Donald Rumsfeld.
Bush wanted Rumsfeld to develop “a new architecture of American defense for decades to come.”Photograph by Martin Schoeller / AUGUST

The improbable restoration of Donald H. Rumsfeld to the seat of American military power was consummated on a cold Friday afternoon in January, 2001, when he was welcomed back to the Pentagon with a full-honors review. The ceremony, a ritual display of martial pomp and fanfare, was staged on the River Parade Field, the vast lawn on the Potomac side of the Pentagon, where Rumsfeld had been welcomed as Gerald Ford’s Defense Secretary a quarter century earlier. Standing before the assemblage, Rumsfeld seemed like a figure lifted from another age. He had run the Pentagon in the time of the Berlin Wall, the era of Brezhnev and Kissinger, and the reunification of Vietnam under a Communist regime. Then he was the youngest-ever Secretary of Defense; now, at sixty-eight, he became one of the oldest, and the only man ever to serve in the position twice. His selection by George W. Bush had been a surprise, not least to Rumsfeld himself (“What in the world am I doing here?” he sometimes asked himself). Rumsfeld had been out of government since 1977, and although he was still well connected, he had not been a particularly close Bush adviser during the campaign. Even so, when General Henry H. Shelton, the chairman of the Joint Chiefs of Staff, who introduced Rumsfeld at the day’s event, and declared him “exactly the right man for the job in this new century,” everyone knew what he meant. Bush and Cheney had campaigned as friends of the military—“Help is on the way,” the candidates had repeatedly proclaimed—and Rumsfeld, a totem of the Republican national-security establishment, was a promise delivered.

At the ceremony, Rumsfeld declared that the new President’s first goal for defense was “to strengthen the bond of trust with the American military.” Earlier that day, at Rumsfeld’s swearing-in at the White House, the President had used the same phrase and promised to give Americans in uniform “the respect they deserve.” The refrain was repeated so often that a reporter finally pressed Rumsfeld to spell out what he and the President really meant: had the “bond of trust” been ruptured by eight years of Bill Clinton? Rumsfeld allowed the question to speak for itself (“I’m looking forward, not back,” he said), but no one missed the point: a central message of the Bush-Cheney campaign had been that the American military had somehow needed to be rescued from its Commander-in-Chief.

Of the several partisan fevers that afflicted the Clinton era, none was more acute than that having to do with Clinton and the military. Some of the disaffection reflected Clinton’s moment in history: the end of the Cold War offered the chance for a “peace dividend” of a smaller military, which Clinton seized, but, in the eyes of many military leaders, he lacked a consistent doctrine for the use of American power. Early in his first term, a relief mission begun by the elder Bush in Somalia became a military mission aimed at capturing the nettlesome clan leader General Muhammad Farah Aydid. When eighteen Army Rangers were killed and one was dragged through the streets of Mogadishu, Clinton pulled out. The following year, when a humanitarian disaster unfolded in Rwanda, the United States stood aside, at the price of countless lives—for which Clinton later apologized on America’s behalf. A “Clinton doctrine” eventually evolved, which defined the national interest partly in terms of humanitarian interventions, a construct that justified bombing in Bosnia, an air war in Kosovo, and the open-ended commitment of American troops to both. To Clinton’s critics, it seemed as if American foreign policy were shaped by the latest ghastly images from the nightly news cycle, and Clinton himself seemed to confirm it, saying his aim was a world where “we don’t have to worry about seeing scenes every night for the next forty years of ethnic cleansing in some part of the world.” The military was deeply skeptical about what it called “operations other than war,” such as peacekeeping and nation-building. The added duty for a smaller force put serious strains on the military’s readiness for combat as well as its ability to retain people. Toward the end of Clinton’s tenure, the Army, Navy, and Air Force each failed to meet its recruitment goals.

The revelation that President Clinton had had sexual relations with a female subordinate in the Oval Office, and then lied about it, struck a particular nerve in the military, and some of the deeply felt resentments began to surface. One Marine major wrote an article in the Navy Times referring to Clinton as “an adulterous liar.” Another marine, a reserve officer, wrote an opinion article, in November, 1998, questioning a military officer’s obligation to obey the orders of “a morally defective leader.” The Marine Corps Deputy Commandant, General Terrence Drake, issued an order to all the Corps’s generals to remind their officers of Article 88 of the Uniform Code of Military Justice—the provision prohibiting officers from using “contemptuous words” about the civilian authority.

And so, that January of 2001, when Donald Rumsfeld was officially welcomed back to the Pentagon, “There was just great relief in the military,” recalled Kenneth Adelman, a former Reagan arms-control official, who was among the invited guests that day. “There was the feeling that now somebody was going to be on their side.”

