A New American Way of War

22. “A New American Way of War”

David Petraeus had once hoped, even expected, that after Afghanistan would come the promotion of his careerlong dream: to chairman of the Joint Chiefs of Staff. But the post went instead to his West Point classmate Martin Dempsey.

With the pending budget cuts and the doctrinal retreats, Petraeus half convinced himself that it was just as well he didn’t get the job. As he told one friend, “I’m better at buildups than drawdowns.”

He’d learned back in December 2010 that the promotion wouldn’t be coming. Bob Gates, the secretary of defense and an ally in the quest, delivered the news during a trip that month to Afghanistan. Petraeus was crushed but not too surprised. He knew about what he called, with a rueful smirk, the “Petraeus problem.” A number of President Obama’s White House aides didn’t trust Petraeus; they thought that he’d tried to box in the president during the deliberations on Afghan war policy; Rahm Emanuel, Obama’s chief of staff, openly worried that the most famous general of his generation might possess a white-knight complex, an urge of destiny to run for president. Petraeus disabused him of this notion.

But even those who dismissed Emanuel’s fears were leery of giving too much power to a general of such prominence, ambition, and agility. The JCS chairman was a position of potentially enormous power; he would have control of the Joint Staff, a multiservice body of nine directorates, manned by several hundred of the military’s smartest officers. The last chairman who’d molded the Joint Staff into a bureaucratic weapon—who’d figured out how to use its collective talents and inside knowledge to win arguments and pursue his agenda—was General Colin Powell, during the presidencies of George H. W. Bush and, briefly, Bill Clinton. It was no accident that almost every chairman since had been fairly unassertive by inclination. Doug Lute, who’d known Petraeus since their days at West Point and who was now Obama’s adviser on Afghanistan, told some of his colleagues, “Dave Petraeus will never be chairman as long as anyone still remembers how Powell ran circles around the interagency process.”

At the same time, any president, especially a Democrat, would want to keep a general like Petraeus on the inside. He might not run for high office, but he could go write policy papers for a think tank like the American Enterprise Institute, where he had many friends. A paper by retired general Petraeus, criticizing the administration’s policy on Afghanistan or the defense budget, would be treated as front-page news. Obama would be asked about “the Petraeus study” at his next press conference. Out of uniform and cast to the winds, Petraeus could set the national-security agenda.

After Gates gave Petraeus the news, they discussed other job possibilities. Petraeus raised one idea: director of the Central Intelligence Agency. He had the right experience, as a consumer of intelligence and as a prodder and a collector, not just in Iraq and Afghanistan—where he’d worked with JSOC, Special Forces, and the entire range of the intelligence community—but also during his time running counterterrorism operations in Bosnia. Gates, who’d been a CIA director himself, thought that Petraeus would be terrific at the job. It would keep him on the front lines of the war on terror (the Agency operated the drone strikes in Pakistan, Yemen, Somalia—every place except Afghanistan, where the military controlled them), and it would fit well with his analytical talents.

Petraeus started lobbying informally for the job—with, crucially, Gates’s strong endorsement. He was eager to demonstrate that he could be a solid team player. Three months after his conversation with Gates, he had a chance. He returned to Washington for congressional hearings on Afghanistan and, during his testimony, defended every aspect of the administration’s policy, including Obama’s decision to set a date when American troops would start to come home, a decision that most COIN scholars—and, privately, many of Petraeus’s fellow senior officers—had sharply criticized.

Toward the end of that trip, he met with the president to discuss his recommendations on Afghanistan—and, on the side, his own future. Obama offered Petraeus the CIA job on condition that he first resign from the military, to avoid conflicting loyalties. Petraeus accepted the deal.

A mere six weeks later, he once again took the long flight from Kabul to Washington, so he could stand in the East Room of the White House while Obama announced the grand shuffle in his national-security team: Gates was retiring; Panetta was shifting from the CIA to replace Gates at the Pentagon; and Petraeus was leaving Afghanistan, several weeks ahead of schedule, to replace Panetta at CIA.

On June 30, the Senate confirmed Petraeus’s nomination, 94–0. He returned to Afghanistan for his last two weeks as commander in a state of rush and ambivalence: on the one hand, eager to turn a new page in his life (Gates had advised him to get out of the car at headquarters in Langley all alone, to cut loose the entourage that he’d cultivated over the past twenty years as an officer); on the other hand, frustrated that his final command post was ending on a less than triumphant note.

