the Second World War." This was hyperbole (Korea? Vietnam?), but it helped to convince the WFP that relief preparations should begin.

On December 11 an ice storm hit the Mid-Atlantic states. For Conrad Crane and his associates at the Army War College, deep in their crash effort to prepare their report on postwar Army challenges, this was a blessing. "The storm worked out perfectly," Crane told me afterward. "We were all on the post, there was no place anyone could go, we basically had the whole place to ourselves."

By the end of the month the War College team had assembled a draft of its report, called "Reconstructing Iraq: Insights, Challenges, and Missions for Military Forces in a Post-Conflict Scenario." It was not classified, and can be found through the Army War College's Web site.

The War College report has three sections. The first is a review of twentieth-century occupations—from the major efforts in Japan and Germany to the smaller and more recent ones in Haiti, Panama, and the Balkans. The purpose of the review is to identify common situations that occupiers might face in Iraq. The discussion of Germany, for instance, includes a detailed account of how U.S. occupiers "de-Nazified" the country without totally dismantling its bureaucracy or excluding everyone who had held a position of responsibility. (The main tool was a Fragebogen, or questionnaire, about each person's past activities, which groups of anti-Nazi Germans and Allied investigators reviewed and based decisions on.)

The second section of the report is an assessment of the specific problems likely to arise in Iraq, given its ethnic and regional tensions and the impact of decades of Baathist rule. Most Iraqis would welcome the end of Saddam Hussein's tyranny, it said. Nonetheless,

Long-term gratitude is unlikely and suspicion of U.S. motives will increase as the occupation continues. A force initially viewed as liberators can rapidly be relegated to the status of invaders should an unwelcome occupation continue for a prolonged time. Occupation problems may be especially acute if the United States must implement the bulk of the occupation itself rather than turn these duties over to a postwar international force.

If these views about the risk of disorder and the short welcome that Americans would enjoy sound familiar, that is because every organization that looked seriously into the situation sounded the same note.

The last and most distinctive part of the War College report is its "Mission Matrix"—a 135-item checklist of what tasks would have to be done right after the war and by whom. About a quarter of these were "critical tasks" for which the military would have to be prepared long before it reached Baghdad: securing the borders so that foreign terrorists would not slip in (as they in fact did), locating and destroying WMD supplies, protecting religious sites, performing police and security functions, and so on. The matrix was intended to lay out a phased shift of responsibilities, over months or years, from a mainly U.S. occupation force to international organizations and, finally, to sovereign Iraqis. By the end of December copies of the War College report were being circulated throughout the Army.

According to the standard military model, warfare unfolds through four phases: "deterrence and engagement," "seize the initiative," "decisive operations," and "post-conflict." Reality is never divided quite that neatly, of course, but the War College report stressed that Phase IV "post-conflict" planning absolutely had to start as early as possible, well before Phase III "decisive operations"—the war itself. But neither the Army nor the other services moved very far past Phase III thinking. "All the A-Team guys wanted to be in on Phase III, and the B-team guys were put on Phase IV," one man involved in Phase IV told me. Frederick Barton, of the Center for Strategic and International Studies, who was involved in postwar efforts in Haiti, Rwanda, and elsewhere, put it differently. "If you went to the Pentagon before the war, all the concentration was on the war," he said. "If you went there during the war, all the concentration was on the war. And if you went there after the war, they'd say, 'That's Jerry Bremer's job.'" Still, the War College report confirmed what the Army leadership already suspected: that its real challenges would begin when it took control of Baghdad.

On January 27, 2003, the chief UN weapons inspector, Hans Blix, reported that "Iraq appears not to have come to a genuine acceptance, not even today, of the disarmament that was demanded of it." Twenty-four hours later, in his State of the Union address, President Bush said that the United States was still hoping for UN endorsement of an action against Iraq—but would not be limited by the absence of one.