Bush had inherited a pressing problem. The American military continued through the nineteen-nineties to train, plan, and equip itself to fight an enemy—the Soviet Union—that no longer existed. It was hardly a secret that the military was badly in need of reform; everyone in uniform knew it, and those analysts and scholars who populated the think tanks of Washington had been fixated on the subject for most of a decade.

As a pro-defense Republican, Bush would have the political capital to bring about genuine, even historic, change. During the campaign, he had vowed to give his Secretary of Defense “a broad mandate to challenge the status quo and envision a new architecture of American defense for decades to come,” and he had chosen Rumsfeld because he believed that he would be more willing and better able than the other candidates to pursue his agenda. The contents and scope of that agenda were not yet known, but Rumsfeld made it clear that his approach to the military was very much hands-on.

Among those gathered at the River Parade Field for the Rumsfeld ceremony was an elderly man with a pleasant, grandfatherly aspect, who, amid the political celebrities and military brass, might have been taken for someone who had strayed from a Pentagon tour group. But within the national-security priesthood Andrew Marshall was something of a legend. He headed a unit called the Office of Net Assessment (he was its first and only director), which had evolved over the years into a sort of in-house Pentagon think tank. That made him the resident deep thinker, and what Marshall, who was in his late seventies, had been thinking about for every President since Richard Nixon (and for two decades before that at the Rand Corporation) was how America could prevail in the next big war.

Marshall’s professional life had paralleled the full sweep of the Cold War. He was admired as a boldly original theorist; in the forty-year strategic chess match between East and West, it sometimes seemed as if Marshall were playing a three-dimensional game. Marshall was among the very few who understood the Soviet vulnerability, and it was largely Marshall who imagined the strategy for exploiting it—the Reagan-era conceit of a winnable nuclear war, based on technologies (such as the unproved missile-defense shield) and levels of expenditure that the Soviets could not hope to match.

Marshall fell out of favor under the Clinton Administration, which saw less call for an esoteric Cold War strategist. But he had already turned his focus on something that he believed was of immense and pressing importance. It was a new way of thinking about the military, an idea with vast implications for every aspect of American defense, from the nation’s weaponry to its global posture, because it would radically change the way America waged war—indeed, it could alter the very nature of war. It was called the Revolution in Military Affairs.

This revolution had begun to unfold in the Cold War’s last stages. In the post-Vietnam nineteen-seventies, the Soviet-led forces of the Warsaw Pact conducted a steady, massive buildup of heavy forces—tanks and mechanized infantry—along the western edge of East Germany and Czechoslovakia. Twenty thousand battle tanks, mostly Soviet, faced west; on the other side, NATO fielded a force of only seven thousand tanks. In the event of a conventional Soviet attack, NATO forces would be forced to wage a fighting retreat until reinforcements arrived, mostly from the United States. One option for countering the Soviet advantage was the deployment of “tactical” nuclear weapons to Europe. Tactical nukes posed obvious political and strategic problems—the NATO allies did not welcome the prospect of even limited nuclear war in Europe, and there was always the chance of escalation into full-blown nuclear conflict.

But the Americans were also working on a non-nuclear weapons system that would change the equation. The top-secret Defense Advanced Research Projects Agency was funding a program called Assault Breaker, which was designed to strike far behind enemy lines, disrupting or destroying follow-on forces, gaining time and cover for the Western alliance to launch a counter-offensive. Assault Breaker was meant to compress the process of locating a target and launching a strike into a synchronized target-and-fire action taking just minutes. DARPA equipped an Air Force plane with an advanced radar system and onboard computers that worked out the target’s coördinates and transmitted them to an Army missile base, which fired rockets toward the target area. In a test of Assault Breaker at the White Sands Missile Range, in New Mexico, the system hit five out of five targets.

In the early nineteen-eighties, Marshall and his colleagues began to notice in their reading of Soviet military literature that the Russians were writing about this new American weaponry with increasing alarm. The Soviets assumed that deployment of Assault Breaker was imminent, and that this American advance represented the dawn of a new military epoch—what the Soviet analysts referred to as a “military-technical revolution.” (The Americans weren’t nearly as attuned to the implications of their own developments. Assault Breaker was not close to deployment; in fact, the Air Force and the Army were disinclined to coöperate, and the joint program eventually died.) Marshall began to analyze this idea, and, after months of study, his office concluded that the Russians had got it right. The Red Army’s Chief of Staff, Marshal Nikolai V. Ogarkov, had recognized that advances such as those tested by the Americans would give conventional weapons many of the strengths of nuclear weapons, without the apocalyptic effects. Implicit in Ogarkov’s insight was the idea that a key breakthrough in technology (for example, microprocessing) could suddenly reconfigure the battle-field—in this case, with accuracy so precise that, Ogarkov wrote, conventional warfare took on “qualitatively new and incomparably more destructive forms than before.”