He retired from the Army on August 31, with a full-colors ceremony on the parade field at Fort Myer, not far from the Pentagon. He was sworn in as CIA director six days later.

In the weeks between his return and his retirement, Petraeus attended a spate of welcome-home parties, with many of his fellow COINdinistas in attendance.

At one party, John Nagl rose to make a toast, hailing the guest of honor as a model soldier-scholar deserving mention alongside Ulysses S. Grant and Dwight D. Eisenhower as American generals who had commanded successfully in two theaters of war—and perhaps someday, Nagl added with a wink and a smile directed at his former mentor, he would follow those generals’ footsteps in another way as well.

Petraeus stared straight ahead with a deadpan expression, avoiding the slightest visible response to both of Nagl’s improbable references: the prospect of a Petraeus presidency and the notion that he’d been successful in Afghanistan as well as Iraq.

Afterward, as the crowd around Petraeus dwindled to just a few friends, Dave Kilcullen approached. “Do you really believe what John said about two successful commands?” he asked.

“No,” Petraeus said, shaking his head slowly. “There’s still a long way to go.”

•  •  •

On June 22, 2011, a few weeks before the end of David Petraeus’s tenure as commander in Afghanistan, President Obama delivered a fifteen-minute address on national prime-time television, announcing that he would withdraw ten thousand troops from Afghanistan by Christmas and another twenty-three thousand by the end of the following summer.

This amounted to pulling out the entire “surge” force: all thirty-three thousand of the extra troops that he’d decided to send in December 2009 after the series of ten meetings with his national-security advisers. These troops were scheduled to rotate out by then anyway, but Obama was now saying that they would not be replaced by a new rotation of troops. The sixty-eight thousand troops that remained after their departure would be it—and their ranks would be reduced steadily until 2014, the year when the NATO ministers had decided, with Karzai’s consent, to end their occupation altogether.

In short, Obama was announcing the end of the surge—and, with it, the end of its underlying strategy.

His top defense advisers were stunned. They’d known that some troops would be coming out: when Obama announced the escalation eighteen months earlier, he also announced that he would begin to pull them out in July 2011. But the advisers had urged him not to withdraw more than a few thousand, at least for the moment.

They’d apparently forgotten, or had never taken seriously, Obama’s words at the last of those ten meetings, the Sunday afternoon session in the Oval Office two days before his West Point address, when he asked these advisers whether the thirty thousand extra troops he’d decided to send would improve things to the point where the Afghan army could lead the fight within eighteen months. He warned them that this would be the final escalation; they shouldn’t count on him to double down or extend their stay if they couldn’t do the job in that time span.

He’d viewed the generals’ COIN strategy as a gamble—a worthwhile gamble, but a gamble nonetheless—and he wasn’t going to bet more lives or treasure than he’d already put on the table. At that Sunday meeting, the defense secretary, the JCS chairman, and the commander on the ground—Gates, Mullen, and Petraeus—had assured him one by one that, yes, they could accomplish the mission with the resources and time that he was giving them.

Now those eighteen months had passed, and the gamble hadn’t paid off. By all accounts, only a handful of Afghan army units were capable of taking the lead. So, as he’d warned, Obama was scaling back the war substantially.

He didn’t put it that way in his TV address. To the contrary, he all but declared victory. The United States and NATO, he proclaimed, “are meeting our goals” of defeating al Qaeda, reversing the Taliban’s momentum, and training Afghanistan’s security forces. So it was time to begin turning the mission over to the Afghans.

By that measure of success, he was right. Two months earlier, on his orders, a team of SEALs had killed Osama bin Laden—the mastermind of the 9/11 attacks, the figure who’d provoked the American invasion from the outset, the wanted-dead-or-alive target that George W. Bush had been chasing for years but had never been able to find.

Yet if Obama’s war aims had been merely to kill bin Laden, push back the Taliban a bit, and step up the training of the Afghan army, he wouldn’t have needed to send so many extra troops to begin with. He’d decided to send thirty-three thousand more troops because that number (augmented by seven thousand more from NATO allies) was the minimum needed to carry out the counterinsurgency strategy that Petraeus and McChrystal were urging him to adopt.

Pulling them all out, and thus returning to the level of troops in place before the surge, meant pulling the plug on COIN.

