6.

GAMBLING ON A “SHITTY HAND”

(Spring and Summer 2007)

We were dealt a really shitty hand, but we’ve played it to the best of our ability,” Col. Peter Mansoor said as he looked back to the troubled beginning of 2007.

They had deplaned into a small civil war, and the streets of Baghdad seemed to grow bloodier by the day. On January 16, two bombs were detonated during the after-school rush at a Baghdad university, killing at least 60 people. Six days later, two more bombs devastated a street bazaar, killing at least 79 more. On January 30, 60 Shiites were killed in multiple attacks across central Iraq. “We had U.S. Air Force F-16s engaging the enemy on Haifa Street, twelve hundred meters from the embassy,” recalled Kilcullen.

It is easy to forget now, after it has become conventional wisdom that the surge worked, at least tactically, how audacious a venture it was. Almost all military experts agreed with the Joint Chiefs of Staff that the U.S. troop presence was an irritant, so more troops likely would only worsen the situation. The liberal position was to withdraw as soon as possible. Hawkish centrists advocated getting smaller and staying longer. Escalation of the sort that was chosen was a radical position advocated by a small minority. At best, it was unclear what a relatively small number of additional troops might do. At worst, many thought, it simply was reinforcing failure—a cardinal sin in military operations. The consensus seemed to be that at best it probably was just delaying a horrific civil war that unfortunately seemed inevitable.

But all that pessimism had one positive side effect, because it created the conditions for strategic surprise—which as Clausewitz, the great Prussian philosopher of war, observes, is the most important and effective kind of surprise. After four years in Iraq, no one seemed to expect the Americans to develop a way to operate differently and more effectively.

The shift was all the more unexpected because it came as President Bush was politically cornered. Usually, “sustained strategic boldness . . . requires a solid foundation of popular support,” Oxford historian Piers Mackesy observed in The War for America, his classic study of how the British managed to lose the American Revolutionary War in 1781 after appearing to have won it just a year earlier. But in agreeing to a troop escalation, Bush was operating from a position of extraordinary political weakness. Not only was he deeply unpopular, he had reversed course at a time when it seemed that stubborn persistence was his sole virtue as a leader: After years of saying he would heed the advice of his military, Bush had split with the overwhelming view of his top military leaders, from the Pentagon to Central Command to the top general in Iraq.

What probably saved Bush was his political opposition—a splintered and confused Democratic Party. The Democrats were close to paralyzed by the Iraq war, wanting to gratify their supporters by questioning it but not wanting to be responsible for the outcome. The major weapon available to them was to cut off funding for the war—but to do that would make them appear antimilitary, which would carry a political price they were not willing to pay. Put bluntly, they wanted to appear to be doing something about it without really doing anything. So, while the House of Representatives voted 246 to 182 in February 2007 to oppose the surge, it wasn’t prepared to follow up that nonbinding resolution with action. This empty-handed approach would prove to be a huge political advantage to Bush, enabling him to launch and continue the counteroffensive.

PETRAEUS AMID THE PESSIMISTS

In news photographs, they are the people on the side, escorting a top official or leaning in to interpret during the photo shoot. In one such image that ran on the front page of the New York Times, two Petraeus aides, Col. Mike Bell and Sadi Othman, flank the Iraqi prime minister and the American secretary of state. These were the people who studied operations, wrote critiques, and drafted papers, the plumbers and mechanics of policy. As 2007 began, few of them believed the surge would succeed. “When I first got here in January,” said Emma Sky, her “sense was, the war was lost, how do we get out?” Nor, with her pacificistic tendencies, was she attracted to the new American strategy. “At the beginning of the surge, I felt violence begets violence. I felt sick. I felt horrible.”

Kilcullen calculated that Petraeus would achieve his goals on security but not on politics. He went on to bet that he could summarize the situation in just ten words and did: “My bottom line: good team, right strategy, possibly too late.” He even had drawn up a paper to give to Petraeus if the situation fell apart. First, he advised, you will need to recognize that you have reached a decision point. Second, act on that recognition in a timely fashion. Third, “credibly communicate” your assessment to the president and other Washington decision makers.

Nor did Capt. Liz McNally, the eager young Rhodes Scholar who was drafting Petraeus’s speeches, think they were on the path to success in Iraq: “Even given the perfect amount of resources, I don’t know if what we’re trying to do is possible,” she confessed one day in the spring.

Hearing that Petraeus might be given command in Iraq, Sadi Othman prayed that he wouldn’t take it, “because the situation was very bad, and because I care about my friend Dave Petraeus.” The stakes were huge, Othman believed, and the odds against success nearly overwhelming. “Let me put it this way, it is very hard to be very optimistic,” he said one day in May 2007, as the casualties continued to rise. “Having said that, I strangely believe it is doable. If, God forbid, Iraq falls apart, I think it will impact the entire region in an unbelievable way. If the problems aren’t solved, I believe the consequence is the whole region up in flames.” It was a chilling thought. He folded his hands in his lap.

One day early in 2007, Col. Bill Rapp, Petraeus’s closest adviser, was in his office watching CNN’s Michael Ware, a reporter he respected, discuss the state of the war. The correspondent gloomily said to his colleague Anderson Cooper that “it just doesn’t seem that there’s any road forward that does not involve the spilling of so much innocent blood or the abandonment of so many of the principles that we of the West hold dear.”

Col. Rapp, who was already worried, “trying to figure if we needed to get out of Dodge,” was so struck by the comment that he wrote it down. Then he picked up a marker and copied it onto the big erasable white board he used with his subordinates to brainstorm. “I wrote it down as a challenge to myself and the CIG [commander’s initiatives group] to help the CG [commanding general] find an alternative. Those days were fairly bleak.”

Their job as the brains behind Petraeus, he instructed them, was “to prove Mick Ware wrong.” Rapp’s deputy, Charlie Miller, arriving in Iraq in February 2007, estimated the chances of success at 10 to 15 percent. By May he considered himself a relative optimist and raised his guess to 35 to 40 percent. It was better but still far from a safe bet.

Soon after he arrived in Iraq, Lt. Col. James Crider, commander of a cavalry squadron deploying in Baghdad, was pleasantly surprised to run into Col. Mansoor, whom he had known and admired for years.

“Hey, sir, I’m pretty optimistic, I think it’s gonna work,” Crider said.

“I’m not,” Mansoor replied, gray-haired and expressionless behind his glasses. “I’m not sure it’s gonna work. In fact, the odds are against it.”

It was a sobering, even frightening exchange for Crider, who had orders to take his unit into one of Baghdad’s toughest neighborhoods. He thought to himself, “This is a guy I know, and he’s General Petraeus’s executive officer, and he’s not sure it’s gonna work?”

One of the few relative optimists around Petraeus was a senior intelligence official who would be interviewed only on the condition that he not be identified by name. “I thought we had a real chance of making it work,” he remembered. At the American military headquarters in Iraq, he said, “A lot of people were thinking ten percent, fifteen percent.” He was at 40 percent, he said.

Despite the odds, they were going to try, especially because they didn’t see a lot of good alternatives. Just because the odds were bad didn’t mean there was a better choice available. There was in this period a sense of being dutiful: They had to cast a cold eye on the blunders of their predecessors while trying to be positive about their own chances. They had to risk their lives and see comrades bleed and die, all the while believing it was likely their efforts would fall short. Mixed with that ambivalence was a determination to at least try, to give it one more shot and at least salvage as much as possible.

Even the principals harbored profound doubts. “I didn’t know,” said Ambassador Crocker. “I thought it could work. If I had thought it absolutely would not I would be insane to come out here . . . I will not be one of those who said I saw this all along. I thought probably it was a long shot, given the levels of violence that had prevailed and the damage they had done to the political and social fabric.”

Odierno also harbored doubts but was at the optimistic end of the scale. “I thought about seventy-thirty, it would work,” he said, looking back. He didn’t think five brigades were enough, but figured that by adding in a Marine battalion, an aviation unit, and various Special Operations units, he could get close to what he needed.

Petraeus, the apotheosis of “can-do”-ism, may have been alone in holding that the new mission was entirely plausible. “I didn’t consider it a Hail Mary pass,” he insisted one day that spring. He saw a series of tasks that needed to be performed, and thought they could be done with some additional troops, some reasonable improvement in the quality of Iraqi forces, and some application of the theory of counterinsurgency. At the ceremony at which he took command, he gave a short talk in which he assured his audience, “this mission is doable.” But a year later he would concede that part of the role of a commander is to stay publicly optimistic.

“THE MESOPOTAMIAN STAMPEDE”

Petraeus’s chosen image of his task was a Frederic Remington oil painting called The Stampede, a 1908 work that depicts a nineteenth-century cowboy riding for his life as a herd of cattle panics under a breaking thunderstorm. The cowboy’s own pony is wild-eyed with fear, all four hooves clawing in the air. Next to the cowboy, cattle with their heads and horns down are driving as hard as they can away from the storm, which already is beginning to douse them with sheets of rain. The sky behind the cowboy and the herd is blackening. One long white streak of lightning is striking near another cowboy and cows in the misty distance, which is murky, a green and black haze of rain and storm. Everything about the painting conveys the threat of chaotic danger. If the cowboy’s pony trips, or throws him to the stony ground, the unfortunate man will be ripped by the horns of the charging cows or pulped by their heavy hooves.

Petraeus included a copy of the Remington painting in a briefing on “The Mesopotamian Stampede” he would give to members of Congress and other visiting Americans. It is “a metaphor really of the need to be comfortable with slightly chaotic circumstances,” Petraeus explained, seeming a bit uneasy, perhaps because of the role the image assigns Iraqis. “A stampede is not always orderly. In that particular painting the ground is rugged, the wind is howling, it’s raining cats and dogs, there’s lightning—and you can use the lightning as a metaphor, it could be an IED, it could be a tasker from higher headquarters, it could be some sort of political challenge in Iraq, who knows what it might be. And the concept of outriders and trail bosses—again the concept of the challenges on the trail, the idea that some issue, some cattle, some tasks, will actually get out ahead of us. They will move on their own and that’s fine. We will catch up with them. But some will also fall behind and we will have to go back and round those up. That some cattle are killed along the way. There’s bad guys out there, rustlers who are trying to kill us and to kill those in the cattle drive. And you can use the cattle to represent any number of different items, from the ISF—the herd is growing, they are getting stronger. There are Iraqi trail bosses out there with us, and we are gradually handing off more of the responsibility for the cattle drive to them.”

He also used the painting to convey to his subordinates his notion of command. “I don’t need to be hierarchical,” he explained. “I want to flatten organizations. I’m comfortable with a slightly chaotic environment. I know that it’s okay if some of you get out ahead of us. Some of the cattle will get out ahead and we will catch up with them. And some will fall behind and we will circle back and we won’t leave them behind.” He didn’t show the image to Iraqis, he said. It was more useful with Americans. “We’re just trying to get the cattle to Cheyenne.”

Lt. Col. Nielsen, one of his aides, added that, in her view, the image is also about the limitations of high command. “A lot of it is about intent, about setting parameters, and an incredible decentralization,” she said. The message, she said, is, “I can’t tell you exactly what to do,” because Iraq simply was too chaotic.

Petraeus adopted a posture of much lowered expectations, and as was his wont, set the tone for his entire command. One of his most striking characteristics is his ability to discern and evaluate the reality of events. That isn’t as easy as it sounds, and it is especially difficult to pick out reality through the fog of war. The first and foremost task of a commander is to understand, with a steady head, the nature of the conflict in which he is engaged. In order to achieve that understanding a commander can be neither overly optimistic nor pessimistic, and especially, not subject to McClellanesque mood swings, seeing every minor victory as a triumph and every partial setback as disaster.

Even more important, Petraeus injected a new spirit into senior commanders. At his first meeting with his division and brigade commanders and senior staff members, in February, he sought to convince them they could succeed. “I was amazed with what Petraeus did,” recalled Keane, who attended the meeting. “He took over a command with a sense of futility and hopelessness about it and almost overnight he changed the attitude and he brought them hope and a sense that we can do this, we can succeed at this.”

Crocker brought a far different self-image to his partnership with Petraeus. In keeping with the morose outlook that led President Bush to dub him “Mr. Sunshine,” he joked once that he saw the general and himself as resembling the lead characters in a movie about two convicts on the run from a chain gang, “shackled” together and so forced to cooperate. He seemed to be referring to The Defiant Ones, a 1958 film starring Tony Curtis and Sidney Poitier.

Both Crocker and Petraeus had served in Iraq before but didn’t know each other until early 2007, when both arrived for their current tours. Crocker’s thought after their initial meeting then was, “I had just gotten very very lucky, given his ability, his drive, his experience, and his intellect.”

At the embassy, Crocker began to oversee and revive a staff that Keane found lethargic. “The whole attitude of the place changed” after Crocker arrived, Keane said. “They had passion. They were taking personal risks. They were connecting with Iraqi officials.”

SMALLER GOALS . . .

One of Emma Sky’s fears in returning to Iraq was that she would be subjected to endless rounds of happy talk at top American headquarters, as she had on her previous tour in Iraq, when she worked in the northeast. “They say ‘Camp Victory’ without any sense of irony,” she noted archly.