Increasingly the question in Washington about war was When? Those arguing for delay said that it would make everything easier. Perhaps Saddam Hussein would die. Perhaps he would flee or be overthrown. Perhaps the UN inspectors would find his weapons, or determine conclusively that they no longer existed. Perhaps the United States would have time to assemble, if not a broad alliance for the battle itself, at least support for reconstruction and occupation, so that U.S. soldiers and taxpayers would not be left with the entire job. Even if the responsibility were to be wholly America's, each passing month would mean more time to plan the peace as thoroughly as the war: to train civil-affairs units (which specialize in peacekeeping rather than combat), and to hire Arabic-speakers. Indeed, several months into the U.S. occupation a confidential Army "lessons learned" study said that the "lack of competent interpreters" throughout Iraq had "impeded operations." Most of the "military linguists" who were operating in Iraq, the study said, "basically [had] the ability to tell the difference between a burro and a burrito."

Those arguing against delay said that the mere passage of time wouldn't do any good and would bring various risks. The world had already waited twelve years since the Gulf War for Saddam Hussein to disarm. Congress had already voted to endorse the war. The Security Council had already shown its resolve. The troops were already on their way. Each passing day, in this view, was a day in which Saddam Hussein might deploy his weapons of terror.

Early in January the National Intelligence Council, at the CIA, ran a two-day exercise on postwar problems. Pentagon representatives were still forbidden by OSD to attend. The exercise covered issues similar to those addressed in the Future of Iraq and Army War College reports—and, indeed, to those considered by the Council on Foreign Relations and the Senate Foreign Relations Committee: political reconstruction, public order, border control, humanitarian problems, finding and securing WMD.

On January 15 the humanitarian groups that had been meeting at USAID asked for a meeting with Donald Rumsfeld or Paul Wolfowitz. They never got one. At an earlier meeting, according to a participant, they had been told, "The President has already spent an hour on the humanitarian issues." The most senior Pentagon official to meet with them was Joseph Collins, a deputy assistant secretary of defense. The representatives of the NGOs were generally the most senior and experienced figures from each organization; the government representatives were not of the same stature. "Without naming names, the people we met were not real decision-makers," Joel Charny says.

On January 24 a group of archaeologists and scholars went to the Pentagon to brief Collins and other officials about the most important historic sites in Iraq, so that they could be spared in bombing. Thanks to precision targeting, the sites would indeed survive combat. Many, of course, were pillaged almost immediately afterward.

On January 30 the International Rescue Committee, which had been participating in the weekly Iraq Working Group sessions, publicly warned that a breakdown of law and order was likely unless the victorious U.S. forces acted immediately, with martial law if necessary, to prevent it. A week later Refugees International issued a similar warning.

At the end of January, Sam Gardiner entered the picture. Gardiner is a retired Air Force colonel who taught for years at the National War College in Washington. His specialty is war gaming, and through the 1990s he was involved once or twice a year in major simulations involving an attack on Baghdad. In the late 1990s Gardiner had been a visiting scholar at the Swedish National Defense University, where he studied the effects of the bombing of Serbia's electrical grid. The big discovery was how long it took to get the system up and running again, after even a precise and limited attack. "Decapitation" attacks on a regime, like the one planned for Iraq, routinely begin with disabling the electrical grid. Gardiner warned that this Phase III step could cause big Phase IV problems.

Late in 2002 Gardiner had put together what he called a "net assessment" of how Iraq would look after a successful U.S. attack. His intended audience, in government, would recognize the designation as droll. "Net assessment" is a familiar term for a CIA-style intelligence analysis, but Gardiner also meant it to reflect the unusual origin of his data: none of it was classified, and all of it came from the Internet. Through the power of search engines Gardiner was able to assemble what in other days would have seemed like a secret inside look at Iraq's infrastructure. He found electricity diagrams for the pumps used at Iraq's main water stations; he listed replacement parts for the most vulnerable elements of the electrical grid. He produced a scheme showing the elements of the system that would be easiest to attack but then quickest to repair. As it happened, damage to the electrical grid was a major postwar problem. Despite the precision of the bombing campaign, by mid-April wartime damage and immediate postwar looting had reduced Baghdad's power supply to one fifth its pre-war level, according to an internal Pentagon study. In mid-July the grid would be back to only half its pre-war level, working on a three-hours-on, three-hours-off schedule.

