By His Own Rules

At the beginning of 2003, the Defense Department received from Bush formal authority for all postcombat operations in Iraq. It was a responsibility that Rumsfeld had requested in October and one even Powell agreed the Pentagon should have. It also represented a marked departure from the way the United States had tended to handle postwar efforts.

In the Balkans in the 1990s, the State Department had taken the postwar lead, while more recently in Afghanistan, responsibility for reconstruction and stabilization had been divided among several nations. Both approaches, though, had led to serious problems. In the Balkans, U.S. forces had been compelled to remain after their mission was complete because of a lag in civil reconstruction over which they had little control. In Afghanistan, the multinational model had proven largely ineffective.

With the Pentagon in charge, Rumsfeld figured civil authority could be reestablished promptly after the invasion, allowing for the near-term withdrawal of U.S. troops. But while the idea made theoretical sense, it overlooked a basic practical point: The Pentagon was woefully out of practice doing this kind of work. The U.S. military hadn’t directly managed an occupied country since World War II.

One perceived advantage of vesting control in the Pentagon was that it would ensure unity of command in Iraq after Saddam Hussein and his Baath Party were swept from power and before a new Iraqi government was established. In keeping with this principle of single leadership—a basic tenet of military operations—the Joint Staff had drafted a plan in the fall of 2002 for postwar Iraq to be administered by a new military headquarters, with a three-star general in charge of a staff of experts drawn from various U.S. government agencies. The thought was that, much as General Douglas MacArthur had controlled postwar Japan, a senior general would administer Iraq during the transition back to sovereignty.

But such a dominant military hand in postwar Iraq could prove politically problematic for the United States. Instead of opting for a single military-run command, Rumsfeld chose a split civilian-military approach. He ordered two distinct authorities: a civilian administrator to oversee the reconstruction and governance of the country, and a U.S. military commander responsible for security and for retraining the Iraqi military. This provided a civilian face on at least a major part of the postwar activity. But it had a serious downside. It would mean a system with divided command, and this, as things turned out, led to conflict and poor coordination between the civilian and military branches.

Further, the civilian element in this postwar planning structure got a late start. Rumsfeld had considered establishing something in October, but he was constrained by Bush, who worried that too many signs of U.S. war preparations could disturb efforts to achieve a diplomatic solution. Not until early January did Rumsfeld move to set up what became the Office of Reconstruction and Humanitarian Assistance, or ORHA. To run it initially, he hired Jay Garner, a retired three-star general who had earned high marks in an earlier postwar operation overseeing relief efforts in Kurdistan after the 1991 Persian Gulf War. Garner had also served on the space policy commission that Rumsfeld had chaired during the Clinton years.

Plans called for Garner to set up the office and eventually be succeeded by a more prominent political figure. Feith, who phoned Garner to pitch the job, told him his role wasn’t so much to generate his own plans as to coordinate and integrate previous efforts and to be ready to form the core of an organization that would administer Iraq after conflict ended.

Rumsfeld related to Garner’s enterprising spirit. “He immediately felt like they were simpatico,” John Craddock recalled. “Garner was talking the language that Rumsfeld wanted to hear, which is can-do—‘I’ve been over there before, and did it after the Gulf War with the Kurds, I’ll need this and that, we can pull it off.’ That was music to his ears because it was action.”

But Garner had little time to accomplish what proved to be a massive assignment. Two months after the announcement of ORHA’s creation, the war started. Garner’s orders had called for him to draft plans to deliver food and other emergency assistance and to restore electricity and other basic services. His office was also supposed to be ready to help reshape the Iraqi military, safeguard Iraq’s infrastructure, and dismantle whatever stashes of weapons of mass destruction might be found. And Garner was to coordinate his activities with the United Nations, nongovernmental relief groups, and Iraqi exiles.

Complicating matters, the staff of ORHA became a source of serious friction between the Pentagon and the State Department, with Garner initially prohibited from accepting the participation of some diplomats offered by State. When Garner sought Rumsfeld’s approval of a proposed list of senior U.S. advisers to key Iraqi ministries, the secretary crossed off eight or nine names that had been sent over from State and asked that several candidates be recommended for each position so that he could choose. Powell, incensed, threatened to hold back all State Department personnel from the effort. Rumsfeld relented, allowing participation by a significant number of State Department officials in ORHA. Years later, though, Powell told Army historians that the planning effort “would have been better served” if the Pentagon “had asked for more help from people outside.”

Tom Warrick, who had directed the State Department’s Future of Iraq Project, was notably kept off the team. Cheney’s office blacklisted Warrick because he had criticized the Pentagon during a meeting with Iraqi exiles in Michigan, warning them against working with Wolfowitz. Rumsfeld told Powell that work on postwar planning had to be done by those who were truly committed to the effort and not those who had written or said things that were not supportive.

The Army’s study of the postwar planning effort, published in 2008, concluded that ORHA’s work “appears to have suffered from this lack of interagency support.” In any case, neither ORHA’s mission nor its relationship with CENTCOM was ever well defined, nor was ORHA given resources adequate for the planning and oversight it was asked to do. ORHA deployed 151 staff members to Kuwait on March 16, only half the number Garner had estimated the organization needed. Its planning had focused on preparing for the four potential crises it considered most likely: oil field fires, large numbers of refugees, food shortages, and outbreaks of epidemics. None of these problems emerged once the fighting started.

Weeks before the war began, James Schlesinger, who had remained a prominent figure in national security circles after being removed by Ford as defense secretary and replaced by Rumsfeld, warned Rumsfeld that planning for the postwar period had not been given its due. Schlesinger served on the Defense Policy Board, which afforded him a window on Pentagon developments. In a private conversation with Rumsfeld, Schlesinger advised that Iraqi society, with its rival groups of Shiites, Sunnis, and Kurds, could prove very difficult to manage, yet many U.S. troops seemed clueless about what they were getting into. Afterward, Schlesinger wasn’t sure how much of an impression his warning had made, although Rumsfeld did ask him to raise his concerns with Tommy Franks and then phoned Franks to say Schlesinger was coming. When Schlesinger met in Tampa with the general and the CENTCOM staff, he received assurances that everything had been thought through.

Others also pointed to serious planning gaps. Numerous studies both inside and outside the government raised concerns about the potential for postwar difficulties and about the need to keep significant numbers of troops in Iraq during the crucial stabilization period. The conclusions of the State Department’s Future of Iraq Project, handed over to the Pentagon at the end of 2002, included the possibility of widespread looting and lawlessness and emphasized the need to quickly repair Iraq’s infrastructure. In December, a joint study by the Council on Foreign Relations and the James A. Baker III Institute for Public Policy at Rice University warned of possible anarchy and of the need for the U.S. military to quickly turn to humanitarian efforts and law enforcement.

The Center for Strategic and International Studies issued a report in January 2003 noting that the success of any war with Iraq “will be judged more by the commitment to rebuilding Iraq after a conflict than by the military phase of the war itself.” The report argued that postwar planning efforts conducted to that point had been incomplete and insufficient. And a paper by the U.S. Army War College in February cautioned that “the possibility of the United States winning the war and losing the peace in Iraq is real and serious.”

Civilian officials in the Pentagon who were focused on postwar peacekeeping considerations were increasingly anxious about what might unfold in Iraq. Joseph Collins, a retired Army colonel who headed the Pentagon’s Office of Peacekeeping and Humanitarian Affairs, feared that too few troops were being sent to adequately maintain supply lines for an invading force, and he also foresaw a more costly and problematic occupation than CENTCOM had anticipated. Chris Lamb, another official in the policy branch with experience in constabulary forces in Bosnia, warned that the U.S. military could undermine a war victory in Iraq if it failed to take care of law and order after the invasion. A fourteen-page memo that he wrote with Feith expressed concern that CENTCOM had not given high enough priority to providing an adequate policing force.

Others, like Robert Perito, an expert on peacekeeping operations at the United States Institute for Peace, a government-financed research center, and David Kay, a deputy to Garner, were urging a sizable international constabulary force. But their calls went unheeded. To the extent that CENTCOM planners had addressed the issue, they assumed that elements of the Iraqi army could be enlisted after the war to help restore order and alleviate some of the burden on U.S. troops.

According to John Craddock, Rumsfeld eventually recognized that CENTCOM “was behind the power curve” in preparing a postwar plan, having been focused on the combat phase and having been unable to work well with Feith’s policy shop. In the final weeks of planning, Rumsfeld began asking more questions about the state of postwar preparations. “I’ll say to Rumsfeld’s credit that he did push to get postwar planning done,” recalled George Casey Jr., then a three-star general directing the Joint Staff. “It was very hard to get anybody to focus on the postwar part of it. Rumsfeld used to say, ‘I’ve had twenty-seven briefings on the war but I’ve had one on the postwar. What’s going on?’”

But Rumsfeld and his senior staff continued to operate on the assumption that U.S. forces could avoid a prolonged occupation. Their guiding notion was that American troops would help Iraqis get back on their feet, not bring the country back to normal themselves. The term most often used to describe this approach was “enabling.”