Instead she was surprised to walk into a marathon conversation among top commanders and advisers about how to lower the goals of the mission. In the course of several weeks early in 2007, she said, “We redefined success in a much more modest way as ‘sustainable stability.”’ This was key: The grandiose goals of the past three years, of turning Iraq into a beacon of democracy that would transform the Middle East, or even of turning Iraq into a dependable ally of the United States, were quietly put on hold. Bush administration rhetoric didn’t always reflect this shift. But on the ground in Iraq, the new goal was simply getting to a more or less peaceful Iraq that didn’t explode into a regional war or implode into a civil war.

As Odierno, Sky, and others talked into the night, hours at a time, three or four nights a week, they focused on the way that parts of the Baghdad government exercised power to further sectarian agendas, undermining the legitimacy of the entire enterprise. “It is a failed state with ungoverned spaces in which the government is part of the problem,” Sky summarized as their conclusion. In particular, they would target Shiite militiamen employed by the Ministry of Health, who among other things were killing Sunnis who sought medical care.

They also decided that they needed to reposition the U.S. government. In February, Odierno would tell his subordinate commanders to conduct “balanced operations targeting groups on both sides of the sectarian divide.” That is, rather than act as an ally of one side, the Shiites, they would recast the American role in Iraq as an arbiter between groups.

As part of that move, Odierno ordered the abandonment of the term “AIF,” for “anti-Iraqi forces,” an Orwellian designation that U.S. officials had given to insurgent groups, as if Americans could decide who was a real Iraqi. They also would carefully release certain leaders of insurgent groups to see if they might begin to cooperate. The message to them would be that the U.S. government recognizes their concerns, which are legitimate, and will work with you, as long as you don’t use violence against us. Finally, they decided that the key indicator of progress in security was Iraqi civilian casualties, not those inflicted on the American and Iraqi militaries.

Odierno also discussed with Keane what do to about Sadr City, the Bronx-sized slum in eastern Baghdad dominated by Moqtada al-Sadr. Keane and he concluded that “we should avoid Sadr City,” and try to deal with it later politically, instead of engaging in another round of block-to-block fighting in a huge neighborhood of hostile Iraqis.

They also decided that there was a hole or a gap in the middle of Iraqi society. The people had needs, especially for security, but the Iraqi government couldn’t provide it, so that opening was being filled by militias. “We need to step into that gap,” Odierno ordered. The way to do that, he said to his advisers, was “to get back out into Baghdad—I want to get my people out there.” In effect, they had reversed the American policy of the previous three years.

In April 2007, Maj. Gen. Fastabend, Petraeus’s strategic adviser, composed a twenty-page essay, “How All This Ends”—that is, the answer to the question Petraeus had posed four years earlier—that captured the revamped approach: The United States, he wrote, needed “to settle for far less than the vision that drove it to Baghdad.”

... AND BIGGER RISKS

The other shoe, Fastabend continued, was to take far bigger risks. He subtitled his essay “It’s Fourth and Long, Go Deep.” In the essay, which isn’t classified but has been held so closely that its existence hasn’t previously been disclosed, he employed the literary device of having Petraeus look back from the future—2009—to recount how he had turned around the situation in Iraq. Never one to waste a moment of his time, Petraeus kept a copy of Fastabend’s essay next to the toilet in his private bathroom, taking it out occasionally to refresh his thinking.

It can end well, Fastabend explained, if the U.S. government would take more risks. But to take risks, we have to think seriously, he continued. Few anti-war critics were as scathing of the conduct of the Iraq war as are members of Petraeus’s staff, such as Fastabend, his chief of strategy. “As a sole superpower, we thought we didn’t have to make hard choices. We thought we could just come here, without thinking about the opportunity costs. When you just write conditions, and never have to say who does what by when—then you don’t make choices and decisions. All you get is conditions: Close the border, end corruption, change the culture.”

It was time, he told Petraeus, “to take some risk—not the ones you’re comfortable with, but gut-wrenching, hold-your-balls risks.”

He recommended six major departures:

• Get rid of extremists by working with them. We had been fighting them for four years, he said, “whacking and stacking them”—but with little to show for all that blood, sweat, and tears. Maybe, he suggested, it was time to replicate the example of Ramadi and cut some deals with tribal leaders and other insurgent organizations. Tell them they aren’t militias, they are neighborhood watches. Parole insurgents to them. “Commanders will object—‘catch and release.’ There will be letters from mothers and fathers—‘They killed Americans.’ You’ll have to take some heat. Make a choice!” Fastabend even called for large-scale detainee releases, which would be considered but eventually was shelved. “Petraeus was comfortable with it, but division commanders weren’t,” Col. Rapp explained. Petraeus agreed to drop the idea because of their concern that it would damage troop morale. But he would go on to implement the idea of local cease-fires with former insurgents.

• Another major risk Fastabend recommended taking was alienating our own allies, the Shiite-dominated central government. Push Maliki hard. And don’t let him shut down the deals with former insurgents. Like Odierno, Petraeus was ready to go further there. In the following months, American commanders would sign up tens of thousands of former insurgents to become local militias, first called Concerned Local Citizens and then later, Sons of Iraq.

• Third was reaching out to Moqtada al-Sadr. Part of these negotiations were even about whether to talk to each other. “They said, we want a date for your exit,” recalled Kilcullen, who was briefed on the initial exchange with Sadr’s representative.

“We can’t do that,” the Americans replied.

“Well, forget it then,” a Sadrist politician replied.

But the Americans were curious. “What date did you have in mind?”

“Well, December 2012,” the Sadrist said. That brought private grins to the Americans—promising to stay in Iraq until then was a position that would have drawn protests from many in the U.S. Congress.

• Fourth was beginning to emphasize reconciliation at the local level, among Sunis and Shiites in towns and provinces, rather than a deal among national leaders, which had hit a dead end. Petraeus and Crocker would go along with this.

• Also, he argued, put the brakes on the transition to Iraq control, stopping the cycle of rushing to failure. “Casey was all about transition,” said Fastabend. “Petraeus has slowed it down.” One risk that surprised Fastabend was how dangerous it was to deal with the Iraqi government. While scheduling appointments, Americans had to worry about whether the government official who was being met would tip off the insurgents to set up an ambush.

Fastabend was even more bothered by how the government had reacted a few weeks earlier to the suicide bombing of the Iraqi parliament, an incident that killed eight in the worst breach of security the Green Zone had ever suffered. One Iraqi official told him, “We’ll show them, we’ll meet tomorrow.”

“Well, I was just over there, and it needs to be cleaned up,” Fastabend responded.

“Oh yeah, it will be,” anotehr Iraqi promised.

He went back the Council of Representatives building the next morning just to check. It was still a bloody mess. “The motherfucker’s legs were still on the floor, and parts of him were all over the walls,” he recalled. He called one of the Iraqis: “You think you’re going to walk the press through here in twenty minutes?”

• Finally, Fastabend was mulling something that was politically explosive back home. “I think you announce a withdrawal schedule.” Moqtada al-Sadr couldn’t live with a big U.S. force in Iraq, which he deemed an occupation. But maybe, Fastabend calculated, he’d want a small presence, of 5 brigades, or perhaps 35,000 troops, just to keep him safe from the Sunnis. “Some people say he wouldn’t accept it. I say, ‘Take a risk.”’ He explained: “So we say to him, we’ll give you a timetable. Maybe come down to twelve [brigade combat teams] by the end of 2008, and five by a year later.”

His bottom line: “If this is the decisive struggle of our time—be decisive.” Fastabend didn’t think he was asking too much. “Considering that the alternative is getting chased out of here, it doesn’t seem that audacious to me.”

There are two kinds of plans, he explained—those that fail and those that just might work. “If we fail, we are getting ready for a pretty major civil war, leading to a regional war. If things work, we are in an accommodation phase” that might lead to a preserved Iraq. So, he said, play according to the stakes.

Fastabend was correctly pointing to a major flaw in the American approach in the war from 2003 to 2006. For years, U.S. commanders had tended to seek strategic gains—that is, winning the war—without taking tactical risks. They ventured little and so gained less. By making the protection of their own troops a top priority, and by having them live mainly on big bases and only patrol neighborhoods once or twice a day or night, they had wasted precious time and ceded vital terrain to the enemy. Also, their priorities undercut any thought of making the protection of Iraqi civilians their mission. That was literally seen as someone else’s job—Iraq soldiers and police.

Capt. McNally, who studied the Fastabend’s essay, concluded that its core message was “It’s put up or shut up time.” The way forward it recommended, she explained, was, “Take risks. Otherwise, we just keep going along, and we lose six soldiers once a week when the Strykers [wheeled armored personnel carriers] get blown up.”

One risk Fastabend didn’t mention in his essay was the internal one Petraeus was taking on as he clashed with his immediate superior, Adm. William Fallon, who had succeeded Abizaid as chief of the Central Command, the U.S. military headquarters for Iraq and the rest of the Middle East. Petraeus was determined to speak his mind, leading to what would amount to a running feud with Fallon, his ostensible boss.

A FOUNDATION FOR STRATEGY

It is axiomatic that good tactics can’t fix a bad strategy, but that a good strategy tends to fix bad tactics, because the inappropriateness of those individual actions becomes self-evident when seen against the larger scheme. For example, in a mission where the top priority explicitly is protecting the people, there would be no excuse for an incident like Haditha.

The biggest single strategic change in Iraq in 2007, the one that preceded all others and enabled them, may also have been the least noticed one: a new sobriety in the mind-set of the U.S. military. It wasn’t just the Bush administration that had taken years to face reality in Iraq. The military also was slow to learn. McMaster’s successful campaign in Tall Afar in late 2005, for example, seemed to be largely ignored by top commanders, or dismissed as irrelevant. Despite the attention given to Tall Afar by the media, there seemed to be no concerted effort in the Army to discern if the success there might be replicated elsewhere. By the beginning of 2007, though, the U.S. military had been fighting in Iraq longer than it fought in World War II. It had been flummoxed and humbled by its struggle in the Land Between the Rivers, trying nearly everything in its toolbox of conventional methods, and not finding much that promised a successful outcome. Finally, it was ready to try something new.

It had to come a long way. In the feel-good days after the fall of the Berlin Wall and before 9/11, and even for some time after, when the U.S. military was the armed wing of “the sole superpower,” Pentagon officials liked to talk about “rapid decisive operations.” That was a term for, as one 2003 study done at the Army’s School of Advanced Military Studies put it, the devastating cumulative effect of “dominant maneuver, precision engagement and information operations.” The technocentric notion behind it was that U.S. forces, taking advantage of advances in sensors, communications, computer technology, and long-range weaponry and precision logistics, all areas in which it excelled, would fight so quickly and adeptly that the enemy would never have a chance to catch up and understand what was happening. Blinded, confused, and overwhelmed, the enemy’s will would break, U.S. forces would triumph, and everyone would live happily ever after. “We need rapidly deployable, fully integrated joint forces capable of reaching distant theaters quickly and working with our air and sea forces to strike adversaries swiftly, successfully, and with devastating effect,” Defense Secretary Donald Rumsfeld said in January 2002. Thus, he continued, we would possess “the option for one massive counteroffensive to occupy an aggressor’s capital and replace the regime.”

One of the people in the audience that day was Gen. Tommy Franks, then the chief of the U.S. Central Command, and 14 months later the commander of U.S. forces invading Iraq. The “rapid decisive operations” approach culminated in Franks’s plan for going into Iraq, in which he sought to substitute speed for control. “Speed kills” became his mantra, repeated endlessly to his subordinates. But as U.S. forces found after they raced from Kuwait into Baghdad, speed could temporarily substitute for mass in military operations but wasn’t the same as control. Once the Americans got to the capital, they stopped moving. Lacking both mass and velocity, they soon lost control of the situation.

Col. McMaster would argue later that the very concept of rapid decisive operations had hamstrung American commanders as they entered the country, because it had “artificially divorced war from its political, human, and psychological dimensions. So, if flexibility hinges on a realistic estimate of the situation going into a complex situation, we were behind at the outset.” It would only be after American commanders and strategists began paying attention to the most basic human elements—tribes, blood feuds, and fights over water, money, and women—that they would begin to understand the war they were in, which Clausewitz maintains is the first and most important task of the military leader.

“Our mindset was not to kill, it was to win,” recalled Lt. John Burns, who led a scout platoon in Baghdad during the Petraeus counteroffensive. “We constantly evaluated our situation and made certain we were fighting the war we had and not necessarily the one we wanted.”

But that sort of seasoned understanding would come only after four years of struggle that more often was counterproductive than not. As the war for Iraq began in earnest in the summer and fall of 2003, U.S. commanders, surprised by the intensity and duration of the resistance to their presence, emphasized capturing and killing their enemies. But every time they captured key leaders, more seemed to spring up. By 2007 the military had realized that this approach was not leading toward success. “I think if the last four years in Iraq show anything, it’s that you can’t get by on brute force alone, and our generals should understand that by now,” Col. Mansoor, Petraeus’s executive officer, said one day in Baghdad late in 2007.

Under Petraeus, many did indeed get it. “You can’t kill your way out of this kind of war,” said Lt. Gen. James Dubik, in a comment that many repeated that year.