On January 19 Gardiner presented his net assessment, with information about Iraq's water, sewage, and public-health systems as well as its electrical grid, at an unclassified forum held by the RAND Corporation, in Washington. Two days later he presented it privately to Zalmay Khalilzad. Khalilzad was a former RAND analyst who had joined the Bush Administration's National Security Council and before the war was named the President's "special envoy and ambassador-at-large for Free Iraqis." (He has recently become the U.S. ambassador to Afghanistan.) Gardiner told me recently that Khalilzad was sobered by what he heard, and gave Gardiner a list of other people in the government who should certainly be shown the assessment. In the next few weeks Gardiner presented his findings to Bear McConnell, the USAID official in charge of foreign disaster relief, and Michael Dunn, an Air Force general who had once been Gardiner's student and worked with the Joint Chiefs of Staff as acting director for strategic plans and policy. A scheduled briefing with Joseph Collins, who was becoming the Pentagon's point man for postwar planning, was canceled at the last minute, after a description of Gardiner's report appeared in Inside the Pentagon, an influential newsletter.

The closer the nation came to war, the more the Administration seemed to view people like Gardiner as virtual Frenchmen—that is, softies who would always find some excuse to oppose the war. In one sense they were right. "It became clear that what I was really arguing was that we had to delay the war," Gardiner told me. "I was saying, 'We aren't ready, and in just six or eight weeks there is no way to get ready for everything we need to do.'" (The first bombs fell on Baghdad eight weeks after Gardiner's meeting with Khalilzad.) "Everyone was very interested and very polite and said I should talk to other people," Gardiner said. "But they had that 'Stalingrad stare'—people who had been doing stuff under pressure for too long and hadn't had enough sleep. You want to shake them and say, 'Are you really with me?'"

At the regular meeting of the Iraq Working Group on January 29, the NGO representatives discussed a recent piece of vital news. The Administration had chosen a leader for all postwar efforts in Iraq: Jay M. Garner, a retired three-star Army general who had worked successfully with the Kurds at the end of the Gulf War. The NGO representatives had no fault to find with the choice of Garner, but they were concerned, because his organization would be a subunit of the Pentagon rather than an independent operation or part of a civilian agency. "We had been pushing constantly to have reconstruction authority based in the State Department," Joel Charny told me. He and his colleagues were told by Wendy Chamberlin, a former ambassador to Pakistan who had become USAID's assistant administrator for the area including Iraq, that the NGOs should view Garner's appointment as a victory. After all, Garner was a civilian, and his office would draw representatives from across the government. "We said, 'C'mon, Wendy, his office is in the Pentagon!'" Charny says. Jim Bishop, a former U.S. ambassador who now works for InterAction, pointed out that the NGOs, like the U.S. government, were still hoping that other governments might help to fund humanitarian efforts. Bishop asked rhetorically, "Who from the international community is going to fund reconstruction run through the Pentagon?"

Garner assembled a team and immediately went to work. What happened to him in the next two months is the best-chronicled part of the postwar fiasco. He started from scratch, trying to familiarize himself with what the rest of the government had already done. On February 21 he convened a two-day meeting of diplomats, soldiers, academics, and development experts, who gathered at the National Defense University to discuss postwar plans. "The messiah could not have organized a sufficient relief and reconstruction or humanitarian effort in that short a time," a former CIA analyst named Judith Yaphe said after attending the meeting, according to Mark Fineman, Doyle McManus, and Robin Wright, of the Los Angeles Times. (Fineman died of a heart attack last fall, while reporting from Baghdad.) Garner was also affected by tension between OSD and the rest of the government. Garner had heard about the Future of Iraq project, although Rumsfeld had told him not to waste his time reading it. Nonetheless, he decided to bring its director, Thomas Warrick, onto his planning team. Garner, who clearly does not intend to be the fall guy for postwar problems in Baghdad, told me last fall that Rumsfeld had asked him to kick Warrick off his staff. In an interview with the BBC last November, Garner confirmed details of the firing that had earlier been published in Newsweek. According to Garner, Rumsfeld asked him, "Jay, have you got a guy named Warrick on your team?" "I said, 'Yes, I do.' He said, 'Well, I've got to ask you to remove him.' I said, 'I don't want to remove him; he's too valuable.' But he said, 'This came to me from such a high level that I can't overturn it, and I've just got to ask you to remove Mr. Warrick.'" Newsweek's conclusion was that the man giving the instructions was Vice President Cheney.