In a speech on February 14 titled “Beyond Nation-Building,” Rumsfeld drove this point home. At a black-tie gala in New York honoring the armed forces, he offered the Balkans as a model of a postwar policy gone wrong. The Clinton administration’s intervention in Kosovo and the lengthy deployment of U.S. and allied peacekeeping forces there had led, Rumsfeld said, to a “culture of dependence” that had made it hard for the Kosovars to stand on their own feet. By contrast, in Afghanistan the Bush administration had resisted the temptation to mount a major nation-building effort or to deploy a large peacekeeping force; instead, it was helping the Afghans rebuild their own country.

Iraq, Rumsfeld said, might be easier to restore than Afghanistan in several respects. Iraq was rich in oil, and the Bush administration had more time to prepare the postwar plan. He touted the services of Garner’s postwar planning office, which he said had been established “to think through problems and coordinate the efforts of coalition countries and U.S. government agencies.” Franks, he added, had also “been working hard on this for many months” in an interagency process.

Rumsfeld’s references to Garner and Franks both were misleading. Garner’s operation had only just begun, and Franks had paid insufficient attention to postwar planning. But the thrust of Rumsfeld’s speech made clear his view that it would be possible to invade Iraq and then manage the aftermath in a way that would allow Iraq to put itself back together while the United States avoided a long, expensive, difficult nation-building effort like the one the Clinton administration had taken on in the Balkans.

The war strategy that CENTCOM devised and Rumsfeld approved at the end of 2002 envisioned that the invasion would begin before all forces had arrived in the region, so it depended on the remaining forces continuing to flow. That would be critical to ensure the protection of vulnerable supply lines and to sustain combat forces as the fight advanced to Baghdad.

But Rumsfeld’s relentless scrutiny of the Army’s troop deployment list continued into 2003 and raised worries among commanders that they would be denied adequate forces to conduct the fight. Rumsfeld, anticipating the possibility that Saddam’s regime could collapse quickly, asked Franks to identify points at which the flow of reinforcements could be stopped—“off-ramping” them, in Pentagon parlance. David McKiernan, the land commander, was reluctant to see planned reinforcements canceled even if the regime collapsed, figuring all the troops called for in the plan could be necessary for the postwar period.

Eric Shinseki was increasingly concerned as well. He had worried for months that the leaner invasion strategy being assembled by CENTCOM did not adequately provide for the considerable logistical demands of pushing to Baghdad. In late January, called to the White House with the other chiefs to convey their views on the war plan, Shinseki offered a qualified assessment. He expressed some reservation about the number of forces that would be in place at the start of the invasion, and he stressed the need to keep reinforcements coming and the logistics lines protected. But he did not press his points and, seeking to give Bush a balanced assessment, declared the plan executable.

“It was the opinion of the chiefs, especially Shinseki, that we were very short on land power,” recalled John Jumper, the Air Force chief. “The worry was, we were setting up the situation, as the land forces progressed, they would be leaving black holes behind them because we didn’t have the people to stay behind and stabilize.”

The chiefs did tend to subscribe to the belief that the war would go quickly, Jumper said. But they, like others, worried that inadequate attention had been paid to managing the aftermath. “We had long talks with Tommy Franks about phase four,” Jumper recalled, using the military term for the period after combat operations. “The chiefs thought that we should be working very hard right after the major combat operation on engineering things—getting the water flowing, getting the electricity, getting the schools and hospitals open—those things that would make the Iraqis think their lives were better. Franks assured us he had this well in hand, but in fact what we saw was that we really didn’t pursue that as an objective after the major combat operation.”

Aside from Shinseki, though, the chiefs did not voice such concerns in the meeting with Bush in January because, Jumper said, the discussion centered on the invasion, not the occupation afterward. “When we got asked our advice and whether we were comfortable with the war plan, it was almost entirely about the major combat operation part.”

The question of the extent to which the chiefs were consulted, and whether Rumsfeld was open to hearing them, would later dog the secretary. He insisted publicly that their views had been solicited, and in memos he wrote for his files, he maintained that the service leaders had put forward no objections. In one memo, for instance, dated September 29, 2004—a year and a half after the invasion—he noted that he and Myers had “met repeatedly during that period with the Chiefs and the Combatant Commander.” Neither he nor Myers, he wrote, could recall Myers “ever raising any differences of views of any of the Chiefs with POTUS [the acronym for president of the United States] on Iraq. The reason no differences were raised is, very simply, because no differences existed. Therefore, any suggestion that the President overrode his military advice is not correct.”

On February 25, Shinseki and his fellow chiefs appeared before the Senate Armed Services Committee, which was examining the administration’s movements toward war. During the hearing, Senator Carl Levin, the senior Democrat on the panel, asked Shinseki how many troops would be needed to control Iraq after Saddam was ousted. Shinseki estimated “something on the order of several hundred thousand soldiers,” adding, “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 a significant ground-force presence to maintain a safe and secure environment, to ensure that people are fed, that water is distributed, all the normal responsibilities that go along with administering a situation like this.”

His estimate drew in large part on his experience as a commander in postwar Bosnia, where the United States had sent fifty thousand troops to quiet five million people, a population one-fifth that of Iraq. The suggestion that the Pentagon was underestimating the difficulty of the task ahead in Iraq and providing an insufficient number of forces made headlines. The next day, Army secretary Tom White received an early-morning phone call from Wolfowitz, who complained that Shinseki was wrong and had no business offering an estimate, since the Army chief was not leading the operation. White argued that Shinseki, in providing his professional assessment, had simply done what Congress had asked him to do.

Two days after Shinseki’s testimony, Wolfowitz, in congressional testimony of his own, called the general’s estimate of several hundred thousand troops “wildly off the mark.” He said he knew of no historical reason to believe that free Iraqis would fight against one another or the United States. “I am reasonably certain that they will greet us as liberators, and that will help us to keep requirements down,” he said.

Moreover, Wolfowitz added, the Pentagon anticipated that many countries then objecting to the use of force would be eager to help once Saddam was gone. “There is simply no reason to assume that the United States will or should supply” the bulk of the military forces needed for postwar stabilization, he said. “I would expect that even countries like France will have a strong interest in assisting Iraq’s reconstruction.”

Rumsfeld also weighed in, making it clear at a Pentagon news conference on February 28 that he disagreed with Shinseki’s estimate, although because Wolfowitz had been so blunt, the secretary could afford to sound less so. “My personal view is that it will prove to be high,” he said of the Army’s chief ’s number.

In fact, Shinseki’s figure was only a little above the total number of forces in CENTCOM’s own plan, once all the reinforcements had arrived in the war zone. Indeed, in giving his estimate, Shinseki referred to “what’s been mobilized to this point” and equated that to what in his view would be required. The CENTCOM plan projected that while the attack would begin with only a portion of the force, as many as 250,000 troops would be in Iraq by the time Saddam was defeated and the United States began to stabilize Iraq.

Nonetheless, the sharp official rebuke dealt Shinseki had an intimidating effect on the rest of the military’s officer corps. It sent a chilling signal to the uniformed leadership about how that kind of military judgment was going to be valued, and it served to silence critics just at a moment when it would have been helpful to have them. “Anecdotal evidence,” a RAND study in 2008 concluded, “suggests that other Army general officers shared General Shinseki’s main concern, namely, that more troops would be needed immediately following combat with Iraqi forces than would be required to defeat those same forces; however, none spoke up publicly before the war.”

With war fast approaching, the Bush administration was still trying to rally as much international support as possible for an invasion force. Yet at times Rumsfeld managed instead to rile U.S. allies. Responding to a Dutch reporter’s question on January 22 about European opposition to the use of force against Iraq, Rumsfeld said the resistance had been centered in Germany and France, which he called “old Europe” and contrasted with the “new members” of “NATO Europe,” new members made up in large part of the formerly communist nations of eastern Europe. “They’re not with France and Germany on this; they’re with the United States,” he said.

The remark generated a flurry of news stories that cast Rumsfeld as the antidiplomat. In later explanations, Rumsfeld said that he had been thinking at the time about how NATO had grown from fifteen members to twenty-six with the addition of the eastern European nations, and that he had meant to say old and new NATO. He never repeated the quip, nor did he ever apologize for it.

The White House claimed that forty-nine countries were part of a “coalition of the willing” in support of military action, but most had promised little more than permission to put their names on the list. Rumsfeld seemed intent on making it clear that the United States could still go it alone if necessary. A week before the invasion, he suggested that even assistance from Britain—the ally that had provided the most support—was superfluous, telling a Pentagon news conference that Britain’s role was “unclear” and that the invasion plan could proceed without British troops.

The defense secretary’s offhand remark, made as Prime Minister Tony Blair was insisting to his own Parliament that Britain’s support was crucial to the effort, sent London reeling and demanding an explanation. The Pentagon quickly issued a “clarification”; Rumsfeld said he had “no doubt” of Britain’s “full support” for the effort to disarm Saddam Hussein. But the incident reinforced a widespread impression that the Pentagon simply was not interested in coalition warfare.