In remaking itself in the 1970s and ’80s as a blitzkrieg force, the Army may have repeated the mistakes of the German army of World War II, observed Andrew Krepinevich, the defense intellectual who wrote the seminal work on the Army’s failure in Vietnam. “In World War II, the Germans were very good tactically, but they were terrible at the strategic level,” he said. Thus the rebuilding of the Army during the 16 years from the fall of Saigon to the beginning of the 1991 Gulf War, rather than being new and innovative, may actually have signaled the end of an era. As retired Army Col. Bob Killebrew, a thoughtful strategic thinker, put it, “We may well look back on the ’90s as the final spasm of blitzkrieg.” That observation casts a new light on the two “thunder run” tank charges that the Army used to penetrate Baghdad during the invasion: They may have been not the harbinger of a new, more agile Army, but rather a last blaze of glory for the heavy conventional force, a miniature version of its glory days of 1944-45 in Europe and 1991 in Kuwait.

After four years of failure in Iraq, the U.S. military began to find effectiveness—at least tactically—as its leaders finally became resigned to the reality that the way to success was conducting slow, ambiguous operations that were built not around technology but around human interactions. “Be deliberate,” Odierno would order his subordinates. Show “tactical patience,” advised a brigade commander. It became common to hear American commanders counsel their frustrated soldiers to take it “Shwia, shwia”—Arabic for “slowly, slowly.” As the new counterinsurgency manual said, they needed to be prepared to take years to succeed. The key to this mode of warfare was slowly seeking accommodation, pulling the population over to one’s side, even if that sometimes meant cutting deals with people who had killed American troops. As Emma Sky said one day, “We are dealing with guys with blood on their hands.”

Sky, who had advised the U.S. military for a year in 2003-4, saw the commanders as having an entirely new mind-set in 2007. “In ’03, the guys were Christian crusaders seeking revenge for 9/11. Today they are advising Iraqis in a way they couldn’t back then. They have completely transformed the way they work with Iraqis. It is a tremendous change. It’s not just the Sunnis or the Shiites who have changed. We all have changed.”

The entire approach was distinctly alien to the rapid, decisive, mechanistic, and sometimes Manichean mind-set that had been taught to a generation or two of American commanders. It had nothing to do with technology and everything to do with dealing with some of the oldest of human traits—eye-to-eye contact and heeding the values and ways of tribes and their leaders. What was going on in Iraq in 2007, as Kilcullen put it, was “a counterrevolution in military affairs, led to a certain extent by David Petraeus.”

The pre-Iraq, triumphalist U.S. military also was fond of talking about “information dominance.” What this tended to mean in reality was amassing data rather than understanding. For most of the time the U.S. military has been in Iraq, it actually has tended to be information poor. As Warren Buffett, the wise billionaire, once observed, if you’ve been playing poker for half an hour and you don’t know who the patsy at the table is, then you’re the patsy. Too often, U.S. troops, cut off both linguistically and physically from the Iraqi populace, operating in a harsh climate amid an alien culture, had been made patsies. This was not their fault but that of their leaders who didn’t understand the task at hand of conducting a counterinsurgency campaign.

Looking back, Maj. Mark Gillespie, a military adviser in Iraq, recalled that in early 2006, he was “reaching terminal velocity and pulling my hair out and trying to figure out why people just don’t get it. Well, it wasn’t them who weren’t getting it, it was me who wasn’t getting it.”

American soldiers would really only start getting the requisite amount of information after they moved out into the population in 2007. In retrospect, this seems like common sense. After all, Clausewitz, often seen incorrectly as the most conventional of war theorists, notes that the people are the greatest single source of information available. “We refer not so much to the single outstandingly significant report, but to the countless minor contacts brought about by the daily activities of our army,” he explained.

The new humility of American commanders amounted to the starting point for the new strategy. After trying it their way for years, they now were going to try it the Iraqi way. So, for example, rather than try to build on their own individualistic core values of freedom and “one-man, one-vote” democracy, they began to rely on Iraq’s more communitarian values, which often revolve around showing and receiving respect. “They felt disrespected, dispossessed, and disgusted,” Petraeus said one day. “All they wanted was”—he began singing the letters in the old Aretha Franklin hit—“R-E-S-P-E-C-T.” (Indeed, one of the new Iraqi political parties that would form called itself “Dignity.”)

With humility came its twin, candor. “There’s a more open environment now,” Capt. McNally said. “People used to maybe think [negative] things, but they didn’t say them.”

This new sobriety was the intellectual context for the reduction in the goals of the war. This is a controversial point, because that shrinkage has never really been announced or even acknowledged. But it was put into practice every day as a smaller, narrower set of aims. The goal was no longer the grandiose one that somewhat murkily grew out of the 9/11 attacks and was meant to transform Iraq and the Middle East—what the old Wolfowitzian Iraq hawks had called “draining the swamp” in which terrorism grew. Instead, the quietly restated U.S. goal was to achieve a modicum of stability, to keep Iraq together, and to prevent the war from metastasizing into a regional bloodbath. That meant finding what one official called “a tolerable level of violence” and learning to live with it.

“Not rhetorically, but in practice we have” limited the goals of the U.S. effort, Mansoor said one day early in 2008. Trying out a phrase Petraeus would use publicly four months later, he said, “We are willing to accept less than a Jeffersonian democracy. . . . The rhetoric of our national leadership is still about freedom, but on the ground, there’s a realization there is going to have to be Iraqis figuring this out.” (In April 2008, Petraeus would tell the House Foreign Affairs Committee: “In terms of what it is that we are trying to achieve, I think simply it is a country that is at peace with itself and its neighbors. It is a country that can defend itself, that has a government that is reasonably representative and broadly responsive to its citizens, and a country that is involved in, engaged in, again, the global economy. Ambassador Crocker and I, for what it’s worth, have typically seen ourselves as minimalists. We’re not after the holy grail on Iraq; we’re not after Jeffersonian democracy. We’re after conditions that would allow our soldiers to disengage.”) Petraeus began keeping an eagle eye on the president’s speeches, using their weekly video teleconferences to convey caution against inflating the rhetoric. He usually succeeded but not always.

There was good reason for this quiet ratcheting down. As Steven Metz, an astute strategic analyst, put it, encouraging democracy was at odds with the larger goal of stability: “Our current strategy is based on the delusion that we can have stable, or modulated democratization,” he said. “Few things are more destabilizing and prone to chaos than democratization. I think we can have either democratization or stabilization. The issue is whether we can tolerate several decades of often-violent instability while democratization takes root.”

THE HARDEST STEP

The surge really began even before the first of the surge brigades arrived. That may sound paradoxical but isn’t, because the surge was more about how to use troops than it was about the number of them. The first new brigade wouldn’t fully arrive until February, but as the bombings increased in January, the 1st Cavalry Division, which already was in the country, escalated its efforts to protect the population, seeking new ways to protect markets, neighborhoods, main roads, and bridges, said Col. Tobin Green, the division’s chief of operations, and a friend and former student of Eliot Cohen. “I believe that was a turning point,” Green said, “a visible sign of commitment to protecting the Iraqi people.”

Moving American soldiers from big isolated bases and into new posts of 35 men (if platoon-sized) to around 100 (if manned by a company) located in vacant schoolhouses, factories, and apartment buildings in Baghdad’s neighborhoods was the hardest step. Essentially, U.S. forces were sallying out to launch a counteroffensive to retake the city.

Seeking to translate the strategy into operational and tactical sense, Odierno was looking downward, monitoring the adjustments of subordinates from division commanders to platoon leaders. “That’s especially difficult with units that were already there,” recalled Keane, his mentor. “He was transitioning those forces from a very defensive strategy to an offensive strategy.” On top of that, having only the minimum amount of troops that he and Keane thought he needed, Odierno began to move them around in order to maximize their effectiveness. “He took risks,” Keane said. “The easy thing would have been to put all the surge brigades into the city.” Instead, following the “What would Saddam do?” approach, Odierno put much of his combat power outside the capital. This was the biggest difference between Odierno’s plan and the one Keane and Kagan had pulled together at the American Enterprise Institute. Eventually, he would split his total available combat power evenly between the city and its surroundings, with six brigades in each.

In February, the 2nd Brigade of the 82nd Airborne Division, the first official surge brigade, was sent into eastern Baghdad. Over the next several weeks, 19 new outposts were established across Baghdad. “Get out of your Humvees, get out of your tanks, your Brads, and walk around,” Army Maj. Joseph Halloran, an artillery officer, later summarized. “Stop commuting to war. . . . The concept of a super FOB [forward operating base] is more damaging to the war effort than any Abu Ghraib or Haditha incident could ever be.”

The first days were surprisingly violent, with an average of almost 180 attacks a day on U.S. forces. “That was the battle of Baghdad,” Petraeus said looking back 18 months later. “It was just very very difficult, very very hard.” During February 2007, Baghdad suffered an average of more than one car bomb attack a day. Between late January and late February, at least eight U.S. helicopters were shot down.

In March, the second surge unit, the 4th Brigade of the 1st Infantry Division, began operations in western Baghdad. One skeptical soldier from the Big Red One told a reporter that he didn’t expect the new approach to work. “It’s getting worse and worse,” he explained to the Washington Post’s Joshua Partlow. “They don’t even respect us anymore. They spit at us, they throw rocks at us. It wasn’t like that before.” In some Shiite neighborhoods, units were greeted by stacked loudspeakers blaring the chants of the Jaysh al-Mahdi, Moqtada al-Sadr’s militia. In Sunni neighborhoods that had been ethnically cleansed, patrolling soldiers often found piles of executed bodies and vacant houses with blood smeared on the walls.

This is how the operations officer for a battalion operating in southwest Baghdad recalled that time to a researcher for the official Center for Army Lessons Learned:

When we first moved into the AO [area of operations], it was house-to-house clearing, and fighting most of the way. It took months before we could drive more than halfway north through the mulhullas without hitting multiple IEDs and taking fire. It got so bad that we twice had to turn over part of our battlespace to Strykers [wheeled armored vehicles]. We focused on establishing a foothold by clearing house by house and holding with a 24x7 presence. Then we began establishing our HUMINT [human intelligence] sources, pulling out bad guys, and building relationships with the people. We also focused on splitting the insurgency. It was composed of two main groups. First were the local mujahadin, who were truly concerned about protecting the neighborhood from the Shiite Militias, particularly Jaysh al-Mahdi (JAM). They were generally actually concerned with the people. The second element was Al-Qaeda in Iraq (AQI). The people were getting tired of all the violence in the neighborhoods, of things blowing up and getting innocents killed. Every time something happened, we’d say “AQI did this too you. Why do you allow it?” When we lost people, we’d stay restrained and not seek vengeance.

The first task was simply surviving. “Our first two weeks were tough,” Lt. Jacob Carlisle, a platoon leader, later said. “We had to clear every day, and we got hit every day.” Indeed, in June, he would be shot in the thigh and hit by shrapnel in the face and arm.

Not all soldiers liked the shift into the population. “My platoon sergeant came to Iraq with the idea that we were going to hide for fifteen months and all come back alive,” recalled Lt. Schuyler Williamson of the 1st Cavalry Division. “When I told him that we were not going to do that, he said I was going to get my soldiers killed.” The balky sergeant eventually was reassigned, Williamson added.

Lt. Col. Crider led his cavalry squadron into the Doura neighborhood in southern Baghdad and lost three soldiers in one week. “We did not know who was responsible for these attacks, and no one would tell us anything,” he recalled. “Our partnered National Police unit was no help as the residents of Doura, our predominantly Sunni neighborhood, hated them.” In fact, he remembered, the locals referred to the police as “the militia.” Bringing them into the neighborhood was seen as a hostile act.

“Doura was a meatgrinder,” recalled Command Sgt. Maj. Marvin Hill.

When Baker Company, a unit in the 2nd Infantry Division, moved into there, it was greeted with “constant enemy small-arms fire, IED, RPG, and grenade attacks, often surprisingly coordinated,” recalled Lt. Tim Gross, a platoon leader. Baker began by spending three nights using shovels, screwdrivers, and tire irons to remove 18 “deep-buried” bombs in its area. The soldiers lied to the locals that they knew where all the bombs were because they had so many local sources of information. “We don’t need any more information because we had hundreds of people cooperating,” was the bluff, as Capt. Jim Keirsey, Baker’s commander, recalled it. In fact, they began with almost no information from the people of the area, who had felt abused by Iraqi police operating in the area. Indeed, Baker later would ban the most abusive of the police, the militia-infested National Police, from entering the neighborhood.

Despite being attacked constantly, Baker Company, with roughly 125 men, began conducting patrols around the clock. It tried to be precise in the use of force. “Shooting the right guy teaches the enemy and population that evil has consequences,” Keirsey wrote. “The corollary is that a poor shot—one that hits an innocent person or leads to collateral damage—is worse than not shooting at all.”

Gross, the platoon leader, called this mind-set “protect the innocent, punish the deserving.” He said it especially impressed the locals when one of his platoon’s patrols, while amid civilians, was ambushed. After a girl was hit, his platoon sergeant picked her up and rushed her to medical care. “An informant reported the incident as a large turning point towards winning the people of our neighborhood,” Gross said.

There was a new savviness to the way American forces were operating. Baker Company’s most effective tactic didn’t involve firepower but instead walking and talking. Its soldiers conducted a thorough census that mapped the 3,500 households in its area of operations, photographing all male inhabitants and collecting their grievances. Dubbed “Operation Close Encounters,” it was done slowly and carefully, with some interviews lasting an hour. Keirsey ordered that the soldier doing the talking should sit down, take off his helmet and sunglasses, accept any drink offered, and speak respectfully. The other members of the patrol should stay in uniform and quietly focus outward on security, rather than join the conversation. In this way, they learned about suspected bomb planters and about Iraqi police abuses. As a result of ethnic cleansing, there were many empty houses. Rather than let them be used by militias, the American troops padlocked their doors and gates.