Marshall was struck by this idea, and, throughout the eighties, he assigned teams of analysts to search for historical instances of such advances. (“What’s amazing,” Marshall told me, “is how much we know, it turns out, about the chariot revolution back in 1700 B.C.”) The example he found most compelling was that of Europe in the years between the twentieth century’s two World Wars. After the Pyrrhic victory of the First World War, France built the finest Army in the world, and an imposing defensive complex of forts, bunkers, and tunnels—the Maginot Line—along its border with Germany. A defeated Germany, on the other hand, had to overcome the constraints of the Treaty of Versailles (forbidding Germany warplanes, tanks, submarines, or heavy guns, and outlawing its general-officer staff) to build Hitler’s Wehrmacht. Yet, in the late spring of 1940, Germany invaded France and won a French surrender in less than six weeks. The stunning German victory was produced by a battlefield innovation, blitzkrieg, that married two relatively new technologies—radio and the internal-combustion engine—to facilitate a tactic of rapid, coördinated movement. The Germans had designed their new panzer divisions to suit this doctrine, and their impact was decisive. The German forces bypassed the Maginot Line, attacked directly through the Ardennes Forest, and dashed to the French rear, sowing chaos en route and forcing a quick surrender.

To Marshall and his associates, the lessons were clear. The side that recognized and exploited such advances gained not just an edge in warfare but an overwhelming advantage; for the side that missed the chance, the consequences could be fatal.

With the impending demise of the Soviet Union, the U.S. was entering a new interwar period. Marshall had no doubt that some new competitor would emerge to challenge the United States, and it struck him as exactly the moment to prepare for the next big war. The U.S. led the world in microchip technology, and the information age promised a dazzling range of military applications, such as advanced sensors, satellite imagery, robotics, and laser systems. The danger would not arise in Europe, Marshall believed, but in Asia—most likely China. The conflict would not be a prolonged ground war, involving massed formations of infantry and tank divisions; rather, there would be long-range precision strikes by “smart” missiles. If there was infantry in the fight at all, it would be in small, specialized units. Marshall supposed, too, that in the global economy these technologies would be available to all. This made it imperative that the U.S. push conflicts to distant battlefields, if possible, and to reduce (or eliminate) such easy American targets as overseas airbases and huge aircraft-carrier battle groups.

In July, 1992, during the race between George H. W. Bush and Clinton, Marshall gave the Pentagon’s senior leaders a formal assessment reflecting his conclusions about the Revolution in Military Affairs. Marshall preferred that term to the Russians’ “military-technical revolution,” because he believed that technology only partly accounted for such bursts of progress. The other critical element was a military’s adoption of entirely new operational concepts, organizational structures, and doctrines. For instance, the French and British had radios, tanks, and airplanes in 1940, but Germany put them to novel use. Marshall wrote that new ideas had to be tested, even if most of them failed. Perhaps most important, the Pentagon needed to stop depending exclusively on the big-ticket weapons that devoured defense dollars and perpetuated the status quo.

Marshall’s assessment came just as the national-security establishment was trying to define America’s posture in a world without a Soviet counterweight. Some, including the first President Bush, Brent Scowcroft, his national-security adviser, and the chairman of the Joint Chiefs of Staff, General Colin Powell, were avowed “realists,” who believed that America’s role was to be part of a new world order, with the emphasis on order. But others believed that the U.S. should embrace and, if possible, enhance its position as the world’s sole superpower. Bush’s Secretary of Defense, Dick Cheney, was inclined to this view, as was his top policy official, Paul Wolfowitz, who argued emphatically that the United States should strengthen its military superiority so that potential rivals would have no hope of catching up. Marshall’s Revolution in Military Affairs offered a means to achieve that goal, and both Cheney and Wolfowitz became converts.

There were also converts within the uniformed military, including a few in the senior ranks, but the services were generally skeptical about the R.M.A., as it was now being called. Some of the resistance, particularly in the Army, reflected the belief that Marshall’s vision of long-distance precision strikes ignored the gritty reality of actual war, with soldiers on the ground. Had President George H. W. Bush won another term, he might have been willing to impose upon the military the upheaval that a revolution in military affairs implied. But the last thing the Army was inclined to do while facing cutbacks under the Clinton Administration was tinker with its revered divisional structure, and the Navy was no less inclined to reduce the number of its aircraft-carrier battle groups. Clinton’s last Secretary of Defense, William S. Cohen, had tried to get the Army to transform itself into a lighter, more expeditionary force. But Cohen, a Republican, was frustrated from the start. “I was coming into a Democratic Administration, and that had its own dynamic,” Cohen recalled. “I must say that President Clinton gave me total authority, so it wasn’t a question coming from him. But just dealing with the issue—how do you push transformation in a Democratic Administration? Is this something that’s weakening the military? The perception on the Hill would be ‘Here they go cutting back on the military powers of the Army.’ ”

Still, Marshall continued to promote his revolution. Using his budget at the Office of Net Assessment, he financed his own futuristic war games. The Revolution in Military Affairs thrived in think tanks and seminars.