This is the place to note that in several months of interviews I never once heard someone say "We took this step because the President indicated ..." or "The President really wanted ..." Instead I heard "Rumsfeld wanted," "Powell thought," "The Vice President pushed," "Bremer asked," and so on. One need only compare this with any discussion of foreign policy in Reagan's or Clinton's Administration—or Nixon's, or Kennedy's, or Johnson's, or most others—to sense how unusual is the absence of the President as prime mover. The other conspicuously absent figure was Condoleezza Rice, even after she was supposedly put in charge of coordinating Administration policy on Iraq, last October. It is possible that the President's confidants are so discreet that they have kept all his decisions and instructions secret. But that would run counter to the fundamental nature of bureaucratic Washington, where people cite a President's authority whenever they possibly can ("The President feels strongly about this, so ...").

To me, the more likely inference is that Bush took a strong overall position—fighting terrorism is this generation's challenge—and then was exposed to only a narrow range of options worked out by the contending forces within his Administration. If this interpretation proves to be right, and if Bush did in fact wish to know more, then blame will fall on those whose responsibility it was to present him with the widest range of choices: Cheney and Rice.

On February 14 Hans Blix reaffirmed to the United Nations his view that Iraq had decided to cooperate with inspectors. The division separating the United States and Britain from France, Germany, and Russia became stark. On February 15 antiwar demonstrators massed in major cities around the world: a million in Madrid, more than a million in Rome, and a million or more in London, the largest demonstration in Britain's history.

On February 21 Tony Blair joined George Bush at Camp David, to underscore their joint determination to remove the threat from Iraq.

As the war drew near, the dispute about how to conduct it became public. On February 25 the Senate Armed Services Committee summoned all four Chiefs of Staff to answer questions about the war—and its aftermath. The crucial exchange began with a question from the ranking Democrat, Carl Levin. He asked Eric Shinseki, the Army Chief of Staff, how many soldiers would be required not to defeat Iraq but to occupy it. Well aware that he was at odds with his civilian superiors at the Pentagon, Shinseki at first deflected the question. "In specific numbers," he said, "I would have to rely on combatant commanders' exact requirements. But I think ..." and he trailed off.

"How about a range?" Levin asked. Shinseki replied—and recapitulated the argument he had made to Rumsfeld.

I would say that what's been mobilized to this point, something on the order of several hundred thousand soldiers, are probably, you know, a figure that would be required.

We're talking about post-hostilities control over a piece of geography that's fairly significant, with the kinds of ethnic tensions that could lead to other problems. And so, it takes significant ground force presence to maintain safe and secure environment to ensure that the people are fed, that water is distributed, all the normal responsibilities that go along with administering a situation like this.

Two days later Paul Wolfowitz appeared before the House Budget Committee. He began working through his prepared statement about the Pentagon's budget request and then asked permission to "digress for a moment" and respond to recent commentary, "some of it quite outlandish, about what our postwar requirements might be in Iraq." Everyone knew he meant Shinseki's remarks.

"I am reluctant to try to predict anything about what the cost of a possible conflict in Iraq would be," Wolfowitz said, "or what the possible cost of reconstructing and stabilizing that country afterwards might be." This was more than reluctance—it was the Administration's consistent policy before the war. "But some of the higher-end predictions that we have been hearing recently, such as the notion that it will take several hundred thousand U.S. troops to provide stability in post-Saddam Iraq, are wildly off the mark."

This was as direct a rebuke of a military leader by his civilian superior as the United States had seen in fifty years. Wolfowitz offered a variety of incidental reasons why his views were so different from those he alluded to: "I would expect that even countries like France will have a strong interest in assisting Iraq's reconstruction," and "We can't be sure that the Iraqi people will welcome us as liberators ... [but] I am reasonably certain that they will greet us as liberators, and that will help us to keep requirements down." His fundamental point was this: "It's hard to conceive that it would take more forces to provide stability in post-Saddam Iraq than it would take to conduct the war itself and to secure the surrender of Saddam's security forces and his army. Hard to imagine."