They also were told that while the area was controlled by insurgents, U.S. funds had helped finance the enemy, because the insurgents got kickbacks from contractors. “People are getting rich working with Americans in Iraq,” said Keirsey. “Make sure they are the right people.” It was an important lesson, but not one that many American officials had heeded in earlier years.

Visiting Iraq, Keane saw not only Petraeus and Odierno but their division and brigade commanders. He would push them. “How many of your platoons are outside the FOB and on the street twenty-four/seven?—that was always a huge dimension for me. And some of those guys would be hedging—they would have one-third of them out there. I said, ‘No guys, you would have to have two-thirds, for sure. And if you could, get them all out there and be protecting yourself back in the FOB using someone else,’” such as support units or contractors.

By May 2007, the 1st Cavalry Division, which was the core unit for Baghdad, at any given point had 75 percent of combat forces off its headquarters post, said Maj. Gen. Joseph Fil Jr., the division commander. The typical cycle for a unit was five and a half days out, followed by one and a half back on post to rest, refit, check e-mail, and clean up. Having troops live where the action was added enormously to their effectiveness, not only in increased awareness but also simply in response time. “You’re not driving and hour and a half to do a ninety-minute patrol,” Fil said.

The 1st Cav’s 1st Brigade, stationed north and northwest of Baghdad, set out to eliminate al Qaeda’s safe havens and crack down on the networks sending car bombs and roadside bombs into the capital. But at first it didn’t feel it had enough troops for those tasks. “I was frustrated because the only thing we were doing was terrain denial—we were so strung out securing the LOCs,” or lines of communication, said Maj. Patrick Michaelis, the brigade’s S-3, or chief of operations. With the troops he had, he explained, trying to keep major roads clear of ambushes and bombs was “all we could do.” Dozens of soldiers were killed. The unit didn’t begin to feel the effects of the surge until mid-May, after it was given an additional battalion from another division, he recalled. Thus reinforced, its operations against al Qaeda would become a model that Petraeus would cite, as the brigade pushed into the areas where al Qaeda fighters and their allies had found sanctuary. “We fought our way in,” he recalled. The enemy was ready, having deeply buried bombs on the roads in the area. One had a full 1,200 feet of copper wire leading to the trigger—far further than U.S. forces were trained to look for the triggerman.

At almost every new outpost established, a series of fights and terrorist actions would ensue. Sentries found it difficult to stop truck and car bombs barreling toward them with mere rifle fire, so were issued bazooka-like anti-tank weapons. Frequently, al Qaeda would overreact to the new patrol bases, said Maj. Luke Calhoun, the brigade’s intelligence officer: “They’d kidnap children, kill women, threaten tribal leaders.” But that counteroffensive usually backfired, he said, because the population was driven into the arms of U.S. forces, who now were available to them 24 hours a day in the new outposts.

One of the hardest hit areas was the town of Sab al-Bor, which had a population of about 60,000. In August 2006, five months before the surge got under way, al Qaeda had begun shelling the town, located on the northwest fringe of greater Baghdad, with big 120-millimeter mortars, aiming at the primarily Shiite northwest corner of the town. But that was the only major security problem with the town, and U.S. forces were facing bigger issues elsewhere as the small civil war grew. In late September 2006, the town was turned over to Iraqi police, “so I could pull B Troop and the IA [Iraqi army] out of the town and move them to other, hotter areas,” recalled Col. James Pasquarette, who commanded the U.S. Army brigade based nearby. But on October 3, soon after that move, the mortar attacks escalated. Shiites in the town retaliated by shooting up Sunni neighborhoods. Thousands fled the town, including the Iraqi police. Soon only about 5,000 inhabitants remained. The young male Shiites who were displaced became willing recruits for Shiite militias, which intensified the cycle of violence.

The turnaround for the 1st Cav brigade, commanded by Col. Paul Funk, began in 2007 after a Marine unit moved to the west side of the brigade’s sector, cutting off al Qaeda’s roads south to Fallujah and north to Tamariyah, Samarra, and Tikrit. Almost instantly the mortar shelling of Sab al-Bor ended. U.S. Army engineers purposely weakened a major bridge so that pedestrians could cross but not vehicles, and the car bombings stopped. By October 2007, Michaelis said, al Qaeda seemed to make a strategic decision to retreat northward to Mosul.

The improvement in security provided multiple benefits. During that period, more and more local militias came over to the American side. Turning over checkpoints and outposts to them freed up the 1st Cav units for other missions. Also, the locals began providing precise intelligence. “The info we were getting from the CLCs [Concerned Local Citizens] was phenomenal,” recalled Michaelis. If they said there were six bombs in a road, and American explosives experts only detected five, the local fighters would insist that they had missed one—and would be proven right. In October, representatives of local Sunni and Shiite militias that had turned also met jointly with representatives of the Iraqi government in what the 1st Cav labeled “The Northwest Baghdad Regional Security Summit.” Michaelis remembers thinking that day, “That’s what ‘right’ looks like.” Finally, elements of local governments began to surface. “We started seeing guys pop up. ‘I’m the water official for the district.’ ‘Well, where the hell have you been for the last fifteen months?’ ‘There’s no way I was gonna stick my head up when al Qaeda was gonna kill me.’ ”

By the end of the unit’s tour that winter, in late 2007, Michaelis said, “We’d start seeing video shops, Internet shops, cigar shops. These are not things you buy when you are at the low end of Maslow’s hierarchy of needs.” And the population of Sab al-Bor was back up to 21,000. The brigade’s transformative experience would be replicated in a dozen other areas in and around the capital in 2007.

THE ENEMY COUNTERATTACK

But anyone still alive to fight the Americans in Iraq in 2007 had learned a lot in the preceding years. In the spring and summer the American surge was met with a counteroffensive involving new tactics and more lethal weapons. This was arguably the toughest period of the war, as the Americans took their last and best shot only to see casualties increase without many signs of security improving. At the very least, it was the hardest part of Petraeus’s time in command in Iraq. At the time he put a positive spin on it while speaking in public, calling himself a “qualified optimist.” But much later he would admit, “There were days that were about the hardest that I have ever experienced.” The U.S. military had committed its reserve. It was taking more risks and losing more people. As the war entered its fifth year in March 2007, there were few signs that the gamble of the surge was paying off, either tactically or strategically.

In north Baghdad, Company C of the 1st Battalion, 26th Infantry Regiment, 1st Infantry Division, spent most of its tour of duty—11 months out of 15—in the heavily contested Sunni neighborhood of Adhamiyah. On May 14, 2007, a bomb blew one of its Humvees into the air. “I never thought I was going to see my buddies running around on fire,” said Staff Sgt. Octavio Nunez, one of two soldiers who would receive the Silver Star for valor that day.

The bombs grew more powerful: In June a Bradley Fighting Vehicle in the same company was hit by a huge explosion, flipping the 25-ton armored vehicle and killing 5 soldiers. The bomb had been placed not far from an Iraqi army checkpoint, a point not lost on the American soldiers. The gunner, Spec. Daniel Agami, was pinned beneath the vehicle. His comrades could hear him scream as he burned to death. Another member of the battalion, Pfc. Ross McGinnis, would posthumously receive the Medal of Honor for jumping on a grenade that lodged inside a Humvee. All told, the battalion lost 31 men during its tour, nearly half of them from Charlie Company.

Col. Galloucis, commander of the MPs in Baghdad, shook his head as he remembered the violent spring of 2007. “We started seeing the introduction of EFPs into Baghdad,” he said, referring to a particularly lethal kind of roadside bomb, the “explosively formed penetrator,” that can lift off the ground heavy vehicles such as the tanklike Bradley Fighting Vehicle. If the entire Iraq war was characterized by roadside bombs, as the spirit of World War I was captured by trench warfare and machine guns, then the spring of 2007 was the campaign of the EFP. These armor-piercing bombs were only the size of a coffee can, and so could be placed very quickly, unlike the big bombs that required much digging. They were used almost exclusively by Shiite militias. The bombs were manufactured in Iran, with the number radically increasing late in 2006, said U.S. officials. “They are harder to make than you think,” said one American bomb expert. Designed to fire a spearlike slug of melting metal at extremely high speed, the bombs didn’t work if milled imprecisely, which causes the metal to fragment prematurely and so diminish in lethality.

Galloucis’s troops were also facing a sniper threat, and that “was having a real psychological impact,” as well as a physical one, because some of the shooters were using armor-piercing rounds that would penetrate American body armor. He remembered moments of despair—“You had a sense that things weren’t working, that whatever we’d do, they’d counter.”

Crider, the cavalry squadron commander in southern Baghdad, soon realized that the time-honored tactic of simply cordoning off an area and searching it not only antagonized the very people whose support they needed but also turned up few signs of the enemy. “Insurgents have learned over five years not to hide things in their homes,” he commented.

U.S. military intelligence officers began to see assaults on Americans—rather than on Iraqis—as a positive sign. “If the attacks are against us, and not against Iraqi Security Forces or the people, then we’re winning,” said one. It was small consolation for those being shot at.

One day Kilcullen was riding with an Iraqi battalion commander who was about to move his unit into northwestern Baghdad for a 90-day tour. They were driving behind the outgoing Iraqi commander, whose Humvee blew up in front of them, turning the old commander into a shimmering cloud of hot pink mist. Kilcullen glanced over to look at the incoming Iraqi commander. “His eyes were like dinner plates,” he recalled. A few days later, a message from the insurgents arrived at the new commander’s headquarters, he said: “Sorry about last week. But you know, it doesn’t have to be that way. You’re only here for ninety days. Can’t we live and let live?” The battalion commander’s reaction, Kilcullen recalled, was “Sign me up.” After that, it became difficult to get that commander to do anything. For the next 90 days, his battalion was ineffective, and the sector effectively was in enemy hands.

Sadr’s Jaysh al-Mahdi was extraordinarily effective in infiltrating Iraqi forces, Kilcullen said. “We did a counterintelligence assessment of an Iraqi army battalion in central Baghdad and found that every senior commander and staff were either JAM, doing criminal activity with JAM, or intimidated by JAM.”

ANALYZING THE PATTERNS of Iraqi violence, Kilcullen concluded that al Qaeda attacked during the day, using car bombs to attack people around Shiite markets and mosques, while Shiite militias retaliated at night, sending death squads into neighborhoods where Sunnis slept. These different pathways of violence required different responses, he argued. The way to deter the al Qaeda attacks was to establish checkpoints at the entryways to markets, mosques, and other public places—and then to count it as a victory if a bomb exploded at a checkpoint and killed two Iraqi soldiers rather than detonating at its target and killing dozens of civilians. Likewise, the answer to Shiite revenge attacks was to protect a dozen of the remaining Sunni neighborhoods, creating “gated communities” surrounded by big cement walls. The new Joint Security Stations would emphasize helping the market checkpoints during the day and backing up the neighborhood checkpoints by night, as well as patrol through their areas.

As thousands of cement barriers were erected—the one separating Adhamiyah from a Shiite area was twelve feet high and three miles long—they were roundly criticized as an imitation of Israeli tactics. That was the most incendiary charge possible in the Middle East. Steve Niva, a Middle East specialist at the Evergreen State College in Washington, charged that they were “dividing neighbor from neighbor and choking off normal commerce and communications.” What they actually were doing was dividing Iraqis from people trying to kill them and choking off the normal movements of death squads. In Adhamiyah, civilian deaths declined by about two-thirds after the wall was erected in April 2007, Kilcullen said. One sign of the value of the walls was that al Qaeda in Iraq vigorously resisted them, noted Lt. Col. Dale Kuehl, commander of a battalion in northwest Baghdad. “We were engaged in a running battle with AQI as they tried to establish holes in the barriers while we tried to keep them intact,” he said.

Taking similar measures in al Anbar Province, the Marines found that the steps to limit the mobility of insurgents produced some unexpected side benefits. “The insurgency is like a shark,” a Marine intelligence report stated, “it has to move to survive. Cut off its freedom of movement and its loses its effectiveness.” As the fighters and death squads shifted to new locations, they were forced to communicate, and signals interception enabled the U.S. military to find them, or to eavesdrop on their reports and planning sessions. Trying to escape the new constraints, some insurgents moved out of the cities and into the desert. This in turn made it easier for the Marines to locate them and then order up air strikes. “Population control measures and the subsequent movement of the insurgency into more remote areas has a secondary positive effect on our operations,” the Marine report continued. “More and more often we found ourselves engaging the enemy on terrain that maximizes kinetic effects.” Also, in the emptiness of the desert, “collateral damage”—that is, killing bystanders—became far easier to avoid.

ONE OF THE SAD realizations brought by the new campaign was how disillusioned Iraqis had become with the Americans after five years. As Col. MacFarland had seen in Ramadi, the locals no longer had much faith in what American officers told them.