As Marshall watched Rumsfeld’s official welcoming ceremony, he was hopeful that the revolution’s moment had arrived. During the Presidential campaign, George W. Bush had promised to build a new American military for the twenty-first century. In a speech at the Citadel in 1999, Bush had said that as President he would instruct his Defense Secretary to conduct a “comprehensive review” of the military, to question everything from its force structure and strategy to its acquisition process. He promised not just to make “marginal improvements” but “to replace existing programs with new technologies and strategies, to use this window of opportunity to skip a generation of technology.” That speech was instantly recognized, by those with a trained ear, as the language of the Revolution in Military Affairs.

Marshall had known Rumsfeld over the years, and he liked him. Shortly after Rumsfeld’s induction, the new Secretary arranged to have lunch with Marshall—not, as Marshall had expected, in Rumsfeld’s private office but in the “Sec Def Mess,” a nearby dining room where the guests seldom went unnoticed. “Oh, I think it was very clear,” Richard Perle, a former Reagan defense official and a close adviser to Rumsfeld, recalled. “It went all over the building that Andy was back. It was like Deng Xiaoping’s return.” The President had pledged to conduct a comprehensive review, identifying probable American adversaries, and when and where the next wars would likely occur. Rumsfeld asked Marshall if he would like to take something like that on. Marshall said he could put a team together right away. The Pentagon’s traditional review process, the Quadrennial Defense Review, was just getting under way, and wouldn’t be finished for another nine months. Hundreds of uniformed staff officers from all the services had spent tens of thousands of man-hours trying to answer essentially those questions. This review tended to be an exercise in justifying the budgets, force size, and programs that the military services wanted to protect. Rumsfeld evidently intended to circumvent that process. He told Marshall that he’d like to have the first draft of his strategic review in six weeks. “We delivered on that,” Marshall recalled.

Word of Marshall’s new assignment rang through the Pentagon like a distress signal, which may have been part of Rumsfeld’s plan. Rumsfeld had in mind for the military, and for the Pentagon itself, an agenda of radical reform. He called it “transformation,” and the return of Andrew Marshall meant that it would be guided by the principles of the Revolution in Military Affairs. Rumsfeld intended to remake the American military into a lighter, more agile, more readily useful force that would be able to leverage new technology to project lethal power over great distances.

Marshall would have been the first to say that technology and the Revolution in Military Affairs had very little application to certain kinds of conflict, such as a counter-insurgency fight against some indigenous guerrilla force. But that was the sort of war that no one—on the new Bush national-security team, or certainly in the American military—had any intention of ever fighting. That would be a war like Vietnam.

When war came, with the invasion of Afghanistan, in late 2001, Rumsfeld had only the barest beginnings of a transformed military, but he had a fully formed philosophy that dictated how America would fight. In Afghanistan, it meant routing the Taliban with small bands of American Special Forces and coördinating long-distance air strikes and Afghan ground troops. For the subsequent invasion of Iraq, in March, 2003, the Rumsfeld vision meant getting to Baghdad and toppling Saddam Hussein’s regime as quickly and with as small a force as prudence permitted. Many professional military men strenuously disagreed with Rumsfeld’s war plan, but, fresh from the validating triumph in Afghanistan, he prevailed.

The man who came to exemplify this new way of war was Major General Buford (Buff) Blount, a courtly Mississippian who commanded the Army’s 3rd Infantry Division—the 3rd I.D.—and who became a popular figure with reporters. ABC’s Ted Koppel rode with Blount, as did the late David Bloom, the NBC News correspondent whose dust-blown satellite video reports provided the signature images of the conflict’s early days, and Michael Kelly, the Atlantic editor, who died in a skirmish near Baghdad. The 3rd I.D., with its tanks and armored fighting vehicles, was, by training and structure, a relic of the pre-Rumsfeld Army. But Blount was a creative, daring general, who, after weathering a brief storm delay, pushed quickly through to Baghdad. And it was Blount who devised and executed the “thunder runs” from the outskirts of Baghdad to the heart of the Iraqi capital, providing the decisive blow in the stunningly short campaign to end the Saddam regime.

But within a few weeks Rumsfeld and his closest advisers came to regard Blount as part of a problem they had not foreseen and were wholly unprepared to handle: the gathering chaos that soon became an insurgency.