Obama wasn’t getting out entirely. The Special Ops commandos who were mounting raids on Taliban hideouts, the drones and other aircraft that were dropping smart bombs on high-value targets both in Afghanistan and across the border in Pakistan—those forces, he emphasized, would stay in place, would not be reduced at all.

But they were the tools of a counterterrorism strategy: the “CT” approach that Joe Biden had favored in the White House debates toward the end of 2009. They had once been elements of a counterinsurgency strategy: tools that would clear an area of insurgents, after which conventional troops would hold the area and then build support for the local government by helping to provide basic services. But with cuts as deep as those Obama was now announcing, “hold” and “build” could not be sustained—not by American troops, anyway, not for long. Pretty much all they could do was go after insurgents and train the Afghan army to go after insurgents once the NATO troops were all gone.

Those forty thousand extra troops that Obama had deployed for the more ambitious strategy—the troop levels and the strategy that his top generals had requested—weren’t able to hold or build much in any case, not even after a year and a half of a COIN-driven surge. The failure had stemmed not so much from the troops or their commanders as from the nature of Afghanistan and its regime. That being the case, there was no reason to believe that another year and a half and thirty thousand troops would make much difference. Nor, probably, would another ten years and one hundred thousand troops.

Back in March 2009, just two months into his presidency, while announcing his first escalation in Afghanistan—the twenty-one thousand extra troops sent to provide security for the country’s elections—Obama had said that he would not “blindly stay the course” or “provide a blank check” for the mission.

He was now making good on that promise.

•  •  •

On September 30, General Martin Dempsey was sworn in as chairman of the Joint Chiefs of Staff. Three weeks earlier, General Ray Odierno had taken the oath as the new chief of staff of the Army.

Had the war in Afghanistan gone better, their promotions might have been seen as a solidification of COIN culture in the highest echelons of the Pentagon.

But the ground was shifting rapidly under their feet.

Within days of his ascension, Dempsey got word from the White House—as did Secretary of Defense Leon Panetta—that the president wanted a formal review of American military policy. Specifically, he wanted to put an end to interventions—large-scale, protracted “stability operations”—like those in Iraq and Afghanistan. Their costs were unsustainable in a time of fiscal pressures; their benefits were uncertain in any event.

For the next two months, the Pentagon’s leaders—first just the chiefs, then an expanded group that included the senior civilian officials—met roughly once a week to discuss a wide range of military issues: a possible shift of forces from Europe to the Pacific, another shift from manned aircraft to drones, the proper mix of active-duty soldiers and reservists, of conventional troops and Special Ops commandos.

And they talked about the future of stability operations, the issue of greatest concern to those who had taken part in the COIN revolution of the previous decade and had seen their own views on the nature of warfare evolve and harden. Among the officials in the room who felt the pangs most intensely were Ray Odierno, the COIN convert who was now the Army’s top general; Michèle Flournoy, the former president of the Center for a New American Security, who was now in her final months (by her choice) as undersecretary of defense for policy; and Robert Schmidle, the Marine three-star general who organized the first Irregular Warfare Conference at Quantico back in the fall of 2004.

Schmidle, who was now deputy director of the new US Cyber Command, had been recruited by Dempsey to run his transition team in the weeks before taking over as Joint Chiefs chairman. Now that a strategic review was in the works, he asked Schmidle to stay on a few more weeks to run it again alongside Flournoy’s deputy, Kathleen Hicks. Schmidle and Hicks had worked together once before, on the team of analysts—including, most prominently, David Kilcullen—who wrote the chapters on irregular warfare for the Pentagon’s 2005 Quadrennial Defense Review.

The officials involved in these new discussions understood the president’s—and the American public’s—reluctance to get embroiled in another Iraq. But they were also keen to preserve what the chiefs were now calling the “lessons learned from ten years of war.” Among these lessons: that conflicts of the future were likely to be a mix of offense, defense, and stability operations; that, in such wars, awareness of the local culture would be as important as an assessment of the enemy’s order of battle; and that, therefore, it was essential to retain officers who were skilled in this sort of warfare—and to educate and train the coming generation of officers in its principles and techniques.

Schmidle knew that, historically, few American presidents had plunged into long insurgency wars on purpose; usually they’d backed into them, or they’d sent troops to a foreign land for a completely different reason, then got sucked into the “stability” mission as the country fell apart. Vietnam was a case of the former; Iraq, the latter. The pattern tended to repeat itself, one way or another, every generation—at intervals just long enough for the people in power during one war to forget the lessons of the previous war.