“The Defense of Jisr al-Doreaa,” an essay written by two Army captains, Michael Burganoyne and Albert Markwardt, and based on The Defense of Duffer’s Drift, the 1905 British military classic about small unit tactics in guerrilla war, vividly illustrates the education of Americans in Iraq—and shows why Iraqis were losing faith. One of the lessons that unfortunately appears more than once is about the failure of American officers to be able to fulfill the promises they make to local Iraqi leaders, or even to keep them alive against insurgent retaliation. “You Americans have been here for years now,” the mayor of a small Iraqi riverside town says in the essay to a newly arrived lieutenant. “It’s promise after promise. . . . Let us just eat so you will not have to lie to me with promises.”

Later, after the fictional lieutenant patiently wins the confidence of the mayor, who tells him where the local insurgents are based, the American unit is ordered to move elsewhere. “I met with the mayor and let him know we were leaving. His face seemed like it lost its color and he almost looked through me.” A few days later, the lieutenant is back on his air-conditioned Forward Operating Base, watching cable news over his breakfast of Lucky Charms cereal, only to see footage of the mayor being executed. “I saw the mayor and all the locals we have developed as informants, their hands and feet tied behind their backs, on the street in front of his house, with two masked men standing behind them. Everyone who had helped us defeat the insurgents was lined up.”

Thus, after getting into the neighborhoods, the new American units of the surge were taking over an operation that was in the red. Before they could do good they had to make up for the mistakes of their predecessors. They had to restore American credibility by delivering on their promises, and demonstrating that they wouldn’t make friends and then abandon them.

“We were wondering if our approach was going to work,” said Lt. Jacob Carlisle. “But when we got hit, we didn’t overreact.” He had studied Petraeus’s counterinsurgency manual and constantly sought to build bridges to local residents. For example, he said, “When we went into houses around the contact, we didn’t point weapons at them and yell and swear like we used to—‘You know what the fuck just happened! Tell me! You know who did this! Tell me!’ Now we went in and asked first if they were okay. Were there bad people who did this around that were threatening them? Why didn’t you tell us there were people digging in explosives in front of your house. . . . Call us next time.”

Carlisle, from Durand, Wisconsin, also found that years of frustrated American reconstruction programs had made Iraqis skeptical. A woman complained to him about raw sewage in the street, and he replied he would fix it. “All American make promises, but nothing ever happens,” she responded. Determined to show that times had changed, he made sure the problem was addressed. “Word gets out,” he said. “The people say, ‘This unit, they tell you something and it happens.’ ”

Even detainees were treated differently, the young infantry officer said. “When people are released, we bring them back to the family. We don’t just dump them out the gate.” During Ramadan, he gave money to widows and children, and to the family of a man he had detained. “All this stuff makes a difference.”

To deepen their awareness, his soldiers were assigned shifts in their neighborhoods. His platoon patrolled during the morning, and the company’s other two platoons the afternoon and evening. “We know what is normal on the streets, and see the same people in the same places every day. We know if something is out of place.” As the days passed, familiarity led to more ease in communication, and even a smidgen of trust. “Now that we were in the neighborhood every day, they believed us that we would keep them safe. More and more started calling us.”

There were three steps of cooperation, said Lt. Col. Stephen Michael, commander of a battalion of the 25th Infantry Division, also posted on the south side of Baghdad. “First people weren’t working with us, then they would work with us covertly, and now most work with us openly.”

It took time—sometimes two or three months—before the Americans and the Iraqis began to grow accustomed to each other. “When we first came over and started planting ourselves in, you wouldn’t see too many people because they didn’t know if we’d be here, they didn’t know to trust us, and basically the extremists were still intimidating and the people were reconning us,” said Col. Wayne Grigsby of the 3rd Infantry Division, commander of the third of the surge brigades, which deployed to the tough area to the southeast of Baghdad. But after about two months, in late spring, people began talking to the American soldiers. Iraqis would begin telling them things, he said, like “Hey, that guy over there has never been in this town before. He drove in with two big trucks,” their cargoes covered with tarps. “I don’t think it’s right, and we don’t want him in here if he’s going to bring trouble. Can you go take care of that problem?”

Down in south Baghdad, Lt. Col. Crider found the same effect. “The days of large cordon-and-sweep operations and hoping to find something . . . were over,” he said. Instead, he sent his soldiers into Iraqi homes to learn who lived in the neighborhood to converse, drink tea, take photographs and census data, and learn about local concerns. “The American soldiers was no longer a mysterious authority figure speeding by in a HMMWV behind two-inch glass who occasionally rifled through their home. . . . After repeated encounters, our soldiers began to learn who was related, which families did not get along, who provided useful insight, and many other intimate details.” They found that in their neighborhood lived an international basketball referee who had worked on the side for Iraqi intelligence. They met a famous Iraqi comedian, as well as a cardiologist fluent in English and eager to help. As they began to know and see more, attacks on them and on Iraqi civilians began to taper off. “AQI could no longer threaten individuals with violence after we left, because we never did,” he observed. Also, locals began to report the emplacement of roadside bombs, which forced insurgents to switch to grenades and automatic weapons, which were riskier to use.

After a detainee was released—legitimately—into the neighborhood, Crider was pleased to receive 11 tips from local citizens about his presence. U.S. soldiers were sent to visit him and talk to him “about how things had changed. . . . He never caused any problems.”

In keeping with the new, more neutral stance of the U.S. military as the arbiter of events, rather than an ally of one side, Crider also reined in the National Police, which at times was indistinguishable from a Shiite militia. “Denying the National Police the ability to unilaterally operate in the neighborhoods greatly increased our credibility,” he said. Commanders also learned to keep a wary eye on those allies, especially as they tried to capitalize on U.S. operations for their own ends. “Once we cleared AQI from an area, Shia extremists would try to follow and claim it as their own, essentially replacing a cleared area with a new threat,” stated an after-action review conducted by Odierno’s headquarters.

The fight was growing more complex. One day in May, Kilcullen noted that, in Baghdad’s Hurriyah neighborhood, there were four factions of Jaysh al-Mahdi, Sadr’s extremist Shiite militia, fighting each other—Noble JAM, Golden JAM, “criminal JAM,” and “ordinary JAM.” U.S. officials sent a message to “JAM Central” in Najaf. “We want these guys out of there.” In response, he said, the JAM headquarters in Najaf sent a hit team to Baghdad to sort out the problem. “Because we treated them as the authority, they cleaned it up.” There also was murky unconfirmed talk that a deal was reached under which the U.S. military would aid Golden JAM in attacking other parts of the militia deemed to have gone rogue. Petraeus stated flatly that no such agreement existed and suggested that it grew out of rumors collected from Iraqis by U.S. intelligence or deals made by local American commanders.

The trends in Shiite southern Iraq also were worrisome. “The British have basically been defeated in the south,” said a senior U.S. intelligence official in Baghdad. They were abandoning their former headquarters at Basra Palace, where an official visitor from London had described them as “surrounded like cowboys and Indians” by militia fighters. An airport base outside the city, where a regional U.S. embassy office and Britain’s remaining 5,500 troops were barricaded behind building-high sandbags, was being hit by rockets or mortars an average of 150 times a month. Was Basra this year a foretaste of Baghdad the next?

ON JULY 4, 2007, Lt. James Freeze, leader of a 2nd Infantry Division reconnaissance platoon based north of Baghdad, celebrated Independence Day by having a glass of sparkling cider and a cigar with his old friend Austin Wilson, another lieutenant and West Pointer who had been the best man at his wedding. They discussed what one word would best characterize the Iraq they knew. They settled on “hopeless.”

By coincidence, Fred Kagan, in many ways the guiding spirit behind the surge, was in the tough south Baghdad neighborhood of Doura a few weeks earlier, visiting one of his former West Point cadets who was now a company commander. “It was a complete combat zone,” he said. “There was no one in the streets. It was a ghost town.” The American brigade commander declined to take him out on patrol because of the danger.

Generals tend to be optimistic by nature, said Gen. Fastabend, Petraeus’s strategic adviser. “The pessimists quit as captains,” he cracked.

But five months into the new strategy, even some of the optimists were feeling gloomy. The Army’s new counterinsurgency strategy required soldiers to be among the people, where they would form new relationships—but it also exposed them to hellacious new levels of violence. “We had some extreme challenges, in May, June, July,” recalled Brig. Gen. Anderson, Odierno’s chief of staff. “We were hedging our bets that the surge would work.” When Iraqi forces were sent into a cleared area to help out, he said, the chances were “fifty-fifty” they were up to the job.

“It kept getting worse,” Rapp recalled. “May had very high casualties. I thought, ‘Holy cow, what is going on here?”’ There was good reason to fret: The possibility was growing that the situation was about to get much worse, with the Americans played out and all the ingredients of a massive civil war coming together—there was oil to fight over, plenty of weapons available, and plenty of Iraqis as well as people in neighboring states who possessed the experience and skills to intensify the fighting.

It wasn’t just Baghdad, either. In May, Gen. Odierno and Emma Sky helicoptered to Baqubah, about 35 miles northwest of the capital, a city both knew from their previous tours. “I knew it wasn’t right,” Odierno said. “It had a black cloud.” As the surge had pushed some fighters out of the capital, they had moved into Baqubah and other parts of Diyala Province.

“We were gobsmacked,” added Sky, using British slang for being stunned into speechlessness.

It was tough having to face the soldiers bearing the brunt of the new strategy. “There was a brief moment of What have we got ourselves into?” recalled Command Sgt. Maj. Hill, the veteran infantryman who had been selected by Petraeus to become the senior enlisted soldier in Iraq. Looking at the casualty reports every night that spring, he said, “would just suck the energy out of you.” His days began to seem like a soul-lashing round of visiting the wounded and then attending memorial services for the dead. He learned to say a prayer under his breath before walking into the military hospitals: “God, give me strength to deal with what I’m about to see.” He kept his calendar open every day from 5 to 6 P.M., on the assumption that at least one service for a dead soldier would be held.

As the casualties continued to mount, Odierno said later, “I was a little nervous.” Col. J. T. Thomson, the career artilleryman who was Odierno’s executive officer, would later recall those dark days as the hardest part of his tour. “May—I mean, the whole month of May,” he said much later. “The wondering—is it going to get any better?”

According to unreleased statistics in the U.S. military database, there were 6,037 “significant acts” of violence in Iraq during May 2007, the highest recorded total since November 2004. “This is a period in which it gets harder before it gets easier,” Petraeus said one day in May as he sipped iced tea in his office, a giant map of the city of Baghdad behind him. He was expecting a long, hard summer of violence, followed by a trek to Capitol Hill to tell Congress how much progress he was making. He was pushing all the American chips on the table, going “all-in,” he said, with the surge. Whatever happened, he was going to ride this thing through to the end. “There’s no combat forces left, at least, I’m aware of,” he said. That is, the United States military simply didn’t have replacement troops available for those he was fielding. “You can’t ask for a brigade that isn’t there.”

Petraeus later would describe this period as “excruciating.” He said he believed that the new approach would work, but “what started to develop as the question in my mind was, when will it start to show demonstrable effects?”

United States’ combat deaths climbed inexorably: 70 in February, 71 in March, 96 in April, and 120 in May, which became the deadliest month for U.S. troops in two years. The additional casualties had been expected as the price to be paid in the short term for moving from big, safe bases to smaller outposts among the population. But they came even as a series of horrific killings of Iraqi civilians occurred. In February, a ton of explosives detonated in a market in a predominantly Shiite area of Baghdad, killing at least 125 and wounding 300 more. It was the single deadliest terrorist bombing ever in the capital. “They were carrying bodies like sheep,” said one Iraqi witness, Abu Lubna.

The insurgents also were introducing worrisome new tactics. In February and March, they forayed into chemical warfare, detonating three trucks carrying toxic chlorine gas in Baghdad, Fallujah, and Ramadi, killing 11 people and sickening hundreds. Col. MacFarland may have found the tipping point in Ramadi the previous year, but there was plenty of fighting left in the city, as his successor unit, the 1st Brigade of the 3rd Infantry Division, led by Col. John Charlton, found in a series of battles in February and March 2007, and then again in June, when a U.S. patrol stumbled across an al Qaeda counterattack as it was forming, resulting in an all-night firefight that was called “the battle of Donkey Island.” It left two Americans dead and more than 30 insurgents, and blunted what likely was a new al Qaeda offensive.

Another horrific new approach appeared in Baghdad. The driver of a car bomb managed to get through a U.S. military checkpoint and into a marketplace because he had two children in the backseat of the vehicle. Troops had been taught that cars with children were no threat. Three Iraqis were killed in the subsequent blast.

Enemy tactics were also more sophisticated, with false IEDs being strewn along with real ones, the better to slow down American troops and set them up for ambushes. “These guys are real smart,” said 1st Lt. Anthony Von Plinsky. “The Iraqi insurgent as a whole has adapted well to our tactics.” By this point in the war, soldiers were fond of saying, all the stupid insurgents were dead. The Americans had come and gone on tours of duty, but many of their enemies had fought nonstop for several years, and those who had survived were fit and adaptive.

The biggest threat to the success of the U.S. mission was the al Qaeda car bomb attacks against Iraqi civilians, which made it seem to many Iraqis as if the Americans couldn’t provide security and that the militias was the only hope. But the biggest threat to the soldiers carrying out that mission was the roadside bombs, especially the highly lethal explosively formed penetrators, or “EFP”s. Also, an increasing number of convoys were being attacked, and American officials worried that enemy fighters were receiving Iranian training in the new tactics used in those attacks.