With American troops in Baghdad, General Tommy Franks, the top commander of the coalition forces, left his headquarters in Doha, Qatar, and flew in a Kevlar-lined helicopter and then a C-130 transport plane to Baghdad. He made his way to the Abu Ghraib North Palace, where he and his commanders shared a victory cigar, and Franks told them to start preparing to leave Iraq.

Blount’s 3rd I.D. had been slated to go home, and his troops, having stormed into Baghdad, considered their mission accomplished. But the days passed, and Blount’s unit found itself faced with a policing mission it didn’t want and hadn’t expected. Scores of looters appeared, and then hundreds and thousands, making off with anything that could be carried away.

“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,’ ” Blount told me. “I mean, it was just never, ever, ever mentioned. Our focus was on fighting the war.” Blount was loath to order his troops to use deadly force on looters. As he saw it, the raiders were indulging an impulse to street justice. “These are the people that we’re liberating,” he recalled. “The oppressed. You know, that have been without for centuries now, and they’ve got nothing, and they’re trying to get a little bit back from the Baath Party.”

Blount ordered patrols to secure hospitals, power facilities, and other key structures, but in many cases the looters had already come and gone. As it became apparent in Washington that the disorder might spin out of control, the 3rd I.D.’s pullout was delayed. At the division’s home base, in Georgia, some of the troops’ spouses began to voice their dismay, and in Baghdad, some of Blount’s soldiers complained to reporters. One, a private named Matthew C. O’Dell, told the Times, “You call Donald Rumsfeld and tell him our sorry asses are ready to go home.”

Blount’s superiors in the Pentagon were furious, blaming his unit’s recalcitrance on a lack of leadership. “We had no orders or plans to occupy Baghdad,” Blount says. “There were just a lot of assumptions made. I’ve looked back at this a good bit at different times, wondering, you know, well, could we have done something differently?”

General Jack Keane, then the Army’s Vice Chief of Staff and one of Rumsfeld’s most trusted advisers in the force, flew to Iraq and was chagrined that the 3rd I.D.’s briefings failed to focus on containing the spreading disorder. But Keane later reflected that Blount’s troops could hardly be blamed. “They were ill-prepared—they weren’t educated to do it, and they weren’t trained to do it, and they weren’t expecting to do it,” Keane said, of the unit’s impromptu policing mission. When the 3rd I.D. eventually left Iraq, four months later, Blount was assigned to a desk job at the Pentagon, a role he served in for the remainder of his active career. He did not get a promotion and a third star, as many had once expected.

As chaos spread in and around Baghdad, Rumsfeld remained publicly serene. “You don’t go from despotism to freedom on a feather bed,” he told reporters on April 24th. Pressed to explain the vivid images of chaos in Iraq, Rumsfeld dryly observed that “freedom’s untidy.” From the start of the new Bush Administration, Rumsfeld had been the Administration’s lightning rod, and he seemed almost to relish controversy. Just a few months into his tenure, as he was making enemies inside the Pentagon’s bureaucracy with his insistent reform crusade, and on the Hill with his reluctance to pay deference, the political press had begun a deathwatch on his career. Maureen Dowd mocked him as Rip Van Rummy, and Slate observed, “Rumsfeld is probably toast.” But by the time Baghdad fell, Rumsfeld believed he had seen his critics proved wrong so often that he had every reason to rely on his instincts.Two weeks before the Iraq invasion, a reporter had asked him, “Can you promise it won’t be a quagmire?”

“I can almost promise you that someone in this room will say it’s a quagmire,” Rumsfeld replied. Indeed, by the second week of the drive to Baghdad, reporters had summoned the “quagmire” scenario, only to be chastened by the rapid fall of Saddam’s regime. The subsequent failure of the search for weapons of mass destruction, whose existence had been a major justification for the war, may have been a blow to the Administration’s credibility (as well as a distraction for troops engaged in the hunt), but it did not seem to affect Rumsfeld’s certainty in his own judgment. More and more, he met questions about the progress of the mission in Iraq with ironic (and, some would say, arrogant) deflection.

In April, as the situation in the streets of Iraq grew messier, Rumsfeld said that “free people are free to make mistakes and commit crimes and do bad things.” More pointedly, he refused to dignify the chaos with a name that might summon memories of Vietnam. To many conservatives, the Iraq war, quite apart from its strategic implications, was a way of exorcising those ghosts, a chance to demonstrate the national will and military prowess that had arisen in the decades since the humiliating withdrawal from Indochina. Long after others, on the ground in Iraq and even within Rumsfeld’s inner circle in Washington, discerned the ominous signs of a gathering insurgency, Rumsfeld insistently declined to call it such.

Jack Keane, the Vice Chief of Staff, says that he was the first to raise the matter in the Tank, the super-secure Pentagon meeting room of the Joint Chiefs of Staff. “I said, ‘This is a low-level insurgency,’ and I told them what the definition of that was, and what it looked like,” Keane said. “They were targeting Americans, and it was organized. It was being done at multiple locations, which meant that there had to be some general guidance. . . . The operations they were conducting had some basic complexity to it. They were planned, and thought out, and well executed, in the sense that they were achieving success. They were killing us.”