The most sensitive alliance issue involved Turkey. While the bulk of the U.S. invasion force expected to enter Iraq from the south through Kuwait, the war plan also called for the U.S. 4th Infantry Division to sweep in from the north, which would require passage through Turkey. By attacking from both the north and the south, U.S. commanders hoped to compel Saddam’s regime to divide its attention and resources. A northern entry would, moreover, speed the arrival of U.S. forces to safeguard the oil fields near Kirkuk and to prevent ethnic fighting between the Kurds and the Arabs.

Turkey had allowed the United States to conduct air strikes from its bases during the 1991 Persian Gulf War, but the economic consequences of that war for Turkey had been severe, and subsequent U.S. assistance had fallen short of what the country had anticipated. The Turkish public was overwhelmingly opposed to another war with Iraq, fearing that it would facilitate the breakup of Iraq and lead to the establishment of an independent Kurdish state, which could fan secessionist tendencies among Turkish Kurds.

Negotiations for arranging the U.S. use of Turkish territory, begun in the summer of 2002, had been difficult, with the Turks placing conditions on the number of U.S. troops and on how they could be transported across Turkish soil and demanding compensation, financial and otherwise. It didn’t help matters that the Pentagon’s precise needs kept changing.

Turkey eventually joined the long list of internal issues dividing the administration. State Department officials worried that defense officials were oblivious to the challenges they were posing to Turkey’s fragile democracy. Rumsfeld, along with Cheney, suspected that the State Department wasn’t trying hard enough to win Turkey’s agreement. Powell was later criticized for not having made a personal trip to Ankara. Rumsfeld never went either, but that was by design. “I remember one of the senior guys at State saying to me several times, ‘God, we can’t send Rumsfeld there, he’ll scare the heck out of the Turks,’” recalled Eric Edelman, a career diplomat then on Cheney’s staff.

On March 1, the Turkish parliament rejected a measure that would have permitted U.S. troops to establish a northern front, forcing a major change in the war plan. To maintain the impression that a northern attack might still occur, Franks kept the 4th Infantry Division floating off the coast of Turkey for a while longer, then brought it around to Kuwait, enabling it to enter Iraq from the south.

Rumsfeld has repeatedly cited the absence of a northern front as his major regret about how the invasion plan played out. He has argued that the delay in getting U.S. troops into Sunni strongholds north of Baghdad, including such cities as Tikrit, Saddam’s hometown, hampered efforts to root out Baathists. Certainly, the loss of that part of the plan cost the U.S. operation something in terms of surprise and perhaps spared some Iraqi forces. But just how many more Saddam loyalists would have been killed or captured and what difference it would have made to the subsequent rise of an insurgency in Iraq remains questionable. The U.S. forces surging up from the south did not stop to do much rooting out there either. Their essential objective was to get to Baghdad and topple the regime, even if it meant bypassing southern cities and towns in the process.

In any case, Rumsfeld’s comments about the refusal of the Turks to open the way for the 4th Infantry Division angered Turkish authorities. Edelman, who subsequently left Cheney’s staff to become the U.S. ambassador to Turkey in mid-2003, tried unsuccessfully to persuade Rumsfeld to stop complaining about the absence of a northern front.

“The Turks say that’s just the Americans blaming us for their multitude of failures,” Edelman said in an interview. “Truth be known, had the division gone in, Saddam’s regime might have crumbled a little quicker than it did, but it did crumble in three weeks anyway. I suppose we might have killed more of the people in the Republican Guard divisions, some of whom instead may have gone on to become insurgents. But no one will ever be able to establish frankly what contributed more to all this. I think it’s a not very productive line of discussion.”

In early March, Rumsfeld met with some of his senior subordinates—Wolfowitz, Myers, Pace, Di Rita, Craddock, Feith, and Ryan Henry—to survey their views on how long the war was likely to last. Several declined to answer at first because Rumsfeld had drilled his staff never to forecast about such operations, often describing the future as essentially unknowable. All remembered how Army general John Shalikashvili, as chairman of the Joint Chiefs of Staff under Clinton, had predicted just ahead of the Bosnia conflict that U.S. forces would be out within a year. As the U.S. presence there persisted for years, Shalikashvili was ridiculed for having misjudged so badly.

Nudged by Rumsfeld to hazard guesses this time, his aides offered estimates ranging from seven to thirty days, reflecting a good deal of optimism about the plan. But Rumsfeld refused to play himself. He told the others that they should be open to the possibility that the war could drag on, noting that the enemy also had a big say in the outcome.

In the early morning, Iraq time, of March 20, four bunker-busting bombs dropped from two U.S. Air Force F-117s and a volley of more than three dozen Navy cruise missiles struck Dora Farms, a compound in a palm grove on the Tigris River outside Baghdad where Saddam and his two sons were thought to be hiding. The Iraqi leader and his family were not there, but the war had begun.

At dawn the next day, U.S. ground forces moved into Iraq, crossing twenty-four hours earlier than planned out of concern that the Iraqis would be sabotaging the oil fields. The total U.S.-led invasion force consisted of fewer than three Army divisions plus a large Marine division and a British division—in all numbering about 145,000 troops. Hours later, U.S. and coalition aircraft and ships launched a furious barrage of bombs and missiles, blasting air-defense sites, command centers, communication links, Baath Party headquarters, Republican Guard facilities, and other sites. The aim, officials said, was to achieve “shock and awe” and deliver a crippling blow to Saddam’s regime.

Iraq’s regular forces proved little problem for U.S. troops, with a majority of the Iraqi army abandoning positions and seeming to dissolve. But to the surprise of American commanders, thousands of irregular fighters called fedayeen put up determined resistance, launching guerrilla-style attacks on rear supply convoys of the advancing U.S. and British forces. While forecast as a threat, the fedayeen had not been expected to be so ferocious and tenacious.

Sandstorms and mud rains enveloped Iraq on March 24 and lasted three days. The foul weather grounded helicopters and stopped many of the invading troops in their tracks. Momentarily stalled, U.S. officers expressed some frustration. “The enemy we’re fighting is a bit different than the one we war-gamed against because of these paramilitary forces,” Lieutenant General William Scott Wallace, who was leading the advancing Army forces, observed at the end of the first week. “We knew they were here, but we did not know how they would fight.” At the Pentagon, Rumsfeld was not pleased with Wallace’s remarks and the doubts raised about the strategy.

With the invasion appearing to falter, a chorus of critics grew louder. They included some prominent former Army commanders from the last war with Iraq. Their critique centered on the size of the invading force, which they said was not large enough to guard ever-lengthening supply lines, cope with the paramilitary units in the south, establish control throughout Iraq, and prepare for an assault on Baghdad.

“Their assumptions were wrong,” said retired general Barry M. Mc-Caffrey, who led the 24th Mechanized Division in the 1991 war and had become a frequent television commentator. “There is a view that the nature of warfare has fundamentally changed, that numbers don’t count, that armor and artillery don’t count. They went into battle with a plan that put a huge air and sea force into action with an unbalanced ground combat force.”

Retired general Ronald Griffith, the commander of the 1st Armored Division during the 1991 war, declared that if it had been up to him, he would have deployed additional armored divisions and more artillery. Others said the war plan had taken significant risks by leaving key units in the United States and Germany at the start. That resulted in an invasion force that was too small and too strung out, underprotected, and undersupplied.

Criticism of the war plan, particularly from the retired military community, angered Rumsfeld. He and his aides tended to perceive much of it as part of a larger fight over the future direction of the military. “It was playing internal politics out in the press during the war when American lives were at stake,” said Ryan Henry.

Henry recalled a small staff meeting during which Rumsfeld grew enraged over some of the remarks that had come from retired officers. “He just couldn’t hold it in any more, and he stood up and said to Myers, ‘It’s your guys, it’s your community, those retired people. This isn’t fair to the country; it isn’t fair to the boys fighting over there. What are we going to do about it?’”

Myers did not remember that meeting several years later when asked about it. But according to Henry, it sparked an unusual public outburst by Myers shortly afterward. During an April 1 Pentagon news conference, the general suddenly appeared uncharacteristically fierce in responding to the attacks on Rumsfeld and the war plan.

A journalist had asked about criticism that the number of U.S. forces in Iraq were insufficient. As Rumsfeld started to comment, Myers stepped forward. He called the complaints “bogus,” said that those who were spreading them were misinformed, and warned that such talk during combat seriously undermined the war effort.

“I’ve been in this process every step of the way,” Myers asserted. He said that Franks had gotten everything he wanted, and he defended the plan as designed to coordinate with sensitive U.N. diplomacy and also maintain a certain element of “tactical surprise.” He said that all the chiefs and all the component commanders under Franks had endorsed the plan. Rapping on the lectern with a clenched fist, he angrily mocked the armchair analysis on television as “a great sport here inside the Beltway.”