The commanders of these past ten years of war—Petraeus, Odierno, Chiarelli, McMaster, and the others—had to excavate the few useful lessons from Vietnam, because the senior Army officers of the earlier generation had deliberately buried the records. And now the new top officers were determined that the next time a president got them involved in a war of this sort, a ledger of lessons, a guide to action—their ledger, their guide—would still be in active circulation.

In one slide of their PowerPoint briefings, Schmidle and Hicks assembled a list of irregular wars, low-intensity conflicts, and counterinsurgency campaigns that the American military had waged over the years. They focused particularly on the “moot-wah” wars of the nineties—among them Somalia, Haiti, Panama, Grenada, Bosnia, Kosovo—the same wars that Petraeus, Nagl, Crane, and others had cited to justify a revival of COIN doctrine. They argued that even another war in Korea—one of the few scenarios that would involve large-scale maneuvers of tanks and artillery—might require stability operations if North Korea’s regime crumbled and the country tumbled into chaos.

By December, the top Pentagon officials were ready to make their case to the White House. Several meetings, in one building or the other, took place over the next few weeks.

Obama himself attended and took an active part in three of these meetings: two in the Oval Office, one in the Pentagon. He quickly grasped the generals’ argument: that no one could predict what sorts of wars we might wind up fighting and that skills in stability operations should be retained in case this particular kind of war arose again for whatever reason. But that left open the question of priorities. How far did the military and the country need to go to make preparations for that kind of war? How many troops and how much money should be pegged to stability operations instead of other types of operations?

•  •  •

By the start of 2012, a consensus had formed around an answer.

On January 5, Obama traveled across the Potomac, stood at the podium in the newsroom of the Pentagon, and—with Secretary Panetta and the entire Joint Chiefs of Staff standing behind him—outlined the main tenets of the new “strategic guidance” for the Defense Department.

No president had ever before held a press conference at the Pentagon; Obama’s appearance to trumpet this policy signaled clearly that it was far from routine.

The eight-page document, titled “Sustaining US Global Leadership: Priorities for 21st Century Defense,” focused mainly on the shift of security concerns and resources from Europe to the Pacific and the role that the American military’s technological edge could play in countering the rise of China’s power.

But one section also dealt with stability operations. Obama summed it up in his opening remarks: “As we look beyond the wars in Iraq and Afghanistan—and the end of long-term nation-building with large military footprints—we’ll be able to ensure our security with smaller conventional ground forces.”

For the Army, the key phrase here was “smaller conventional ground forces.” For the COINdinistas, it was “the end of long-term nation-building with large military footprints.”

The document noted that “for the foreseeable future,” the United States would “take an active approach” in countering threats posed by al Qaeda and other jihadist terrorist groups around the world. But, it went on, this approach would “emphasize nonmilitary means,” and, to the extent that arms were necessary, it would focus more on assisting other nations’ security forces so that they could stave off homegrown threats on their own.

Occasionally, the guidance acknowledged, American forces would need to get involved in “directly striking the most dangerous groups and individuals.” But even then, it added, in italics: “Whenever possible, we will develop innovative, low-cost, small-footprint approaches to achieve our security objectives.”

In the key passage, it stated that the American military would remain “ready to conduct limited counterinsurgency and other stability operations if required” and, therefore, would “retain and continue to refine the lessons learned, expertise, and specialized capabilities that have been developed over the past ten years . . . in Iraq and Afghanistan.”

“However,” it proclaimed, again in italics, “US forces will no longer be sized to conduct large-scale, prolonged stability operations.”

Back in late 2005, David Kilcullen, Bob Schmidle, and the other authors of the Pentagon’s Quadrennial Defense Review had called for boosting the Army’s size and resources, so that it could “conduct a large-scale, potentially long-duration irregular warfare campaign.” Now, six years later, Obama was reversing course. With his strategic guidance, he wasn’t merely saying “Never again” to another war like Iraq or Afghanistan, at least as long as he was commander in chief; he was telling the generals that when they calculated how many troops were needed for various contingencies, they were not to assume that one of those contingencies might be a long war involving lots of troops engaged in anything like a COIN campaign.