It was not a welcome observation, as General Richard B. Myers, then the chairman of the Joint Chiefs, made clear, according to Keane. “I got some pushback from the chairman, and he said, ‘We’ve gotta be careful about that word,’ ” Keane said.

Rumsfeld’s deputy, Paul Wolfowitz, also sensed that what was happening in Iraq was more than the untidy impulses of a newly free people. In testimony before the House Armed Services Committee, two months after the fall of Baghdad, he conceded, “There’s a guerrilla war there, but . . . we can win it.”

Wolfowitz, too, got pushback, both from Rumsfeld and from Tommy Franks, who said three weeks later that “guerrilla and insurgency operations are supported by the people, and I’ve demonstrated to my own satisfaction that the people of Iraq do not support the violence that we’re seeing right now.”

Keane said, “They thought maybe it wouldn’t be that bad, and there’s no sense using those names associated with it, because you’re gonna stomp it out in a couple, three months—it wasn’t worth all the rhetoric.”

In a remarkable exchange with reporters at a Pentagon briefing in late June, a reporter read Rumsfeld a dictionary definition of “guerrilla war” and asked why he was so reluctant to name it as such. “Can you remind us again why this isn’t a quagmire?” the reporter asked. Rumsfeld replied, in his best vice-principal tone, as if patronizing a particularly dull student, “I guess the reason I don’t use the phrase ‘guerrilla war’ is because there isn’t one, and it would be a misunderstanding and a miscommunication to you and to the people of the country and the world.” He explained that the disturbances in Iraq were being caused by looters, criminals, remnants of the regime, foreign terrorists, and Iranian agents. “Doesn’t make it anything like a guerrilla war or an organized resistance. It makes it like five different things going on that are functioning much more like terrorists. . . . Now, that is not—it doesn’t fit that word. So, I think, that if one analyzes what is going on in that country, they would find a different way to characterize it. I know it’s nice to be—have a bumper sticker, but it’s the wrong bumper sticker.”

At this, the reporter pulled out the official Defense Department definition of “guerrilla war” (“I knew I should have looked it up!” Rumsfeld said. “I could die that I didn’t look it up!”): “ ‘Military and paramilitary operations conducted in enemy-held or hostile territory by irregular, predominantly indigenous forces.’ This seems to fit a lot of what’s going on in Iraq.”

To which Rumsfeld replied, “It really doesn’t.”

Two weeks later, Rumsfeld’s top commander on the ground, General John Abizaid, a Rumsfeld favorite, who had just taken over the leadership of the Army’s Central Command from Tommy Franks, addressed the question squarely. The insurgents who were attacking Americans, Abizaid said, “are conducting what I would describe as a classical guerrilla-type campaign against us. It’s low-intensity conflict in our doctrinal terms, but it’s war however you describe it.”

Abizaid, a Lebanese-American who speaks Arabic, went on to note that the insurgents were mostly mid-level Baathists, members of Saddam’s Special Republican Guard and security services, and that they had organized regionally into cells. For his efforts, Keane says, Abizaid “got his ass handed to him.”

The American invading force had prepared for a chemical assault from Saddam’s Army, for a set-piece battle with Saddam’s armored units, and for a long, bloody battle for Baghdad. “What we didn’t plan for,” says Jack Keane, “is what happened.”

The American military had spent thirty years avoiding the thought of fighting a counter-insurgency, a subject that hadn’t been part of the curriculum at West Point since the Vietnam era. The proud, professional volunteer force constructed since Vietnam was not built for a protracted occupation, an eventuality that, in any case, the nation’s political leadership reflexively rejected out of hand. Bill Clinton fought a war in Kosovo by announcing at the outset that the commitment of ground troops was not an option; the Bush Administration came into power renouncing the idea of nation-building as a foolish misuse of military resources.

But the inability quickly to recognize and formulate a coherent strategy to fight the insurgency that arose in Iraq reflected more than the military’s institutional biases, or Donald Rumsfeld’s intransigence. It also partly reflected a crucial policy decision in Washington, which effectively dashed any chance that the coalition forces would be received as liberators rather than as occupiers.