He was as emphatic and demonstrative as the normally unflappable Myers had ever been at a news conference during a year and a half of briefings on terrorism and war. Asked about his performance several years later, Myers said he considered it one of his most important and described his remarks as “premeditated.” “I couldn’t take it any longer,” Myers said, recalling his feelings at the time. “Because it’s not the secretary’s plan, it’s the military’s plan, and the military ought to speak out.”

Rumsfeld also weighed in against his critics, insisting to the Pentagon press and several television interviewers that he was not the author of the war strategy. Everyone had agreed that the old plan for invading Iraq had been flawed, Rumsfeld argued. The new plan, he declared, had been produced by Franks. “I keep getting credit for it in the press, but the truth is, I would be happy to take credit for it but I can’t,” Rumsfeld said. “It was not my plan. It was General Franks’s plan, and it was a plan that evolved over a sustained period of time.”

He seemed irritated when it was suggested that in saying it was Franks’s plan, he was distancing himself from it. “Goodness gracious!” Rumsfeld exclaimed. But the downplaying of his involvement was certainly belied by the facts. Politically, he may have felt reason to do so, seeing some advantage for the administration’s civilian leadership to be able to say that they were just implementing the military’s plan. Personally, such posturing afforded him a way of avoiding responsibility and placing it fully on his commanders should that become expedient. More probably, Rumsfeld, who genuinely seemed to view his role in the planning as more contributor than architect, was simply trying to ensure that Franks received sufficient credit.

For the record, Rumsfeld sent Myers a private memo later that day that sought to spell out his view of the war-planning process and also to address criticism of his handling of the deployments. “There has been no small number of articles and stories in the media over the past week alleging that I have imposed an unworkable, ‘high tech’ war plan on the military, destroyed the ‘carefully crafted’ time phased force deployment plan (TPFD), disapproved and ‘delayed proposals’ from Gen. Franks for forces to flow, and cut in half the forces Franks requested, with the result that U.S. and coalition forces are being killed unnecessarily and the U.S. is losing the war—all of this after 10 or 12 days of war,” Rumsfeld wrote in the memo. “The articles and breathless statements on television by retired Army generals suggest that the TPFD was a brilliant piece of work and that Gen. Shinseki ‘wisely’ advised the president that the plan was bad because the forces were inadequate and not heavy enough and that he and they are now being proven correct. I don’t remember the past several months quite that way.”

Discussing the lengthy process that led to development of the war plan, he insisted the responsibility for it rested with Franks. “Gen. Franks received a lot of advice, some of which he took, some of which, properly, he did not,” Rumsfeld wrote. “Press reports have repeatedly said it was ‘Rumsfeld’s war plan.’ As much as I would be delighted to take credit for the war plan, because I believe it is excellent, the fact is I don’t remember imposing any of my opinions on Gen. Franks with respect to the plan. We talked many, many times, to be sure. I raised many questions, and I energetically offered my views. Sometimes I suggested thoughts on particular aspects of the plan. I helped connect Tom’s work to the goal of having a plan that complemented and did not compete with the President’s diplomacy in the UN.”

Moving on to the controversy over the TPFD, Rumsfeld recalled cautioning against preparing a single, giant deployment order—the MOAD—in the fall of 2002 in light of then-ongoing diplomatic efforts. “Instead, we sought a steady flow of forces over the period,” he wrote. “That decision was made early enough to preclude any adverse delay on the total forces that arrived by D-day. As to my dismembering the TPFD, my recollection is that it was Gen. Franks who decided he wanted to accelerate the flow of some TPFD forces, not me. He recommended that certain elements be pulled forward and deployed in advance of the TPFD flow.”

Rumsfeld noted that he had refused to sign only two deployment orders: One was the MOAD, and the other involved a Puerto Rican National Guard unit slated to go to Germany for protection of U.S. forces there. “I suggested we ask the Germans to provide the force protection,” Rumsfeld wrote. “We did so, and the Germans agreed.”

The secretary told of being presented with deployment orders with as few as five days between notification and activation. He had been surprised by the orders coming to him so late and said Myers and Pace had also been surprised. The Joint Staff’s senior operations officer was instructed to avoid such minimal notification for reservists. “I had never been told that was going on until I attended an Army generals conference and was told that the Army average was five days,” Rumsfeld wrote. “I was shocked. I should have been told the system was that screwed up.”

He also noted a meeting in the Tank in the fall of 2002 at which a plan for reserve mobilization was discussed. The plan, he said, would have led to as many or more reserves being called up as were activated for the 1991 Persian Gulf War. The proposal had concerned him, as well as Wolfowitz, Myers, and Pace, all of whom, Rumsfeld said, saw it as providing too little flexibility and as being likely to end up unnecessarily disrupting the lives of many reservists right before the holiday.

“Further, we routinely asked Gen. Franks if he had all that he needed,” Rumsfeld wrote. “In every instance, Gen. Franks said he had what he needed.” Additionally, on three or four occasions Bush asked Myers, Pace, and the chiefs whether they agreed with the plan, and each said he did, Rumsfeld said. And, he noted, in a meeting with Myers, Pace, Franks, and each of CENTCOM’s component commanders just before the invasion, Bush asked whether they had everything they needed to win, and each said yes.

Rumsfeld concluded with a paragraph saying he found the second-guessing by Army generals revealing. The seeming pervasiveness of anonymous negative comments about the plan did not augur well for future U.S. Army leadership, he said, adding that general officer promotions would have to be examined closely to ensure higher-quality leadership.

That last comment was a stark expression of the conflict in which Rumsfeld found himself with the Army. He did not elaborate on how he thought those senior officers who were critical of his own leadership could be weeded out in the selection process. But his thought of using the process to create a more supportive and amenable group of Army commanders is evident.

In the end, Iraqi forces proved no match for the U.S. military juggernaut marching toward Baghdad, and after several days of battles with Saddam’s tougher Republican Guard forces on the outskirts of the capital, American troops reached the city center on April 9, nearly three weeks after they had crossed the Kuwaiti border. Saddam himself managed to escape into hiding, and scattered fighting continued around the country, but on April 15, coalition forces secured Tikrit, bringing Iraq effectively under U.S. control. American casualties had been relatively light, with 139 dead and 542 wounded. Although ground troops went into battle with special suits to protect them from Iraqi biological and chemical attacks, no weapons of mass destruction were deployed.

On April 16, Franks visited Baghdad. He sent a message to Rumsfeld recommending that the United States declare a military victory in Iraq and announce that it was transitioning to postwar operations. He also instructed his officers to be prepared to reduce forces rapidly. In line with the prewar planning and the general euphoria at the rapid crumbling of the Saddam regime, Franks expected a very limited role for U.S. ground forces in Iraq.

But the situation on the ground was hardly calm. After several days of euphoria following the arrival of U.S. troops in Baghdad, the Iraqi capital erupted in a frenzy of crime and looting. American soldiers, without orders, manpower, or a plan to intervene, stood by while government facilities were stripped of everything from furniture and fine art to plumbing and electrical wiring. Because of the speed of the American victory, fewer Army divisions than planned were on hand in Iraq to help stabilize the situation. And the 3rd Infantry Division that had led the U.S. Army’s charge into Baghdad was equipped not with the military police and civil affairs units useful in policing but with tanks and armored fighting vehicles meant for combat.

“Never, from the first day that we ever started planning this until we got to Baghdad, in all the processes, rehearsals—nobody ever mentioned the word ‘looter,’” Major General Buford “Buff” Blount III, the division’s commander, told a journalist later. “I mean, it was just never, ever, ever mentioned. Our focus was on fighting the war.”

Moreover, Blount and other commanders were reluctant to issue shoot-to-kill orders to quell the looting, seeing the mayhem as probably a short-term indulgence in vengeance after years of mistreatment under Saddam. Similarly, White House and Pentagon officials in Washington, fending off rising criticism of inadequate postwar planning, described the destruction as a normal turn of events that would play itself out quickly. In a news conference on April 11, Rumsfeld minimized the images of looting that were being shown on television.

“It’s the same picture of some person walking out of some building with a vase, and you see it twenty times, and you think, ‘My goodness, were there that many vases?’” The press corps laughed. “Stuff happens! But in terms of what’s going on in that country, it is a fundamental misunderstanding to see those images over, and over, and over again of some boy walking out with a vase and say, ‘Oh, my goodness, you didn’t have a plan.’ That’s nonsense. They know what they’re doing, and they’re doing a terrific job. And it’s untidy, and 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.”

But the scale of the destruction that followed the fall of Baghdad went far beyond what the term “looting” normally conveys. Whereas the war itself had caused little damage to infrastructure because U.S. forces had relied heavily on precision weaponry, the looting destroyed almost three-quarters of the government ministries and many other structures and facilities.

When Garner and his team arrived in late April, they found that seventeen of twenty-three government ministries had been completely destroyed. Also damaged were most of the city’s hotels, department stores, museums, schools, universities, and hospitals, as well as Saddam’s palaces. A notable exception was the Ministry of Oil, which was heavily guarded by a company of U.S. Marines. In neighborhoods where U.S. forces were positioned, looters were deterred, but there were far too few American soldiers to make a notable difference.