There seemed no limit to the forms of violence. American troops operating a new outpost in Diyala Province befriended a donkey that hung around, giving it food and water. Then “the insurgents assassinated him,” said Spec. Josiah Hollopeter. “That really irritated me.”

THE BATTLE OF TARMIYAH

As the new American outposts proliferated, they did appear to draw some of al Qaeda’s firepower away from civilians. The more remote stations were especially enticing. For example, according to Col. David Sutherland, as sectarian killings and kidnappings declined in the late winter and spring of 2007 by about 70 percent in Diyala Province, northeast of Baghdad, attacks on U.S. and Iraqi troops increased by the same amount.

One of the most spectacular attacks was launched against 38 soldiers manning an isolated American outpost in the town of Tarmiyah, just north of Baghdad. The town of about 40,000 actually had been relatively calm until the summer of 2006, when it was destabilized by ethnic cleansing in the capital that sent thousands of Sunnis fleeing there. Al Qaeda’s power in the town grew, and in December it ordered the Iraqi police there to leave—which they promptly did. The 1st Cavalry Division then established an outpost in the abandoned police station. In mid-February it was being manned by members of D Company, 2nd Battalion, 8th Cavalry Regiment, 1st Cavalry Division. It was the northernmost position in the division, poking into an area that had been a relative safe haven for Sunni insurgents.

At precisely 7 A.M. a rocket-propelled grenade detonated on the corner of the small outpost, followed by some AK-47 fire. Lt. Shawn Jokinen, who had gone to sleep two hours earlier, jolted awake in his cot. Staff Sgt. Jesus Colon, the sergeant of the guard, shouted that they were under attack. Jokinen ran to the front door of his barracks with his M-4 carbine and saw a small white “bongo” truck crash through the sliding blue front gate and roar straight toward him. He emptied the M-4’s magazine into the windshield, causing the truck to swerve slightly away from the entrance, but before the driver died he detonated about 1,500 pounds of Ukrainian-made military-grade explosives, sending bits of concrete and glass sailing through the compound. “The explosion threw me against a wall and I got covered with debris,” Jokinen remembered. The blast dug a crater twenty feet wide and six feet deep, shattered every window in the compound and the surrounding area, and dropped the front wall of the compound.

The battle that followed resembled the movie Zulu, in which a small detachment of British soldiers fends off thousands of African warriors. At first the dust was so thick that no one could see or breathe. “Everything was black, then brown,” said Staff Sgt. James Copeland. He took a knee until he could get some air. Several soldiers were covered in rubble. Those not covered pulled their buddies out, then grabbed their weapons, helmets, and body armor, and ran upstairs to the roof. Some would fight for hours in their boxer shorts. Two medics began treating those with life-threatening wounds. “The rest of us wrapped up each other,” Jokinen said. Copeland told the injured they were needed to shoot if they could, then grabbed a wounded soldier’s M-249 light machine gun and ran to the roof, where he realized that his gear was buried and that he didn’t have a helmet.

Lt. Cory Wallace, D Company’s executive officer, had been walking out of the compound’s command post, where he had been processing six members of an alleged al Qaeda sniper cell nabbed in an overnight raid, when the blast hurled him into a wall. The compound’s 500-gallon fuel tank exploded into a fireball, knocking him out and killing Sgt. Colon. He regained consciousness and scurried back into the command post, where he saw Pfc. Pao Vang trying to stop blood squirting from a laceration on his neck. Wallace looked outside. “I noticed the front half of our barracks were destroyed. Several soldiers were staggering out of their patrol base. They were covered with dust and blood. I was still a little dazed from the blast so it took me awhile to notice that the enemy was throwing hand grenades and improvised mortar rounds over the walls.” Black smoke joined the dust and grenades in the air.

Wallace shouted to Pfc. James Byington, who had picked himself up from the ground, to call the battalion headquarters for help. “Byington informed me that the radios were not working,” the XO recalled. It turned out that the compound’s generator had been knocked out. Wallace told Vang, who had a shard of glass protruding from the side of his head, to fix the generator. Vang tried to do it while under direct fire from a nearby building but couldn’t, so said he would find batteries for the radio. They were buried under some rubble, so Vang dug with one hand to find them, the other pressed against his neck wound, which was spraying blood every time he moved.

At some point—Wallace remembers it was 90 minutes, but battalion records say far sooner—Wallace was able to transmit a situation report to his headquarters. “Once the radios were functional, I called battalion and informed them that our patrol base was under heavy attack and that our company had multiple wounded with one KIA,” he said.

One of the soldiers on the roof yelled down to him, “Sir, don’t let battalion pull us out, we’re going to hold this motherfucker!” There was no fear of running out of ammunition, because the platoon sergeants wisely had insisted that the unit keep on hand about three times the daily requirement. On top of that, soldiers had themselves prepared “Armageddon Boxes”—extra ammunition and some water for unexpected emergencies—and kept them in their Humvees. “The only problem was our ammo holding area was located on the second floor of the barracks,” Wallace said. “Soldiers kept sprinting down an exposed staircase, filling up sleeping bags with extra ammo, and running back up to their fighting position on the roof.” Copeland ran from soldier to soldier on the roof, distributing ammunition and assigning sectors of fire.

The radio was in the command post, so Wallace couldn’t see outside to guide the AH-64 Apache attack helicopters appearing overhead toward their targets. He had Vang assemble a portable Harris PRC-117 radio and take it to the roof, where Staff Sgt. Freddie Housey, a veteran of the capture of Baghdad in 2003, directed the air counterattack. One of the Apaches was hit and pulled away with one of its pilots wounded and his flight suit on fire. Other helicopters conducted devastating strafing runs with their 30-millimeter cannons.

Lt. Col. Scott Efflandt, the battalion commander, had been eating breakfast 12 miles away at his headquarters at Taji when he felt the concussion of the explosion and then, moments later, heard the boom. He checked with his tactical operations center, or TOC, but was told there was nothing to report, so assumed it was artillery fire involving another unit. He didn’t know then that soldiers from his D Company were fighting for their lives.

Wallace reported in a few minutes later. Efflandt raced to his TOC. As he arrived, he recalled, “the streaming video from the UAV [drone reconnaissance aircraft] came online and our hearts skipped a beat.” He called Wallace, found him “in charge and unflappable,” and told him help was on the way.

At around 8 A.M., a unit of Stryker armored vehicles from the 2nd Infantry Division came to the rescue. One of the Strykers backed up to a hole the blast had made in the compound wall, dropped its ramp, and loaded the six most severely wounded D Company soldiers. Another unit arrived and secured a landing zone, or LZ, for medical evacuation helicopters. Wallace realized his compound wouldn’t be overrun, and he would survive the day. “With our litter-urgent soldiers medevac’d and armored vehicles occupying a perimeter around the patrol base, I knew the enemy was beaten,” he said.

But the battle wasn’t over. “As we headed to the LZ I still heard small-arms fire, friendly and enemy,” said Copeland. “The LZ was hot with the Stryker and air assets still firing as we were moved to the bird and continued as we flew away.”

Efflandt, a working-class son of Rock Island, Illinois, who had gone on to teach at West Point, got to Tarmiyah later in the morning. “When we entered the town, I was stunned. It was as if we were in the wrong place, as everything looked different—battle-damaged buildings, debris everywhere downtown, no people out and about. Arriving at the patrol base I was aghast.”

The outpost was destroyed. It may have been defendable but it was uninhabitable. Efflandt decided to stay and fight it out, requesting immediate delivery of a big logistics package, including thousands of tons of concrete barriers. He issued orders to take over a school building 200 meters north of the destroyed outpost and get a new patrol base up and running by sundown. “It sent a message to the insurgents that we would not be defeated and we weren’t going anywhere,” recalled Maj. Robert Rodriguez, the battalion’s executive officer. “It was a tactical decision with strategic implications.” Leading from the front, Efflandt spent the next 24 hours in the new post commanding the operation to retake the town. He had in mind Odierno’s dictum that any land taken would not be given up.

When the fight was over, of the 38 soldiers who had been in the outpost, 2 were dead and 29 others had been wounded. Those who weren’t hospitalized moved back to their battalion headquarters at the big base at Taji, a few miles to the southwest. The next morning Wallace woke up and went to eat. “I didn’t realize what had happened until I walked out of the chow hall,” he said. “For some reason, that was the best breakfast I had ever tasted in my life.”

But other feelings from the fight lingered. Looking back on it now from the United States, Wallace said, “I feel guilty. I keep thinking there were a hundred things I could have done to prevent it.” He is scheduled to return to Iraq in November of 2009.

When Efflandt left the battalion in 2008, his officers memorialized his command tour in Iraq with a print of Gen. Meade at Gettysburg titled Stand and Fight It Out. Sporadic fighting would continue in Tarmiyah through that year, at one point leading to a friendly fire shootout between American soldiers and Iraqi soldiers and police, killing 6 of the Iraqis.

BLACK THURSDAY

As the surge intensified, with the majority of the additional brigades in country, the situation actually worsened. Thursday, April 12, stands as perhaps the toughest day of this period. The previous day, news had broken in Washington that three retired generals had turned town the job of coordinating Iraq policy for the White House. It was a stunning vote of lack of confidence in the new strategy in Iraq. One of those who refused the job, retired Marine Gen. John Sheehan, explained his decision by saying, “The very fundamental issue is, they don’t know where the hell they’re going. So rather than go over there, develop an ulcer and eventually leave, I said, ‘No, thanks.’ ”

On the same day, Defense Secretary Gates announced that all soldiers in Iraq, as well as those on their way, would serve 15 months there, rather than the one-year tour that had been the norm. Soldiers now had to tell their families to revise those homecoming plans, many of which involved long-planned trips to see family members or vacations at resorts. As the news spread among troops in Iraq, their reaction was expectable. “It flat out sucks, that’s the only way I can think to describe it,” said Pvt. Jeremy Perkins, a member of an engineering battalion in Baqubah.

On the morning of April 12 itself, a truck bomb dropped part of a key Baghdad river crossing, the Sarafiya bridge, dumping cars into the Tigris and killing 11 people. This appeared to be the first step in a campaign to prevent Shiite death squads from crossing the river into west Baghdad, or perhaps to limit the mobility of U.S. and Iraqi reinforcements. Several other bombs would hit major bridges in the following weeks. It was one more way to pull apart the carcass of a once-great city.

That afternoon, a bomber managed to get past multiple checkpoints, bomb-sniffing dogs, and body searches, and into the Green Zone building where parliament was meeting, killing a member and seven other people.

Back in Washington on the same day, Senator Joseph Biden, the Democratic chairman of the Senate Foreign Relations Committee, pronounced the surge doomed. The next day, Friday the 13th, Senate Majority Leader Harry Reid said, “I believe myself . . . that this war is lost. . . . The surge is not accomplishing anything.” Even some supporters of the war were beginning to talk about what “Plan C” might look like. Would it be, one hawk asked, a fallback to the core missions of attacking al Qaeda, protecting the embassy and providing air cover and other support to Iraqi forces?

The bad news seemed relentless. On April 14, a car bombing at the entrance to the main bus station in the Shiite holy city of Karbala killed 32. Four days later, bombings in mainly Shiite areas of Baghdad killed more than 150.

The assaults against new outposts continued. On April 24, a U.S. patrol base in an old schoolhouse in Sadah, near Baqubah, came under complex attack, with small-arms fire and rocket-propelled grenades from several directions. As the soldiers on the roof of the base returned fire, they saw two explosives-laden dump trucks coming at them. The drivers couldn’t be shot because they were cocooned in steel, with only a slit to see through. The first, carrying 1,500 pounds of explosives, blew up outside the gate, leveling the obstacles leading up to it. The second one barreled through the breach just made and detonated 2,000 pounds, collapsing a building. All told, 9 U.S. soldiers were killed, all of them from the 82nd Airborne Division; 20 more were wounded. “It was the worst day of my life, to have to literally dig with your hands and carry your kids out,” recalled Col. Sutherland, commander of the 1st Cavalry brigade to which the 82nd Airborne unit was attached. “That was extremely hard.” The Islamic State of Iraq, a group affiliated with al Qaeda, boasted in a subsequent statement that it had sent “two knights” to attack “the Crusader American base.”

One day in April, a senior non-commissioned officer in the 1st Battalion, 2nd Brigade, 1st Infantry Division allegedly led some soldiers in the execution of four Iraqi detainees. According to preliminary testimony by other soldiers, 1st Sgt. John Hatley, the top sergeant in the battalion’s A Company, had four blindfolded and handcuffed Iraqis kneel by a canal. They had been captured after what Stars & Stripes, the official U.S. military newspaper, termed “a brief exchange of fire” and a search that turned up “heavy weapons,” which in Iraq usually means mortars or rocket-propelled grenade launchers. Hatley told his men that if they passed along the Iraqis to a detention facility, they simply would be released, testified Pfc. Joshua Hartson. The Iraqis then were shot and their blindfolds and handcuffs were removed. “We then pushed the bodies into the canal and left,” Sgt. Michael Leahy wrote in a statement given to Army investigators. Back at Combat Outpost Angry Dragon, Hatley gathered his troops and ordered them not to discuss the incident. He also told some soldiers to burn the blindfolds and the handcuffs, which were plastic, and to clean out the Bradley Fighting Vehicle in which the detainees had been moved. The incident only came to light in January 2008. Criminal proceedings began months later. Hatley, Leahy, and a third soldier eventually were charged with committing premeditated murder. At the time of publication of this book, they had not gone to trial, while two other soldiers pleaded guilty to lesser charges.