The circle of defense advisers that had most ardently advocated the Iraqi invasion, including Richard Perle and Newt Gingrich, had imagined a strategy that wouldn’t require a lasting American presence in postwar Iraq. The plan depended on the recruitment and training of “free Iraqis” to participate in the combat phase of the operation, and the imposition of a provisional government, run mostly by Iraqi exiles, after the war. Something like that had worked in Afghanistan, and, the reasoning went, the approach stood an even better chance of working in Iraq; Iraqi exiles had been planning for such an eventuality for more than a decade. But the program to train Iraqi fighters produced fewer than a hundred recruits; it also ignored the reality that prominent exiles like Ahmad Chalabi had less credibility, and less of an indigenous base, than those whom the U.S. had relied on in Afghanistan. The Defense Department’s plan to set up a provisional Iraqi government was abandoned after a bitter interagency argument within the Bush Administration that lasted until the very eve of the war. The State Department and the Central Intelligence Agency refused to endorse the imposition of a provisional government composed of Iraqi exiles, arguing that it would not be seen as legitimate. In the end, Rumsfeld surrendered on the point—to the lasting distress of the hawks nearest him. “I think he made a serious mistake,” Perle, a member of Rumsfeld’s advisory Defense Policy Board, recalled. “I think he underestimated the importance of getting those matters right.”

President Bush, with Rumsfeld’s approval, ultimately decided that postwar Iraq would be governed by an American-led Coalition Provisional Authority, to be headed by a veteran diplomat, L. Paul Bremer. With that appointment, and the implicit personal authority conveyed therein, came a critical, and not entirely intended, shift in American postwar policy. Bremer became the American proconsul in Iraq, technically reporting to Rumsfeld’s Defense Department but exercising a degree of authority that came to surprise even Rumsfeld. Bremer began his tenure, in May, 2003, by issuing a series of edicts that included the disbanding of the Iraqi Army and the removal of senior Baath Party members from government jobs. (Bremer said that he was acting on instructions from the Pentagon.) The edicts signalled that Baathists would have no place in the new Iraq, but they also crippled the bureaucracy and eliminated the most important instrument of Iraqi unity and a crucial tool in establishing order. In effect, half a million men, many with guns, were sent into the streets.

These moves had a decisive impact on the coalition’s response to the widening insurgency. A huge instant bureaucracy was set up inside the walls of Saddam’s former Republican Palace, where Americans laboriously laid plans for undertakings ranging from the design of a new Iraqi flag to the restructuring of the Iraqi monetary system. Meanwhile, no coherent, unified plan to fight the insurgency emerged, which rendered such plans increasingly abstract. “It was Alice in Wonderland,” recalled Gary Anderson, a defense specialist who was dispatched to Iraq by Paul Wolfowitz to help set up an Iraqi civil-defense corps. “It was surreal. I mean, I was so depressed the second time we went there, to see the lack of progress and the continuing confusion. The lack of coherence. You’d get two separate briefs, two separate cuts on the same subject, from the military and from the civilians.”

To Wolfowitz and others who had advocated the quickest possible turnover of authority to Iraqis, the C.P.A. was a maddening obstacle to the ever-dwindling hope of replicating the Afghanistan success. “There was an explicit, coherent strategy,” recalled Newt Gingrich, another influential neoconservative member of Rumsfeld’s Defense Policy Board. “It was the strategy of staying light, getting in, rebuilding the Iraqis quickly, and getting out. The war plan was right. Bremer thought he was MacArthur in Japan. And Bremer, in thirty days, destroyed the strategy. And neither Bush nor Cheney nor Rumsfeld would intervene decisively enough, and neither would Franks or Abizaid. It was one of the most amazing strategic mistakes I’ve ever seen.”

Such comments may be the retrospective rationale of men who’d advanced a war policy that had gone bad. Events in Iraq have provided no compelling evidence that the war hawks’ plan stood any real chance of success, even if it had been followed. But Bremer did come to symbolize an occupying force that did not seem intent upon leaving anytime soon. “When one now looks back,” Douglas J. Feith, the Under-Secretary of Defense, would later observe, “and says, huh, we had an insurgency, and the insurgents were able to capitalize on the widespread perception that the United States wanted to occupy, own, exploit, and stay for a long time running Iraq—and that enormously valuable piece of information operations by the bad guys was facilitated by the way we organized the C.P.A. I mean, one can look back, make that point, and then say, what a dumb, obvious error.”

As the weeks turned into ever more bloody months, it became increasingly clear that the coalition could lose in Iraq. General Abizaid, the CENTCOM commander, knew that the coalition had to have a single, coherent counter-insurgency strategy, and it had to have one soon.

On November 11, 2003, Abizaid sent a memorandum to Rumsfeld, in which he provided the Secretary with the definition of “counter-insurgency,” taken from the Pentagon’s own dictionary of military terms:

Sir, our doctrine states: “Counterinsurgency—those military, paramilitary, political, economic, psychological and civic actions taken by a government to defeat insurgency.”. . . Clearly we must integrate elements of national power in any effort to defeat an insurgency.