Rumsfeld’s failure to grasp the scale of the destruction and its significance was compounded by his public effort to play it down. In the process, he not only undercut U.S. military efforts to deal with the violence but conveyed an image of U.S. insensitivity. “Some senior officials in Washington chuckled about a ‘new spirit of freedom’ that had suddenly sprouted . . . among ‘grateful,’ liberated Iraqis,” wrote Fred Ikle, who served as the Pentagon’s policy chief during the Reagan administration and who still had high standing in conservative circles. “America lost most of its prestige and respect in that episode. To pacify a conquered country, the victory’s prestige and dignity is absolutely critical.”

The violence grew worse, transforming from initially unorganized, individual activity into increasingly systematic, large-scale crime. McKiernan considered declaring martial law and issuing orders to shoot looters on sight. He told Pentagon investigators later that he hadn’t been able to bring himself to issue a shoot-to-kill order out of concern that innocent women and children might be harmed. But other accounts have portrayed the general as having also been dissuaded by top Washington officials.

Major General William Webster, the deputy ground commander, fielded queries from Iraqi leaders about the possibility that martial law might be declared. In an interview with Army historians, he remembered McKiernan responding, “The president and the secretary of defense have said that we will not declare martial law. We are not going to put our military in a position of enforcing Iraqi laws.”

Despite the unsettled situation in Iraq and evident shortfall in U.S. forces, Rumsfeld approved a recommendation from Franks to cancel the planned deployment of the 1st Cavalry Division, which had been scheduled to go as part of a reinforcement flow that also included the 1st Armored Division. In the weeks leading up to the war, Rumsfeld had indicated a desire to off-ramp some Army units if Iraqi forces were quickly defeated, and after the fall of Baghdad, he had questioned Franks as to whether both the 1st Cavalry and 1st Armored needed to deploy.

Initially, Franks wanted both divisions in Iraq to help extend the military’s reach into northern and western parts of the country and to start the process of reconstituting the Iraqi army. But after discussing the matter for several days with Rumsfeld as well as with Myers, Pace, and his own staff, Franks changed his mind and recommended fewer reinforcements. The 1st Armored Division was sent, but the deployment of the 1st Cavalry Division was officially canceled on April 21. General Jack Keane, the Army’s vice chief of staff, supported the decision.

Speaking to the American Enterprise Institute, where he was a resident scholar, and also to the Washington Post, Gingrich assailed the State Department as “ineffective and incoherent” and a “broken instrument of diplomacy.” He blamed the department for a list of U.S. diplomatic defeats, among them the inability to win a U.N. resolution authorizing force in Iraq, the failure to win Turkey’s acceptance of U.S. troops, and the ineffectual effort to stymie France’s campaign against the war. He criticized State Department officials for undue deference to the United Nations and for tolerance of terrorism in Syrian-occupied Lebanon. Contrasting State’s “pattern of diplomatic failure” with what he described as the successes of the Defense Department under Rumsfeld, Gingrich said the Pentagon had “delivered diplomatically and then the military delivered militarily.”

Gingrich’s remarks reflected deep frustration among conservatives with Powell’s State Department, which they saw as not sympathetic to the president’s policies. Because of Gingrich’s association with Rumsfeld, his assault was interpreted as reflecting an intensified struggle between the Pentagon and the State Department over the direction of U.S. foreign policy. Conservatives, suspicious of Powell from the start, had viewed him and his subordinates at State as repeatedly undercutting the president’s policies.

In response to Gingrich’s broadside, State Department officials defended their record. They cited the department’s efforts before the invasion of Iraq to secure foreign basing rights for U.S. troops, enlist allies in the military campaign, and find financial backing for rebuilding Iraq. They also let loose a biting quip about the former House leader, a quip that Powell and Armitage had drafted together. “It’s clear that Mr. Gingrich is off his meds and out of therapy,” Armitage declared.

In the immediate aftermath of the fall of Baghdad, U.S. authorities were on track to set up some kind of Iraqi interim government. Garner and his team organized a political conference in Nasiriyah with indigenous civilian leaders on April 15, and Ryan Crocker, a senior diplomat, and Zalmay Khalilzad, an NSC official designated as a presidential envoy, were dispatched to help find individuals interested in joining a new government.

On April 23, the Bush administration announced plans to have an appointed interim authority in place by early June and pledged to continue meeting with emerging Iraqi leaders and political organizations. ORHA began calling weekly meetings of Iraqi leaders and coalition forces. News reports described plans for the creation of a “transitional government” with power over “nonsensitive” government ministries such as education and health care, to be followed by the formation of a “provisional government” within six months to two years after the establishment of an interim authority.

But on the evening of April 24, only three days after Garner had reached Baghdad, Rumsfeld called him to say Bush planned to appoint L. Paul “Jerry” Bremer III, a terrorism expert and former career diplomat, as his permanent envoy to Iraq. For Rumsfeld, the move was all part of the original plan. He had expected before the war to put in place a civilian administrator with a political or diplomatic background and, by late April, had judged that the effort in Iraq required a diplomatic dimension outside Garner’s experience. Before settling on Bremer, Rumsfeld even mused in a meeting with top aides that he himself should be the one to take over in Iraq. “It was a recognition of how important he considered the next phase could be,” Steve Cambone recounted.

In some ways, Bremer was an odd choice. Although an acknowledged expert on terrorism and a career foreign service officer, he had little previous experience with Muslim society except for a stint in Afghanistan in the 1960s. He had never served in the military, spoke no Arabic, and had never worked in the Middle East or in a postwar occupation or reconstruction effort. Nor had he ever managed any large budget or organization.

Weighing heavily in his favor, though, at least in Rumsfeld’s mind, was his State Department experience. Rumsfeld told Feith that picking Bremer would make it easier for Defense and State officials to work together on Iraq. When Powell and Armitage were informed of Bremer’s selection, they were relieved. They considered Bremer, who had retired from the Foreign Service in 1989 after twenty-three years, someone they could talk to.

There had been complaints about Garner’s operation. State Department officials worried that Garner, who had been organizing meetings with Iraqi political figures even before he and his staff had shifted their headquarters from Kuwait to Baghdad, was moving too quickly to hold local elections and establish a new government amid the unanticipated chaos. At the Pentagon, some groused that Garner was not an effective manager and that ORHA was floundering. They were also critical of a number of State Department recruits on Garner’s staff. “The neocons were saying that the people the State Department had offered up were their B team, that they weren’t putting their very best out there, so that the effort would not succeed and they could then come in with their A team and say, ‘We’re the ones to be able to run it,’” Ryan Henry recounted. “That was a school of thought, and they were part of a chorus that was influencing the secretary, persuading him that things weren’t working out with Garner.”

When Bremer’s appointment was formally announced by Bush in early May, it was considered a hasty decision, made quickly after Bush and his advisers decided that both the perception and the reality of reconstruction efforts were deteriorating. Rumsfeld did not regard Bremer’s appointment as a rejection of Garner, and he had hoped that Garner would stay and work under Bremer, who was being given broader responsibilities as head of an occupation administration being called the Coalition Provisional Authority (CPA). But Garner, who thought he was on the verge of a political breakthrough in setting up an interim government, was frustrated by the timing of Bremer’s appointment and left.

Over the next few months, Bremer’s actions produced a major shift in the approach to the governing of postwar Iraq from the Bush administration’s original plans. ORHA had been conceived as a temporary organization designed to assist a new Iraqi government during a short transitional period. By contrast, the CPA possessed all the powers of an occupation authority, and Bremer exercised those powers to establish a longer-term and more complicated reconstruction effort than U.S. policy makers had anticipated.

Bremer’s moves have become a source of lingering dispute between his team and Rumsfeld’s. The Rumsfeld group contends that the envoy, without fully consulting Pentagon officials, abandoned the original plan to set up an Iraqi interim government and quickly pass sovereignty back to the Iraqis. Bremer, for his part, insists he was just carrying out Bush’s wishes.

Determining even the basic facts is complicated by an evident lack of communication and coordination among top officials in Washington during that period. A number of decisions appear to have been taken without the full involvement of principals. “The interagency process in Washington was not working, which explains a lot of the confusion,” Bremer said in an interview.

This breakdown in the process was evident at the outset of Bremer’s mission when, even before departing Washington, the new envoy managed to establish his own primacy over U.S. efforts in Iraq during a direct conversation with Bush, without discussion in any interagency meeting. Plans had called for Bremer to share responsibilities with Khalilzad, who on the ground in Iraq was still spearheading political arrangements for a new government. But in a meeting with Bush, Bremer insisted that Khalilzad be removed from the Iraq account, telling the president that he needed “full authority.” In Bremer’s view, it made simple sense that there be only one presidential envoy in a country. Bush agreed.