At 4:40 on the morning of May 12, insurgents ambushed an American unit in the “Triangle of Death” area southwest of Baghdad, first bombing it and then raking the survivors with gunfire. Five soldiers died and another three were abducted, with two of the bodies discovered finally a year later. Nine more soldiers were killed on May 23. Another 10 died on the 28th, which was Memorial Day, most of them in an incident in which an OH-58 Kiowa Warrior helicopter was shot down north of Baqubah and the mission sent to respond to the crash was hit by big two roadside bombs.

Senator Gordon Smith, the Oregon Republican who had come out against the war so vigorously the previous December, traveled to Iraq in May. He believed his emotional speech on the floor of the Senate had helped push Bush toward the surge. A White House aide, he recalled, had told him, “We recognized with your speech that not only were we losing the war, we were losing the Republicans we needed.” But after touring the country and talking to Petraeus, he was no more optimistic. After he and Senator Orrin Hatch, the Utah Republican, left Petraeus’s office, they were strapping into their seats in a Black Hawk helicopter for the short flight back to the Baghdad airport. “So what do you think?” Smith said to Hatch.

“We could lose this thing,” Hatch glumly replied.

On June 25, a wave of bombings hit Iraqi allies of the U.S. effort. Two car bombs targeted the police station in the refinery town of Bayji, killing 30. Another attack killed 8 policemen in Hilla. But the most politically significant incident of the day was a suicide attack on a group of Anbar Awakening sheikhs meeting at the Mansour Hotel, just a short walk from the northern entrance to the Green Zone. Six of the tribal leaders were killed, as well as 6 other people.

The last of the surge brigades and their support troops finished arriving in June, elevating the U.S. troop level in Iraq to 156,000—plus another 180,000 contractors performing functions that once were done by soldiers. (Most of these were cleaners, cooks, and so on, but about 20,000 were private security guards.) By July it was beginning to look to many that the surge was failing, adding to pressure to move to a withdrawal plan. The most precious commodity Petraeus and Odierno had was time. “Everything takes time,” noted Maj. Gen. Rick Lynch, who was commanding the 3rd Infantry Division in the belt south of Baghdad. “And everything takes longer than you think it’s going to take.”

A growing chorus of voices was saying they had run out of time. Retired Gen. Sir Michael Rose, one of the most prominent British officers of recent years, called on the Americans to “admit defeat” and bring the troops home. Senator Smith predicted that “a dozen Republican senators . . . will be with me in September.” And a poll found that nearly 60 percent of Americans thought the surge would not help restore civil order to Baghdad. Tom Donnelly remained a strong supporter of the surge that he had helped design, but conceded “it’s the eleventh hour and the fifty-fourth minute.”

Political support for the surge, never strong, appeared to be collapsing. Senator Reid, who in April had pronounced the war lost, now attacked Petraeus personally, charging, somewhat oddly, that the general “isn’t in touch with what’s going on in Baghdad”—as if he could discern better from Washington, D.C. Senior Republicans weren’t far behind him in heading for the exits. Senator Richard Lugar, the centrist Indiana Republican, took to the floor of the Senate on June 25 to call for an end to the surge. “I believe that the costs and risks of continuing down the current path outweigh the potential benefits that might be achieved by doing so,” said Lugar, one of the most respected voices in Congress on foreign policy. “Persisting with the surge strategy will delay policy adjustments that have a better chance of protecting our interests over the long term.” A week later he would be joined by Pete Domenici of New Mexico, who called for following the Iraq Study Group’s recommendations and getting U.S. combat forces out of Iraq by early 2008. Senator George Voinovich of Ohio also was backing away from the president.

“The war in Iraq is approaching a kind of self-imposed climax,” warned Henry Kissinger.

Al Qaeda was chortling. “Today, the wind—by grace of Allah—is blowing against Washington,” Ayman al-Zawahiri, the terrorist organization’s second in command, said in a video posted on a jihadist website.

DEAD MAN WITH AN IPOD

The morale of American troops seemed to be waning as they doubted if their new mission was working. “We’re tired of being lost,” said Sgt. 1st Class Michael Eaglin, who was operating from a small base in Sadr City. “Have you ever been lost and at the same time getting shot at? It’s miserable. . . . I want to be here for a reason, not just a show of force.”

In Yusifyah, a tough little town near the southern edge of Baghdad, Spec. Yvenson Tertulien told the Los Angeles Times that “I don’t see any progress. Just us getting killed. . . . I don’t want to be here anymore.”

Lt. Gregory Weber, an infantry platoon leader in the 2nd Infantry Division, recalled responding to a bombing and RPG ambush of a U.S. patrol in southern Baghdad that summer:

We passed the top half of a HMMWV [Humvee] turret. 1st Squad was so focused on security and assaulting/clearing up to the blast site that they didn’t even see [in the turret] the KIA [killed in action] Soldier, covered in soot, ACH [helmet] blown off, IBA [body armor] barely on, but an iPod headphone still in his ear. On site, there were three HMMWV destroyed. One upside down from an 8 foot deep, 15-foot-wide blast crater, 25 meters away, burning with the remains of 4 soldiers left inside. Another HMMWV was in the blast crater, partially submerged in water from a water main rupturing, and the other HMMWV 25 meters the opposite direction with its back end blown off. It was the most horrific subsurface IED detonation I saw the entire deployment.

Five soldiers were killed in the incident, but the image that haunted Weber was the first thing he saw, the dead soldier in the blasted turret, “iPod still in his ear.” He still wonders, “Did his leadership know he was distracted by music; not being able to hear the battlefield?”

Indeed, there were growing signs of such demoralization and indiscipline. In the hard-hit 1st Battalion of the 26th Infantry Regiment, which had lost five soldiers in one bombing in June, life got even worse in July. The first sergeant of its Alpha Company, while on patrol, said, “I can’t take it anymore,” put a weapon under his chin, and shot himself in front of his men. A few days later, members of a platoon in the battalion’s Charlie Company refused to go out on a mission, saying they were afraid of becoming abusive with Iraqis.

In another unusual act that verged on insubordination, seven 82nd Airborne soldiers placed an opinion piece in the New York Times that called the surge a failure. “We see that a vast majority of Iraqis feel increasingly insecure and view us an occupation force that has failed to produce normalcy after four years and is increasingly unlikely to do so as we continue to arm each warring side.” Legally they were entitled to express their opinions, but for soldiers to write a newspaper piece on policy during a war is almost unprecedented. Three weeks later, two of the writers, Sgt. Yance Gray and Sgt. Omar Mora, would be killed after their truck veered off an elevated highway in western Baghdad and dropped about 30 feet.

The governor of Puerto Rico, Anibal Acevedo Vila, addressing the National Guard Association’s annual conference, called for a new strategy in Iraq that would lead to a withdrawal. He received a standing ovation.

“I have never seen in twenty years the sort of resigned attitude I am hearing from my active-duty counterparts,” reported one Army Reserve colonel. “They are conveying a ‘game over’ attitude where they are going to continue saluting the flag and doing what the NCA [national command authority] wants, but not without realizing it is all horseshit at this point.” After the American military left Iraq, he added gloomily, the Iraqis will “turn on each other like a pack of weasels.”

A SLOW TURNING

In retrospect, it appears that the pattern of the battle of Baghdad from March to June resembles, on a vastly larger scale, that of the assault earlier in the year on the Tarmiyah outpost. In both places, the new U.S. strategy was pushing into enemy strongholds and eliminating safe havens. The enemy reaction was to hit back as hard as it could. Indeed, the U.S. counteroffensive could be said to have triggered some of the bombings, as the enemy faced a “use it or lose it” prospect with its arsenal of prepared car bombs and stashed explosives. “They have previously been, you know, frankly, elusive when we actually got into an area and started to clear it, and we’re seeing that in this area of east Rashid, they are standing and fighting,” Maj. Gen. Joseph Fil Jr., commander of the 1st Cavalry Division, noted in June. Both sides were throwing everything they had into the fight.

Visiting Iraq at the beginning of April, Senator John McCain expressed “cautious, very cautious, optimism” about the effects of the new strategy. “We’ve made tremendous mistakes,” he said in Baghdad on April Fool’s Day, “but we’re finally getting it right. And is it too little, too late? I don’t know, but I don’t think so.”

His traveling companion, Senator Lindsey Graham, added another thought: “We’re doing now what we should have done three years ago.”

The two were mocked for citing a walk through a Baghdad market as evidence of improved security, but in fact they were right. There may have been soldiers protecting the market, but the market was there, with merchants and goods, because of that military presence.

As Kissinger had said, the war was approaching a climax—but not of the sort he envisioned. Quietly, in various corners of Baghdad and its environs, even as the high-profile bombings were escalating, the new strategy was beginning to show results in hundreds of ways. Every day, American troops found that more Iraqis were beginning to talk to them. Better intelligence was coming in, and was being acted on more quickly, by units that lived on the next block instead of on the outskirts of the city. A unit getting a tip on a house where enemy fighters were gathering would begin watching it, not necessarily to hit it immediately, but perhaps to see how it fit into a larger network. With that knowledge, it might then be able to cripple a gang that often had been intimidating and extorting area residents. Maj. James Allen learned this lesson in an odd way as the Iraqi troops he was advising ambushed an insurgent planting a roadside bomb. They aimed to kill the would-be bomber, but their weapons were so poorly maintained that they couldn’t fire. “The dude who was emplacing the IED froze, though, so they walked over and bagged him,” Allen recalled. “He rolled over on the supplier, the supplier rolled on someone else, and we essentially shut down IEDs on that stretch of road for eight weeks.”

Also, having American troops in residence often dramatically improved the effectiveness of their Iraqi counterparts. Having Americans available to come to their aid—and perhaps to feed and outfit them—made Iraqi soldiers more comfortable about being out in the neighborhoods. “They feel as long as the Americans are there, they can pretty much handle anything that’s going on,” said Sgt. Maj. Michael Clemens, who served with the 82nd Airborne in Diyala Province from mid-2006 to mid-2007. Many of these new “partnered relationships” would begin to show results by midsummer. Of course, the locals also generally found it easier to talk to the Iraqi troops, who often would pass along the information they gleaned to the Americans with whom they shared a post.

Familiarity bred knowledge. One squad of American troops living in a Sunni area began to examine what was being sold in the markets as an indicator of the mood of the population. For example, it noticed one day that heavy portable heaters were being offered in their local market, which they interpreted—correctly—to mean that people were planning on staying there, which in turn meant that the pressure on the population to move brought by Shiite militias must be declining.

Even the language that American leaders used was changing. “There’s a lot less cowboy lingo in the force—‘toss the compound,’ ‘take ’em down,’ ‘roll ’em up,’ ‘get the bad guys,”’ observed Lt. Col. Yingling, on his third tour in Iraq. Col. Grigsby, commander of the third surge brigade, still introduced himself like a traditional armor officer as “Hammer Six,” but his orientation was different. “The quality of life in Jisr Diuala, one nahiya in the qadha,” was improving, he told reporters one day. He also was proud that “we worked out of eight patrol bases and four joint security sites in the middle of the population centers, [and so] we never commuted to work.”

The improvements in American operations were technical as well as doctrinal, tactical, and cultural. One of the reasons that redeploying the troops into small outposts could work in 2007 better than it would have in previous years was that brigade commanders had far more aerial surveillance assets available and under their control. During 2007 the number of these drone reconnaissance aircraft operating in Iraq would increase tenfold, according to an after-action review by Odierno’s headquarters. During his first tour in Iraq, in 2003-4, Odierno noted, the most that could be counted on was two drone reconnaissance aircraft available to him in all of Iraq, and they had to be shared with other divisions. In 2007 all 18 Army combat brigade commanders had their own RQ-7 Shadow UAVs, and could request more surveillance and strike aircraft as needed. This made it far easier, for example, for a commander to keep an eye on potential threats to his outposts.

In addition, in a highly classified operation, new information about al Qaeda and insurgent leaders began to get distributed much more quickly to tactical units. The officer responsible for the change was a military intelligence specialist, Lt. Col. Jen Koch Easterly, who reorganized the collection and analysis of intercepted telephone and computer communications in order to coordinate it better with other intelligence operations and with what units were doing on the ground. She also focused more on going after the networks that were assembling, delivering, and detonating roadside bombs, which has been the single greatest killer of U.S. troops during the war. According to one senior officer, her military intelligence unit’s successes became the undisclosed key to the success of the surge. Her work still remains largely unknown because so much of what was done remains highly classified. But as one operations report by the 1st Cavalry Division put it, “synchronization of ISR/HUMINT/SIGINT [intelligence, surveillance, and reconnaissance /human intelligence/signals intelligence] has significantly reduced IED cells and threat.” Asked about that, Maj. Patrick Michaelis said, “It was a major factor. . . . Cryptological support from Colonel Easterly was critical.” She declined to be interviewed, citing classification issues.