Abizaid attached to the memo a one-page primer called “Elements of Successful Counterinsurgency” (“worthy of your time to digest,” Abizaid noted to Rumsfeld), which included the advice to “develop a coordinated, integrated plan based on an accurate assessment of the insurgency’s goals, techniques, and strategies.” It is difficult to imagine that, at that late date, the Secretary of Defense needed to be told the meaning of “counter-insurgency,” but the likely purpose of Abizaid’s memo is no less remarkable. Seven months after the fall of Baghdad, the Administration still lacked a strategy for countering an enemy it did not fully understand. Abizaid’s memo underscoring the need for an “integrated” strategy arrived at the same moment that Bremer was summoned to Washington for a series of meetings that effectively reined him in. He returned to Baghdad with instructions to hasten the timetable for establishing Iraqi sovereignty. (Bremer said that the new timetable was his idea.)

Insurgency and the means to combat it soon became the hot new subjects in military circles; books on counter-insurgency, such as John Nagl’s “Learning to Eat Soup with a Knife,” became required reading around the Pentagon and among junior military officers. Rumsfeld himself never quite surrendered to the language of Vietnam—preferring terms such as “terrorists” to “guerrillas” or “insurgents”—but he’d long since realized that Iraq had devolved into something much worse than untidy. Just a month before Abizaid’s memo reached him, Rumsfeld had written to four top Pentagon officials, conceding that the war in Iraq would be “a long, hard slog.”

By the end of George W. Bush’s fifth year in office, the inner core of true believers around Rumsfeld, such Iraq hawks as Wolfowitz and Feith, were mostly gone from the Pentagon. Rumsfeld, meanwhile, was poised to become the longest-serving Defense Secretary ever, and was already a powerful token of the war itself, freighted with all its failures and controversies, from the abuses at Abu Ghraib to the harsh interrogation policy that Rumsfeld had personally endorsed. Among Democrats, insistence upon Rumsfeld’s resignation became a kind of litmus test of antiwar seriousness. By the time of the 2006 midterm elections, more than a few Republicans were inclined to concur.

To the end, Rumsfeld seemed to embrace the controversy. When, on the day after the election—an emphatic no-confidence vote on Rumsfeld’s war—he joined Bush at the White House and his departure from the Pentagon was announced, Rumsfeld could not resist comparing himself to another civilian warrior who believed himself misunderstood by his critics. “I have benefitted greatly from the criticism,” he said, invoking Churchill, “and at no time have I suffered from the lack thereof.”

In some way, Rumsfeld’s long tenure has brought the relationship between the military and its civilian leadership full circle. At the end of the Clinton Administration, some uniformed critics of the President had to be warned about possible insubordination; Rumsfeld inspired similar passion among some senior military leaders. This spring, what appeared to be a coördinated wave of criticism of Rumsfeld by retired generals moved some of their peers to suggest that they, too, be mindful of the prohibition on “contemptuous” remarks. On March 19th, General Paul Eaton, who had been in charge of training Iraqi troops before his retirement, last January, wrote in the Times that the Defense Secretary “has shown himself incompetent strategically, operationally and tactically and is far more than anyone else responsible for what has happened to America’s mission in Iraq. Rumsfeld must step down.” Lieutenant General Greg Newbold, who retired in 2002, called for Rumsfeld’s resignation in Time in an essay that cited “McNamara-like micromanagement” and stated that “a fundamentally flawed plan was executed for an invented war, while pursuing the real enemy, al-Qaeda, became a secondary effort.”

Rumsfeld had arrived at the Pentagon convinced that the American military had to be broken down and rebuilt, and he had done just that. In Rumsfeld’s six-year tenure, he had initiated a dizzying list of reforms, some deemed necessary even by his critics, from the restructuring of the Army’s once-cumbersome fighting formations to the repositioning of American forces abroad, shifting them away from Western Europe. Less visible change was brought to the military culture, especially in the Army, which now sees itself as the force of first resort—an attitude once reserved for the Marines. But it is also true that the Iraq war has severely stressed the Army, and Rumsfeld’s reluctance to significantly increase the size of the force has it stretched to its limits. It is a war-honed and immeasurably more knowledgeable Army, precisely because it is bound up in a conflict with no apparent end, facing a resourceful enemy who daily presents some new lethal challenge. It has truly been transformed, but not in the manner that Andy Marshall would have prescribed, or foreseen. It is caught in a de-facto civil war that has cost the lives of more than twenty-eight hundred service members, and those of many times that number of Iraqis.

Donald Rumsfeld might have made a fine peacetime Defense Secretary. But it may also be that Rumsfeld would never have lasted in peacetime. He survived in the job because war came, and he used the wars to force change upon a recalcitrant professional military. In a real sense, the war in Iraq allowed Rumsfeld to create his transformed military, even while presenting it with a mission that it may not be able to win—a war like Vietnam. ♦