Khalilzad learned his mission had been canceled just minutes before Bremer’s appointment was announced on May 6. Powell was stunned when he was advised of the cancellation and called Rice, noting how damaging it would be to lose Khalilzad’s involvement. Powell described Khalilzad as “the only one there who knows what’s going on or can relate to those people in Iraq.” Rice replied that Bremer had made his own preeminence a condition for taking the job. Rumsfeld, too, professes not to have known that Khalilzad’s assignment was going to be canceled.

Arriving in Baghdad in mid-May, Bremer quickly dashed expectations raised by Garner and Khalilzad for an early transfer of authority when he informed Iraqi leaders that the creation of a new Iraqi government would have to wait. His first two decrees—one banning members of Saddam’s Baath Party from official positions, the other disbanding the Iraqi army—were meant to signal a clean break from the past, make it clear that the Baathists were not coming back, and begin the process of reestablishing political and military institutions with a fresh slate.

But the exclusionary nature of the decrees reversed initial plans that had counted on retaining some elements of the old order to get Iraq rapidly back up on its feet. With thousands of Baath Party members banned, many government institutions were immediately left without their top several levels of management, and in time, lower bureaucratic levels were also depleted of such critical personnel as engineers, physicians, and schoolteachers. With the Iraqi army disbanded, U.S. forces lacked a potential source for local security and reconstruction assistance while hundreds of thousands of unemployed men roamed the chaotic and dangerous streets. Most significantly, the decrees, which became Bremer’s most controversial actions, have since been blamed for creating a pool of disaffected and unemployed Sunni Arabs that later served to fuel the insurgency.

Bremer insists that the measures were coordinated with Rumsfeld and other Pentagon officials, and the record shows that the senior civilians at the Defense Department as well as Franks were advised. Feith has acknowledged backing both actions. Indeed, his office drafted the de-Baathification order, which defense officials regarded as a limited measure meant to target only the top slice of the party’s leadership and only about 1 percent of the party’s members. The problem, they say, came when the order was implemented too widely and aggressively by a committee of Iraqis chaired by Ahmed Chalabi. But the decree was issued over the strong objections of Garner and the CIA station chief in Baghdad, who warned the move would needlessly fuel anti-American sentiment and undercut national reconciliation efforts.

As for the disbanding of the army, Bremer argued that the action merely reflected facts on the ground, since Iraqi forces had essentially dissolved during the invasion. Senior officers close to Saddam had gone underground, conscripts had abandoned their units, and the military’s facilities had been ransacked and dismantled. It would better serve U.S. interests, the argument went, to create an entirely new Iraqi army because calling back the old Sunni-dominated army, first, might not succeed, and second, would certainly risk political problems with the Shia and the Kurds.

Bremer sent a draft of the disbanding order to Rumsfeld on May 9 and to Wolfowitz, Haynes, Feith, Franks, and Garner the next day. Walter Slocombe, the CPA’s director of security and defense, discussed the proposed order with Feith, and Bremer briefed Rumsfeld on it several times, forwarding a final draft to him for his approval on May 19. That same day, Rumsfeld sent Bremer a classified set of planning instructions on how to establish what was to be called the New Iraqi Corps, envisioned to grow to three divisions and forty thousand troops.

But while Rumsfeld and other senior Pentagon civilians had been consulted in advance about the disbanding of the old army, some key players were blindsided by the edict. Myers, Pace, and the military service chiefs have said they did not know it was coming. The decrees weren’t discussed in interagency meetings by deputies. And McKiernan as well as John Abizaid, who had served as deputy CENTCOM commander during the war and was preparing to replace Franks, considered the decision an abrupt and unwelcome departure from their previous planning.

McKiernan and the CIA had been working on reviving the Iraqi army, meeting with former commanders on ways to reconstitute some units. Garner’s team, too, was engaged in an effort to use the former army as a source of labor and had arranged for contractors to retrain some troops. His staff had collected the names of fifty thousand to seventy thousand military and police personnel who might be drawn back. Bremer and Slocombe were aware of U.S. military contacts with former officers, but did not view these officers as the nucleus of a new Iraqi command, merely as holdovers from the old force.

The poor coordination between the civilian and military commands in Baghdad was symptomatic of the problem created by the earlier decision to separate the two. The rushed move to disband the Iraqi military also reflected a failure to think through how postwar events might unfold. By eliminating the prospect of a ready pool of Iraqi troops to help restore order, Bremer exacerbated the security vacuum already created by Rumsfeld’s decision to reduce the flow of scheduled reinforcements. Now, not only was the number of outside troops coming into Iraq limited, but the number of Iraqi forces immediately available had dried up.

At the time, Abizaid was sharply critical of the relatively small size of the new army being planned. He persuaded Rumsfeld of the need to create an Iraqi Civil Defense Corps, consisting of ad hoc paramilitary units, that would complement the army and assist U.S. and allied forces in dealing with local security problems. These units gave U.S. operations an “Iraqi face” and soaked up unemployed young men to keep them out of the insurgency, but they proved of widely inconsistent abilities and never amounted to a national force capable of filling the security gap in Iraq. Eventually, the corps was converted into a national guard before it was finally folded into the regular army.

Over the next few months, Bremer moved farther away from the idea of arranging for the transitional Iraqi authority that Garner and Khalilzad had pursued and that Rumsfeld and his staff thought was still a main objective. Bush had endorsed the plan for an Iraqi Interim Authority at an NSC meeting shortly before the invasion on March 10. But Bremer contends that by the time he was asked to take the CPA job in late April, fresh doubts had arisen—at least among some top U.S. officials—about trying for a rapid transfer of sovereignty.

While senior Pentagon officials still favored ceding authority early to a small group of Iraqi exiles, State Department and CIA officials were arguing again for a longer-term effort to manage the deep divisions in Iraqi society. Bremer understood from his discussions with Bush and from NSC meetings that the president had tilted toward the take-your-time advocates and had come to favor a longer-term process to build support for democracy in Iraq.

“Sometime after March 10—I’ve been told by various participants it was early April—Bush decided the idea of a short occupation and a quick handover wasn’t going to work,” Bremer said. “But the problem was, I don’t think it was ever documented. No one ever put a piece of paper in front of him and said, ‘Here’s the new guidance,’ and so there was a lot of confusion.”

Rumsfeld himself appeared somewhat ambivalent. Although he believed a sooner transfer was better than a later one, he was not as eager as Wolfowitz, Feith, or some conservatives outside the Pentagon, such as Richard Perle, to see an immediate handover. Rumsfeld’s initial focus was on ensuring U.S. forces were free to root out terrorist elements and locate suspected stockpiles of weapons of mass destruction. He worried that establishing a new Iraqi governing entity too soon could interfere with these efforts.

Rumsfeld also anticipated a lengthy new political process ahead in Iraq. In memos in May to the NSC and to Feith, he cautioned that the transition to democracy in Iraq would not happen fast or easily and urged that elections not be rushed. Feith recalls Rumsfeld expressing concern to him about the Iraqis pressing for power before they had expanded their leadership council. Noting reports of a lack of discipline among Iraq’s emerging leaders, Rumsfeld cautioned as well against creating an Iraqi governing entity that might undermine Bremer’s ability to do his job.

Nonetheless, Rumsfeld expected a relatively early turnover of sovereignty to the Iraqis and assumed that Bush’s approval of the plan to establish an Iraqi Interim Authority as soon as possible after liberation remained in force. Bremer, for his part, has pointed to a paper trail indicating that he kept Rumsfeld apprised of the shift in approach and that Rumsfeld raised no objection.

On May 22, for instance, Bremer outlined to Rumsfeld a plan to move first to a constitution and then to elections. Bremer wrote several more memos to Rumsfeld and the Pentagon through the rest of the spring and summer and into the fall, reporting on talks with Iraqi officials and repeating plans for drafting a new constitution first, estimating that elections couldn’t be held for at least a year. He also spoke publicly about these plans—in a June 22 address to the World Economic Forum in Amman, Jordan, for instance, and in a July 23 appearance at the National Press Club in Washington.

But evidently neither Rumsfeld nor his senior staff realized that Bremer’s talk of putting off full sovereignty until after the elections meant there would be no sovereign control in the meantime. In an interview for this book, Rumsfeld said he has no recollection of Bremer giving indications in the summer of 2003 of an intention to put aside the Interim Iraqi Authority plan. Instead, he said he understood the creation in July 2003 of the Iraqi Governing Council, consisting of twenty-five Iraqis selected by the CPA through a nationwide search, to be the first phase toward establishing the interim authority. Bremer even called it an “interim administration,” although it was given far less power than the Interim Iraqi Authority had been intended to have.

Rumsfeld does allow for the possibility that Bremer, in separate discussions with the president, State Department officials, and NSC staff, may have received guidance, or what he interpreted to be guidance, in pursuing an approach for a longer occupation—or at a minimum to use his own judgment. But Rumsfeld asserts that Bremer did not receive such guidance from him.