In midsummer, intel people picked up some interesting indications that the insurgency was running out of steam. One smart U.S. Army intelligence officer in Baghdad said that he just didn’t see the signs of a vibrant counterattack forming. “There’s nothing that shows any kind of [enemy] surge in the making,” he said. On intercepts of telephone conversations between insurgent leaders, he noted, “There’s a lot of bitching and moaning, ‘What have you done today?”’ The response, he said, was often along the lines of, “I haven’t done anything, there are too many around, I can’t move.”

One of the emerging lessons was that the increase in regular U.S. troops on the streets of the city improved the effectiveness of the Special Operators who were targeting al Qaeda. It also helped that Odierno was an old friend of Lt. Gen. Stanley McChrystal, the head of Special Operations in Iraq, whom he had known since their days together at West Point. Until that point, “We didn’t see how essential conventional forces are in the counterterror fight,” remembered Rapp. American commanders, he said, were surprised to see that having their troops moving around effectively sponged up the sea in which al Qaeda swam. As insurgents found it more difficult to move, they began to communicate more electronically, in part because as senior leaders were caught, they tended to be replaced by younger, less experienced men, which in turn made them more vulnerable to Lt. Col. Easterly’s signals interception operation. “When you stay in the neighborhood, they have no place to stay, they have to talk more, because they’re mobile, so we can catch them, ” Rapp said.

Even as U.S. troop deaths increased, Iraqi civilian deaths appeared to be declining, decreasing steadily from January on. Essentially, by moving out into the population, the military had interposed itself between the attackers and the people. And some of the attacks on them that succeeded were not as bad as previous ones. For example, in March and April, the bombs that detonated were hitting more checkpoints and fewer of the markets and mosques those checkpoints were intended to protect. Roadside bombs also were becoming less effective, for two reasons. Partly, emplacers had less time to dig holes, and some bomb cells were resorting to the hasty method of simply lowering small pressure-detonated bombs through a hole in the floor of a car and then driving off. Soldiers comfortably dismissed those relatively ineffective devices as “drop ’n’ pops.” Even some of the bigger devices used low-grade homemade explosives, indicating that the bombmakers were running low on more lethal material.

COUNTERINSURGENCY INSIDE THE PRISON CAMPS

Two other institutional initiatives also were beginning to have an effect. These were how the Americans handled prisoners and how they raised Iraqi forces. Neither one held the excitement of combat operations, but improving them was essential if the American effort was to become more effective.

For years handling detainees had been the Achilles’ heel of the American operation. Holding and treating prisoners decently didn’t seem the hardest of tasks, but their abuse and torture at the Abu Ghraib facility had been one of the biggest embarrassments, and strategic setbacks, of the war. “We have learned an enormous amount, the very hard way,” Petraeus said later. One hard lesson he listed was that “you cannot safeguard our values by violating them in another country in an endeavor like this.”

Despite that, his counterinsurgency manual didn’t offer much new on the way to deal with detainees. It advised that they should not be abused, but didn’t really have much to say about what to do with them besides that.

Not long after Petraeus took command, he picked Maj. Gen. Douglas Stone, a Marine reservist who had worked for Hewlett-Packard and IBM, to take over the detention operation. At first, there was some question about Stone, with a few officers recommending against him. Petraeus called Gen. James Mattis, a Marine who has had a kind of parallel career to his—first when they were both assistants to a top Pentagon official, then when they both commanded divisions during 2003 in Iraq, and finally overseeing their services’ educational and training establishments. Most important, they are two of the most highly educated generals in today’s military. “Jim, what’s the deal?” Petraeus asked. “Some people advised not to take him.”

“He is the kind of guy you need,” Mattis reassured him. “There will be some degree of care and feeding required, but knowing you, Petraeus, and knowing him, you will be a great team.”

Stone would rewrite the book on effective detention operations. His beginning insight was that there was an insurgency inside the prison camps, and that simply warehousing the prisoners only intensified the opposition there, creating more insurgents out of civilians and more dedicated ones out of existing insurgents. As Stone later put it, “by not emphasizing population protection and the exemplary treatment of detainees, our facilities became breeding grounds for extremist recruitment.” In an official review, he termed U.S. detainment policies for the first several years of the war “an abject failure, a strategic risk to the MNF-I mission and a failure from a counterinsurgency perspective.” In April 2007 alone, the month he took command, there were 10,178 acts of detainee violence inside U.S. prison camps in Iraq.

What was needed, Stone thought, was a campaign that paralleled the larger counterinsurgency effort overseen by Petraeus. He dubbed it “COIN inside the wire.” Stone told his guards to secure the prison population, and especially isolate the roughly 1,000 extremists who had been intimidating the 20,000 other inmates, to the point of holding “trials” of inmates who refused to join them. The first step in separating out the hard core was to learn more about the prisoners, who until that point had been separated by sect but not by ideology. (The sectarian split was about 80 percent Sunni, 20 percent Shiite.) Despite Western perceptions, only a tiny percentage were foreigners. Who were they? What motivated them? The answers they gave in surveys surprised their captors. They were tribally oriented, with 78 percent reporting that they would use their tribal leaders to solve problems. They were not strikingly religious—only 28 percent reported attending services at their mosque on most Fridays. More than 10 percent had been police, soldiers, or security guards at the time of arrest. Most important, only about 4 percent were deemed to be hard-core cases. The vast majority, it seemed, were motivated not by ideology or a sense of grievance, but by minor economic necessity. They planted bombs not to feed their families, but for the cash al Qaeda would pay them, so they could buy small luxuries such as air conditioners or DVD players.

The second move was to begin providing services to the prison population. Basic literacy courses were offered. A civics course was made mandatory. Some 160 Muslim clerics were hired to begin teaching moderate Islam, in courses offered on a voluntary basis. Other courses were given in Arabic, English, history, science, geography, and math. “There is a danger that the insurgency is becoming a vocation,” warned a briefing prepared by Stone’s headquarters, so vocational training was begun in carpentry, textiles, and masonry. The notion was to provide a pathway back to a life in the civilian world where they would not seek to benefit from violence. Stone even proposed giving released detainees a stipend of $200 a month for six months, just to get them on their feet and keep them away from temptation, but that idea died for lack of sufficient support. Instead, he opened a brick factory inside one prison, Camp Bucca, and paid the prisoners for their work, enabling them to build small nest eggs. Typically, the flamboyant Stone had each brick stamped in Arabic, “Rebuilding the nation brick by brick.” After concluding that prisoners who saw their families were less likely to become violent, he set out to enable family visits, running regular bus trips from the cities to the camps.

Stone also urged his subordinates to recognize their own cultural limitations. “Our own individual view of the world tends to limit our perceptions,” he wrote in an overview document, “creating risk when we make the mistake of judging a detainee’s actions in the context of our own culture rather than his own. This is one of the most significant challenges we face in detainee operations.” One of the best ways to defuse a confrontation with guards, they found, was to turn on the large-screen television and play a video of a recent soccer match. And, in another move to reduce tensions and also improve understanding, a pamphlet in Arabic was created to explain the detention process to new arrivals, with a comic book version created for those who were illiterate.

American commanders also seemed to be getting the word. Preparing to lead a brigade of the 10th Mountain Division into Iraq, Col. David Paschal made a point during a training exercise by wearing a dishdasha—the Arab robe that most Iraqi men wear—and playing the role of “an uncooperative detainee.” He threw food and insults at his guards until he finally was tackled and handcuffed. “By participating in the training I was able to experience the level of professionalism and proficiency of my soldiers while at the same time see how they are maintaining our detainees’ safety as well as dignity,” he explained.

Such training continued to be important, because there continued to be a hard core of such uncooperative Iraqis. Even after all of Stone’s improvements were implemented, incidents still occurred. One day many months later, six Navy personnel working as prison camp guards grew tired of having inmates’ feces hurled at them and locked the offenders in a room, then set off pepper spray and turned off the ventilation.

But the strategic view of how to handle detainees had changed, probably irretrievably, as had the atmosphere in most parts of the camps. In April 2008, there were 178 acts of violence recorded in the prison camps, about one-tenth the figure a year earlier. At his farewell ceremony in June, Gen. Stone commented: “History has shown us that leaders often rise from the most difficult of times and circumstances, and we should not be surprised if Iraq’s future leaders are today being held in coalition force custody.” The way they were treated today might shape the country’s policies in the future, he warned.

SURGING THE IRAQIS

An old military aphorism holds that amateurs talk tactics, but professionals talk logistics. In fact, real military insiders often focus on larger personnel issues—raising, training, and equipping the force—because that is the key to long-term, sustainable success.

The U.S. effort to create a new Iraqi military had never gone particularly well. Part of that grew out of the political obstacles facing Iraq: A member of the Mahdi Army, for example, might not be well equipped or trained, but he knew what he was fighting for. By contrast, a member of the Iraqi army, despite having reliable American gear, didn’t know if the government for which he fought would even exist a year later. Even under Petraeus in 2004-5, the training effort had a slow, haphazard feel to it. This was one reason he seemed to shy away from discussing that tour of duty. He maintains that he succeeded then. “It was a massive task and what we inherited was a pretty modest effort,” he said. But still that tour carries about it a whiff of something inconsistent with the rest of his stellar military career.

Years later there was still plenty of room for improvement. In April 2007 a platoon of American soldiers was pinned down outside a mosque in western Baghdad’s Kadhimiyah neighborhood and looked around for some help from Iraqi soldiers. “Of the twenty-seven hundred Iraqi security forces that are in Kadhimiyah, no Iraqi unit would respond,” said Lt. Col. Steven Miska, deputy commander of a U.S. brigade.

Early in 2007, Petraeus asked for Lt. Gen. James Dubik to come out and take over the program to train, equip, and advise Iraqi army and police units. Dubik is an unusual figure, lower key than Petraeus, but like him a light-infantry intellectual. He had spent about half his military career in infantry and paratroop units, and the other half studying and teaching at military schools such as West Point and Fort Leavenworth School of Advanced Military Studies, and civilian universities, including Harvard, Johns Hopkins, and the Massachusetts Institute of Technology. Before joining the Army he had intended to become a priest and had spent a year at seminary.

In an echo of Petraeus’s “Mesopotamian Stampede,” Dubik called his training effort “Meanwhile, Back at the Ranch.” That is, he explained, “We all feel we’re the part of the movie where the spotlight isn’t. The posse is going after the rustlers, rescuing the stagecoach, and running the bad guys out of town.” Meanwhile, Dubik was trying to create a more effective Iraqi security force. In some ways, this was a key counterinsurgency move, because it is axiomatic that it is indigenous forces that finally put down insurgencies, not foreign militaries. Dubik wasn’t particularly taking a page out of the new counterinsurgency manual, but he was consistent with the new strategy in another, larger way: He increased the risk taking in his part of the effort.

First, he deemphasized the transition to Iraqi control. “It’s the indirect approach,” Dubik said. “It’s right out of Aristotle: If you want a happy life, don’t aim for happiness, aim for virtue.” In other words, create an effective Iraqi force, and the transition will follow naturally, without being forced. In Maoist terms, Iraqi forces would not be given power, they would take it.

Second, rather than downplay the infiltration of Iraqi forces by Shiite militias, especially in the National Police, Dubik confronted it, purging its ranks. This wasn’t just a matter of personalities and personal connections, but also of the politics of the country. As Stephen Biddle, who advised Petraeus on the issue, put it, “The problem is, in a country at war, the same pressures will exist against the next commander. The guys in the black baklavas will visit him at midnight.” So the issue was not how to go after individual commanders but how to reduce sectarian influence—again, an instance of the indirect approach. The key, Biddle said, was to initiate a “virtuous cycle” where militias were weakened, so their pressures were less, so Iraqi commanders acted in less sectarian ways, and so the Iraqi population’s opinion of Iraqi forces improved, making those forces stronger.

But there also plainly were some commanders who had to go. “We have gradually cleaned them up,” Petraeus later said. In the National Police, he said, “They replaced the overall commander, both division commanders, all of the brigade commanders, and about seventy to eighty percent of the battalion commanders—and in some cases did it twice.” In the course of those removals, Dubik noted, some 15 internal affairs investigators at the Ministry of Interior, which oversees the police, were killed, and another 14 were wounded.

Third, and probably most important, Dubik accelerated the growth of Iraqi forces, knowing that they might not be as effective at the outset. His goal was “sufficient quantity of sufficiently capable.” He halved the time dedicated to basic training. This was essentially a step away from the professional U.S. military approach of the last 20 years and toward the World War II approach of churning out troops and letting quality show itself and rise up. Under Dubik, the size of the Iraqi security forces increased from 400,000 in June 2007 to 560,000 a year later—actually becoming larger than the active-duty United States Army.

FINISHED BUSINESS

There was also one more American commander who had to go. In June, just as the surge was about to take full effect, Defense Secretary Gates effectively fired the chairman of the Joint Chiefs of Staff, Gen. Pace, who was the last member of the old Rumsfeld team still in place, having been vice chairman of the Joint Chiefs starting in October 2001 and then chairman since October 2005. Gates blamed the removal on Congress, saying he decided not to renominate Pace for the customary second term because “the focus of his confirmation process would have been on the past, rather than the future, and further, there was the very real prospect that the process would be quite contentious.” That may be so, but Gates was also effective at ridding himself of unneeded trouble. Pace became the first chairman in more than 40 years to serve such a short term. With him went the vice chairman of the Joint Chiefs, Adm. Edmund Giambiastiani, who had been seen as even closer to Rumsfeld than Pace.