None of the government working groups that had seriously looked into the question had simply "imagined" that occupying Iraq would be more difficult than defeating it. They had presented years' worth of experience suggesting that this would be the central reality of the undertaking. Wolfowitz either didn't notice this evidence or chose to disbelieve it. What David Halberstam said of Robert McNamara in The Best and the Brightest is true of those at OSD as well: they were brilliant, and they were fools.

At the beginning of March, Andrew Natsios won a little-noticed but crucial battle. Because the United States had not yet officially decided whether to go to war, Natsios had not been able to persuade the Office of Management and Budget to set aside the money that USAID would need for immediate postwar efforts in Iraq. The battle was the more intense because Natsios, unlike his counterparts at the State Department, was both privately and publicly supportive of the case for war. Just before combat he was able to arrange an emergency $200 million grant from USAID to the World Food Programme. This money could be used to buy food immediately for Iraqi relief operations—and it helped to ensure that there were no postwar food shortages.

On March 13 humanitarian organizations had gathered at USAID headquarters for what was effectively the last meeting of the Iraq Working Group. Wendy Chamberlin, the senior USAID official present, discussed the impending war in terms that several participants noted, wrote down, and later mentioned to me. "It's going to be very quick," she said, referring to the actual war. "We're going to meet their immediate needs. We're going to turn it over to the Iraqis. And we're going to be out within the year."

On March 17 the United States, Britain, and Spain announced that they would abandon their attempt to get a second Security Council vote in favor of the war, and President Bush gave Saddam Hussein an ultimatum: leave the country within forty-eight hours or suffer the consequences. On March 19 the first bombs fell on Baghdad.

On April 9 U.S. forces took Baghdad. On April 14 the Pentagon announced that most of the fighting was over. On May 1 President Bush declared that combat operations were at an end. By then looting had gone on in Baghdad for several weeks. "When the United States entered Baghdad on April 9, it entered a city largely undamaged by a carefully executed military campaign," Peter Galbraith, a former U.S. ambassador to Croatia, told a congressional committee in June. "However, in the three weeks following the U.S. takeover, unchecked looting effectively gutted every important public institution in the city—with the notable exception of the oil ministry." On April 11, when asked why U.S. soldiers were not stopping the looting, Donald Rumsfeld said, "Freedom's untidy, and free people are free to make mistakes and commit crimes and do bad things. They're also free to live their lives and do wonderful things, and that's what's going to happen here."

This was a moment, as when he tore up the TPFDD, that Rumsfeld crossed a line. His embrace of "uncertainty" became a reckless evasion of responsibility. He had only disdain for "predictions," yes, and no one could have forecast every circumstance of postwar Baghdad. But virtually everyone who had thought about the issue had warned about the risk of looting. U.S. soldiers could have prevented it—and would have, if so instructed.

The looting spread, destroying the infrastructure that had survived the war and creating the expectation of future chaos. "There is this kind of magic moment, which you can't imagine until you see it," an American civilian who was in Baghdad during the looting told me. "People are used to someone being in charge, and when they realize no one is, the fabric rips."

On May 6 the Administration announced that Bremer would be the new U.S. administrator in Iraq. Two weeks into that job Bremer disbanded the Iraqi army and other parts of the Baathist security structure.

If the failure to stop the looting was a major sin of omission, sending the Iraqi soldiers home was, in the view of nearly everyone except those who made the decision, a catastrophic error of commission. There were two arguments for taking this step. First, the army had "already disbanded itself," as Douglas Feith put it to me—soldiers had melted away, with their weapons. Second, the army had been an integral part of the Sunni-dominated Baathist security structure. Leaving it intact would be the wrong symbol for the new Iraq—especially for the Shiites, whom the army had oppressed. "These actions are part of a robust campaign to show the Iraqi people that the Saddam regime is gone, and will never return," a statement from Bremer's office said.