Bremer described himself as “flabbergasted” by the claims that Pentagon officials were unaware of his plan. Citing his voluminous reports to Washington, he said not once did Bush, Rumsfeld, or any other U.S. official object to the CPA’s course. “If Rumsfeld says he doesn’t remember giving guidance, I guess silence is some of kind of guidance,” Bremer remarked. “It’s simply not credible. There is simply no case to be made that they didn’t know what I was doing.”

Even so, Rumsfeld and Feith maintain that only when an op-ed article by Bremer was published in the Washington Post on September 8 did they realize that the CPA leader was taking a different route and was not ready to hand over responsibility for governance of Iraq to an Iraqi leadership. In the article, Bremer outlined seven steps that he called Iraq’s “path to sovereignty.” They included such formidable and uncertain goals as a ratified constitution and nationwide general elections. The article also asserted that the CPA would stay in existence until the country had achieved all seven steps—which CPA officials had estimated privately could take two years.

Rumsfeld said the article had not been reviewed by Pentagon officials beforehand, nor had Bremer told him that he envisioned a two-year process before the CPA could be dissolved. In the weeks that followed, Rumsfeld, along with other administration officials, pressed for an earlier termination of the CPA, and Bremer ultimately concurred. Agreement was reached in November for the CPA’s term to end in June 2004.

Feith, Perle, and other conservatives have pointed to this turn toward an imposing, lengthy rule by the CPA as perhaps the most ill-fated decision of the postwar period. They blame it for the failure to achieve the early postwar peace they had envisioned under an exile-led government. In their view, the decision dashed any chance that the coalition forces would be received as liberators rather than as occupiers, and it hobbled the coalition’s response to the widening insurgency.

This rationale for what went wrong in Iraq is too simplistic an explanation—and suspect, coming from those who had so aggressively pushed for the invasion in the first place. Events in Iraq have provided no compelling evidence that a provisional government, quickly installed, would have thwarted the rise of an insurgency or, once confronted with one, would have taken more-effective action against it earlier. Bremer has continued to defend his approach as correct, saying that what it sacrificed in immediate Iraqi sovereignty was more than made up for by giving the Iraqis time to put in place an interim constitution and political structures and to organize politically to build a new democracy.

Rumsfeld, for his part, places much of the blame for veering off into a prolonged occupation authority on the deep split in views in Washington and on a lack of firm coordination over how best to proceed. “If you have six or eight hands on the steering wheel, no one’s got their hands on the steering wheel,” he said in an interview.

He described an ongoing “tension” between essentially “two groups of thinking” among those advising Bush. One group, he said, favored going into Iraq, removing Saddam, ensuring the country did not return to making weapons of mass destruction and engaging in terrorism, and then leaving. “That way of thinking argued for a relatively light footprint, a relatively short period of being involved, and a sensitivity to not being seen as too much of an occupying force,” Rumsfeld said.

The other view, Rumsfeld noted, basically argued, “We’ve got to make darn sure this thing works out, we’ve got to make darn sure that they end up with a democratic government and a government that is at peace with its neighbors.” Such thinking, he added, can lead to a prolonged occupation with a “very American” face. “You start making a lot of decisions for them and you start having a larger group doing it and you run the risk of creating a dependency and anger and opposition.”

“There’s no roadmap, there’s no guidebook that says this is what you should do,” Rumsfeld observed. He personally favored the smaller-footprint, lighter-handed, shorter-duration approach. But he hinted at some doubt in his own mind about whether this would have worked any better than what resulted with the CPA.

Two weeks after Baghdad fell, Tom White, the Army secretary, was told to report to Rumsfeld’s office for a late-afternoon meeting. When White walked in, Rumsfeld came right to the point. “I want to make a change,” he said.

White’s earlier resistance to terminating the Crusader program and some of his other views on the Army’s future had become an unpardonable offense for the defense secretary, who continued to believe the service was too enamored of large forces and slow change. White told Rumsfeld he would write a short resignation letter that day. The Army secretary had planned to stay through the end of Shinseki’s term, but a week later Wolfowitz called to say that was no longer acceptable, and White left on May 9.

White was the only Pentagon official Rumsfeld ever fired during his time as defense secretary under Bush. Rumsfeld lost confidence in others and came close to dismissing them but settled for simply freezing them out in hopes they would leave of their own accord. At the same time, in deciding promotions to coveted joint command positions, he showed little hesitancy about blocking those uniformed officers who got on his bad side. That was the case with two senior Army commanders who had helped lead U.S. forces into Iraq: David McKiernan, who had responsibility for all ground troops, and William Scott Wallace, who headed the Army’s V Corps.

Wallace had upset Rumsfeld with his remarks to reporters, one week into the war, about the unpreparedness of U.S. forces to deal with the attacks they were facing from the Saddam fedayeen. McKiernan, in turn, had rubbed the secretary the wrong way during a visit Rumsfeld made to Baghdad shortly after firing White. Rumsfeld thought McKiernan was keeping his distance from him. On plane rides into and out of Iraq from Kuwait, the general chose not to sit near the secretary and talk about any issues, and in Baghdad, McKiernan shunned a request from Rumsfeld’s military assistant to introduce the secretary at a town hall meeting with troops, passing the task off to Wallace. As the secretary spoke, McKiernan stood on stage looking stern, his arms folded across his chest. Rumsfeld was angered by McKiernan’s behavior and complained to Craddock and others.

“Rumsfeld never forgot that, and it was a long time—a couple of years—before McKiernan recovered and received his fourth star,” recalled Staser Holcomb, the retired admiral who assisted in the selection of senior officers. “Rumsfeld does make some fairly quick judgments about people, and he made a judgment about McKiernan—that McKiernan was a grouch and resisting the secretary of defense and resisting civilian control.” Wallace, too, had to wait two years before receiving his fourth star. In Rumsfeld’s eyes, according to Holcomb, both Wallace and McKiernan were seen as loyalists to White and Shinseki, which was a big strike against them.

Army leaders had presumed that McKiernan, then the senior officer in Baghdad, would be picked to stay on and head the task force being set up in the Iraqi capital to oversee military occupation forces. But in May, Franks instructed McKiernan to leave Iraq along with the staff of his land-war command, which had helped plan and direct the invasion. Like other senior U.S. military and civilian authorities, Franks assumed the major fighting was over, and he wanted McKiernan’s command staff out of Iraq and returned to their normal role of support of land operations throughout the CENTCOM region.

A new headquarters was established to command the military forces in Iraq, and Lieutenant General Ricardo S. Sanchez was named to head it. Sanchez had led the 1st Armored Division into Iraq before being promoted and picked to succeed Wallace as the head of V Corps. It was this corps headquarters that was to serve as the nucleus of the newly established command known as Combined Joint Task Force 7, or CJTF-7. But the group had never expected to run a massive peacekeeping and reconstruction effort and was hampered from the start by personnel shortages and by a poorly coordinated transfer of responsibilities.

The belated decision to allow not only McKiernan but his whole headquarters of senior, seasoned officers to leave and to replace them with an ill-prepared team led by the Army’s newest three-star general was another of the major mistakes of the postwar period. “The lateness of this decision suggests the degree to which senior civilian and military leaders within DoD underestimated the challenges that would confront coalition military forces after the defeat of Iraqi forces,” RAND’s study of the war concluded.

Explaining the move several years later, Franks told military historians that he acted so quickly to redeploy McKiernan’s headquarters as a way of spurring the Pentagon to send another task force to work with ORHA. He said he had told top Pentagon officials what was needed and, further, that it was their responsibility to ensure the new headquarters was rapidly installed. “I thought it was sufficient to tell Don Rumsfeld and Dick Myers, ‘Here is what we are going to do in Iraq. Here is what we need in Iraq. We need a joint headquarters, a CJTF. You figure it out,’” he said. “So that is a task that John Abizaid and I very simply laid on Washington and said, ‘Figure it out. Do it fast.’”

But the effort didn’t quite work as planned, and Franks decided to make V Corps the senior headquarters in Iraq. That decision was taken over the objections of Keane, the Army’s vice chief, who had assumed that McKiernan’s headquarters would oversee the occupation in Iraq. Learning of the switch to Sanchez, Keane raised his concerns with Abizaid, who had been picked to succeed Franks as the head of Central Command. “I said, ‘Jesus Christ, John, this is a recipe for disaster,’” Keane told Army historians. “I was upset about it to say the least, but the decision had been made and it was a done deal.”

Rumsfeld had approved the promotion of Sanchez to lieutenant general, but he has denied responsibility for the decisions to remove McKiernan’s headquarters in the summer of 2003 and to make Sanchez the senior U.S. commander in Iraq. The changes, he insists, were never brought to his attention at the time.

In his memoir, Sanchez writes of two meetings in which Rumsfeld professed his noninvolvement. Hearing the claim the first time, in September 2004 shortly after Sanchez had left Baghdad, the general was incredulous. Rumsfeld also asserted he was never told that Sanchez’s headquarters had remained at less than half its stipulated manning levels during the year it was in operation.

“Why didn’t you tell somebody?” Rumsfeld asked. Sanchez replied that he not only frequently argued for more support but appealed to every leader who visited Iraq, and he said that “every senior leader in the Pentagon knew the status of CJTF-7.” When Sanchez later told Army leaders of this exchange with Rumsfeld, they contended that Rumsfeld had himself been the problem behind the staffing shortage.

Sanchez recounted another meeting with Rumsfeld, nineteen months later, in which the secretary again insisted he had been in the dark. Rumsfeld showed Sanchez a two-page memo he had written asserting he had not taken part in the decision to move McKiernan’s headquarters out and put Sanchez in charge of the entire mission. When Sanchez expressed his disbelief that Rumsfeld hadn’t known, the secretary grew agitated.

Sanchez concluded that this was a pattern on the secretary’s part of trying to deny involvement, assign responsibility to others, and put it all down in writing. “In essence, Rumsfeld was covering his rear,” Sanchez wrote. “He was setting up his chain of denials should his actions ever be questioned. And worse yet, in my mind, he was attempting to level all the blame on his generals.”

Given Rumsfeld’s reputation for hands-on management, it is indeed difficult to believe that he had no role in the selection of the general who would be commanding postwar operations in Iraq and the composition of his headquarters staff. Myers himself said in an interview that it was “possible but not likely” that Rumsfeld was not involved “because all those key personnel decisions we worked pretty carefully.” If the decision did somehow occur without Rumsfeld’s involvement, it would at least suggest a grave breakdown in communication and oversight.

A month after White’s departure, Shinseki retired. In his June 11 farewell address, the general got in a few last jabs at Rumsfeld. “You must love those you lead before you can be an effective leader,” he said in a parade-ground speech delivered on a humid morning. “You can certainly command without that sense of commitment, but you cannot lead without it. And without leadership, command is a hollow experience, a vacuum often filled with mistrust and arrogance.”

Shinseki did not refer directly to his clashes with Rumsfeld and Wolfowitz over the number of U.S. troops in Iraq, the Crusader, or the selection process for senior military officers. But he alluded to his tensions with the department’s civilian leadership. “The Army has always understood the primacy of civilian control,” he said. “In fact, we are the ones who reinforce that principle with those other armies with whom we train all around the world. So to muddy the waters when important issues are at stake—issues of life and death—is a disservice to all those in and out of uniform who serve and lead so well.”

The general also sided solidly with White. “When they call the roll of principled, loyal, tough guys, you will be at the top of the list,” he said to the former Army secretary. Further, with an eye toward Rumsfeld’s continued interest in reorganizing the Army and perhaps trimming its ranks,

Shinseki warned against cuts in the fighting force. “Beware the twelve-division strategy for a ten-division Army,” he said. “Our soldiers and families bear the risk and the hardship of carrying a mission load that exceeds what force capabilities we can sustain.”

After leaving office, Shinseki kept a strict public silence, telling associates that he didn’t want to criticize while soldiers were still fighting and dying in Iraq. In 2008, however, a copy emerged of a memo he had written to Rumsfeld just before leaving. In it, Shinseki offered what he called some “closing thoughts” on several of the controversies that had so roiled relations between the two men. “While our disagreements have been well-chronicled, and sometimes exaggerated, these professional disagreements were never personal, never disrespectful, and never challenged the foundational principle of civilian control of the military in our form of government,” the general wrote. “When the discussions were about the national security, I felt it was my duty to provide my best professional military advice.”

He said his February 2003 testimony estimating the forces required to stabilize postwar Iraq had been misinterpreted. “I didn’t believe there was a ‘right’ answer on the number of forces required to stabilize Iraq until the commander on the ground had the chance to conduct both his mission analysis and a troop to task assessment,” Shinseki wrote. He explained that he had deliberately chosen a high number in an effort to avoid imposing a “force cap” and foreclose options for Rumsfeld and Franks. It was unfortunate, he added, that he hadn’t had the opportunity to explain the rationale before the matter blew up into a public issue, although he noted that neither Rumsfeld nor Wolfowitz had ever discussed the issue with him “despite all the commentary in the press.”

He also wrote that his actions on the cancellation of the Crusader system were misinterpreted or misconstrued by Rumsfeld and the secretary’s aides. Noting the decision to eliminate the weapon had come “without any consultation or forewarning,” Shinseki said his subsequent testimony was meant to defend not the Crusader specifically but simply the Army’s continuing need for some kind of cannon artillery fire.

In a discussion of the direction that U.S. military change should take, he stressed the importance of developing human leadership over new technologies. He noted that Rumsfeld had repeatedly questioned the need for multiple echelons of Army command and had pressed to eliminate layers and headquarters, and he suggested the secretary think twice before doing so. “I would recommend that we ‘make haste slowly’ here and that technology theorists not be allowed to hold sway over practical analysis and operational experience,” he wrote.

Shinseki observed that the Army’s multiple command levels had proven useful in Iraq in executing a constantly changing plan. “Again, it’s about leadership, not fixed organizational designs,” he wrote. He also cautioned against trading “current capabilities” for “future possibilities,” urging Rumsfeld and the secretary’s aides to base their decisions on analysis and to “listen to commanders with combat experience.” Further, he advised Rumsfeld to keep in mind the many contributions the Army makes to other forces, whether fuel deliveries to the Air Force, artillery support to the Marines, or logistical help to Special Operations forces. “They are consistently overlooked,” he wrote, “which skews the perception” of the Army as a service staffed by too many support troops.

He warned against cutting back on troop strength, saying, “It is near impossible to undo the effects of bad end strength or force structure decisions.” And while he did acknowledge that “the Army is still slower to deploy than we would like,” he said the problem is an insufficient number of aircraft to transport troops, not too many troops.

But he reserved his sharpest criticisms for some of the ad hoc processes that Rumsfeld had instituted at the expense of long-established conventional ones. In the absence of Tank sessions with the secretary, he said members of the Joint Chiefs of Staff “are not given the opportunity to express their best military judgment as often as they should.” He also complained about “a lack of strategic review to frame our day-to-day issues” and “a lack of explicit discussion on risk in most decisions.” Additionally, he objected to “a tendency to compartmentalize analysis” and to senior-level meetings that lacked “structured agendas, objectives, pending decisions and other traditional means of time management.”

With the Army’s top military and civilian leaders gone, Rumsfeld was finally free to install the kind of people he wanted to carry out the transformation he hoped to see in the Pentagon’s oldest and largest armed service. For secretary of the Army, he recommended James Roche, the Air Force secretary, whom Rumsfeld regarded as an innovator, although that nomination ran into trouble over Roche’s fierce advocacy of a controversial deal to lease tanker aircraft rather than buying them outright. For chief of staff, Rumsfeld had wanted Keane, who declined the offer and retired because his wife, Terry, was ill with Parkinson’s disease. The secretary then tried unsuccessfully to recruit Franks and Abizaid before settling on Peter Schoomaker, a retired general who had headed the Special Operations Command from 1997 to 2000.

The selection of Schoomaker, who had spent much of his military career in Special Operations and once led the elite Delta Force, underscored Rumsfeld’s interest in reshaping the Army into a more agile, rapidly deployable organization. The decision to skip over the ranks of three- and four-star generals still on active duty caused some grousing and was taken as a further sign of Rumsfeld’s low opinion of the Army’s military leadership. But it was difficult to quarrel with Schoomaker’s credentials as a commanding presence and unconventional warrior.

The same day the choice of Schoomaker was announced, Rumsfeld recommended that Myers and Pace be nominated for second terms. “The secretary feels this team has been terrific,” a senior Defense Department official said, citing the work by Myers and Pace and the senior civilian leadership in redrafting military strategy, reshaping the forces, and redrawing regional commands.

Shortly after the invasion of Iraq, Newt Gingrich, in an effort to persuade Rumsfeld to pace himself, reminded him that General George C. Marshall had carefully managed his days during World War II to avoid exhaustion. Rumsfeld could relate to Marshall. On March 30—within days of the war’s start—Rumsfeld had surpassed the record set by Marshall as the oldest serving secretary of defense. Marshall had left office at the age of seventy years, nine months, and twenty-one days. Rumsfeld, approaching seventy-one, was still fit and vigorous but was beginning to question how much longer he could sustain the six- and seven-day weeks he had been working since taking over at the Pentagon.

To get her husband to ease up a bit, Joyce Rumsfeld had started hunting for a house outside Washington that could serve as a weekend retreat. She found one in the neatly maintained waterside village of St. Michaels on the Eastern Shore of Maryland, an hour-and-a-half drive from the nation’s capital.

The Rumsfelds paid $1.5 million for the brick Georgian. Located on four acres at the end of a gravel drive, the five-bedroom, four-bathroom, five-fireplace house is named Mount Misery. Legend attributes the name to the gloomy temperament of the Englishman who originally built the house in 1804. By 1833, Mount Misery’s owner was Edward Covey, a farmer notorious for breaking unruly slaves for other farmers.