The case against wholesale dissolution of the army, rather than a selective purge at the top, was that it created an instant enemy class: hundreds of thousands of men who still had their weapons but no longer had a paycheck or a place to go each day. Manpower that could have helped on security patrols became part of the security threat. Studies from the Army War College, the Future of Iraq project, and the Center for Strategic and International Studies, to name a few, had all considered exactly this problem and suggested ways of removing the noxious leadership while retaining the ordinary troops. They had all warned strongly against disbanding the Iraqi army. The Army War College, for example, said in its report, "To tear apart the Army in the war's aftermath could lead to the destruction of one of the only forces for unity within the society."

"This is not something that was dreamed up by somebody at the last minute," Walter Slocombe—who held Feith's job, undersecretary of defense for policy, during the Clinton Administration, and who is now a security adviser on Bremer's team—told Peter Slevin, of The Washington Post, last November. He said that he had discussed the plan with Wolfowitz at least once and with Feith several times, including the day before the order was given. "The critical point," he told Slevin, "was that nobody argued that we shouldn't do this." No one, that is, the Administration listened to.

Here is the hardest question: How could the Administration have thought that it was safe to proceed in blithe indifference to the warnings of nearly everyone with operational experience in modern military occupations? Saying that the Administration considered this a truly urgent "war of necessity" doesn't explain the indifference. Even if it feared that Iraq might give terrorists fearsome weapons at any moment, it could still have thought more carefully about the day after the war. World War II was a war of absolute necessity, and the United States still found time for detailed occupation planning.

The President must have known that however bright the scenarios, the reality of Iraq eighteen months after the war would affect his re-election. The political risk was enormous and obvious. Administration officials must have believed not only that the war was necessary but also that a successful occupation would not require any more forethought than they gave it.

It will be years before we fully understand how intelligent people convinced themselves of this. My guess is that three factors will be important parts of the explanation.

One is the panache of Donald Rumsfeld. He was near the zenith of his influence as the war was planned. His emphasis on the vagaries of life was all the more appealing within his circle because of his jauntiness and verve. But he was not careful about remembering his practical obligations. Precisely because he could not foresee all hazards, he should have been more zealous about avoiding the ones that were evident—the big and obvious ones the rest of the government tried to point out to him.

A second is the triumphalism of the Administration. In the twenty-five years since Ronald Reagan's rise, political conservatives have changed position in a way they have not fully recognized. Reagan's arrival marked the end of a half century of Democrat-dominated government in Washington. Yes, there has been one Democratic President since Reagan, and eventually there will be others. But as a rule the Republicans are now in command. Older Republicans—those who came of age in the 1960s and 1970s, those who are now in power in the Administration—have not fully adjusted to this reality. They still feel like embattled insurgents, as if the liberals were in the driver's seat. They recognize their electoral strength but feel that in the battle of ideology their main task is to puncture fatuous liberal ideas.

The consequence is that Republicans are less used to exposing their own ideas to challenges than they should be. Today's liberals know there is a challenge to every aspect of their world view. All they have to do is turn on the radio. Today's conservatives are more likely to think that any contrary ideas are leftovers from the tired 1960s, much as liberals of the Kennedy era thought that conservatives were in thrall to Herbert Hoover. In addition, the conservatives' understanding of modern history makes them think that their instincts are likely to be right and that their critics will be proved wrong. Europeans scorned Ronald Reagan, and the United Nations feared him, but in the end the Soviet Union was gone. So for reasons of personal, political, and intellectual history, it is understandable that members of this Administration could proceed down one path in defiance of mounting evidence of its perils. The Democrats had similar destructive self-confidence in the 1960s, when they did their most grandiose Great Society thinking.

The third factor is the nature of the President himself. Leadership is always a balance between making large choices and being aware of details. George W. Bush has an obvious preference for large choices. This gave him his chance for greatness after the September 11 attacks. But his lack of curiosity about significant details may be his fatal weakness. When the decisions of the past eighteen months are assessed and judged, the Administration will be found wanting for its carelessness. Because of warnings it chose to ignore, it squandered American prestige, fortune, and lives.