The strategic guidance nullified another key COIN document from late 2005: Defense Department Directive 3000.05, which declared that stability operations were “a core US military mission” to be “given priority comparable to combat operations.”

That was clearly no longer the case. Planning and preparing for those sorts of wars would be a part of the armed forces’ repertory but not a centerpiece—not a scenario to shape or justify its size or its budget. It was no longer a “core mission” of the American military.

•  •  •

Exactly how to go about retaining and refining the lessons learned from ten years of war, Obama left to the chiefs. Odierno and Dempsey spent a lot of their time filling in the blank spaces.

One space they filled was to transform the Army’s Maneuver Center of Excellence, a school for armor and infantry officers at Fort Benning, Georgia, into a training ground for “full-spectrum operations,” combining tank maneuvers with counterinsurgency and humanitarian assistance, devising scenarios and exercises in which junior officers would have to switch back and forth from one mode of warfare to another, teaching them how to make judgments and decisions in a complex environment.

As the center’s new director, Odierno appointed H. R. McMaster, now a two-star general, who’d been the peerless tank-killer in Desert Storm and the COIN commander par excellence in Tal Afar, Iraq.

Odierno also drew up new rules for the Army’s promotion boards, a long list of skills and experiences that colonels must amass before advancing to the rank of brigadier general. Scoring well at commanding a combat battalion would no longer be sufficient; they’d also need experience at posts that had required cultural awareness and complex decision-making. And to make sure that the guidelines were followed, Odierno would personally pick the generals who served on the boards—generals that he knew were in synch with his approach.

But the larger picture was still hazy. There had never really been an official, thorough study on the military implications of a post–Cold War world. Now the Army was facing a post–Iraq War world as well. What were its roles and missions in this world? If it wasn’t going to fight the Red Army in Europe, or wage large-scale, prolonged stability operations, what should it be sized, structured, trained, and equipped to do? What priority should be given to options for short-term or small-scale stability ops? And what should those ops involve specifically: Would regular Army soldiers be trained and organized to fight in small wars or merely to “advise and assist” allied armies, embedding small teams of trainers and enablers in their midst? Special Operations forces had focused on these missions in the past; to what extent would, or could, the regular Army now join along?

During the initial meetings where the chiefs discussed these ideas, Dempsey wondered out loud whether they were all teetering toward un-charted territory: some nexus of Special Ops forces, drones, networked intelligence, and cybertechnology that added up to “a new American way of war.”

It was reminiscent, though he might not have known it, of the moment just before Operation Desert Storm, more than twenty years earlier, when Andy Marshall, the Pentagon’s director of net assessment, wondered if the US Air Force, with its emerging smart bombs and high-tech sensors, might be on the edge of a “revolution in military affairs”—a concept and phrase that Donald Rumsfeld drew on when planning his light, swift shock-and-awe invasion of Iraq.

In some respects, the Obama-Panetta strategic guidance—with its vision of future forces that were “smaller and leaner . . . agile, flexible, ready, and technologically advanced”—bore a striking resemblance to the Marshall-Rumsfeld language of “military transformation.”

There were, of course, differences. Obama saw force as a means of stabilizing the international order, and in his surprisingly numerous decisions to use force, he stressed the legitimizing role of fighting alongside allies in coalitions; Rumsfeld, by contrast, saw force as a tool for toppling rogue regimes that were impeding American primacy, and he celebrated the new wonder weapons as gateways to unilateralism.

But both tugged on technology as an alluring anesthetic that numbed the senses to the pain of warfare: the grit, grime, and mayhem of its consequences on the ground. In this sense, neither vision was so different from the old “American way of war,” with precision bombing now serving as a substitute for annihilating firepower. The modern variant was more humane, less grotesquely destructive, but each had the effect, in its own way and time, of rendering war easier, more comfortable.

In the article that he ghostwrote for General John Galvin in the mid-1980s, David Petraeus, then an ambitious young major, noted the military’s tendency “to invent for ourselves a comfortable vision of war, a theater with battlefields we know, conflict that fits our understanding of strategy and tactics . . . that fits our plans, our assumptions, our hopes, and our preconceived ideas.” Galvin and Petraeus had urged the military’s leadership to prepare for what they called the “uncomfortable wars” ahead: wars of subversion, terrorism, and guerrillas that conformed to no assumptions, confounded all preset plans, and required a whole new way of looking at warfare. Petraeus spent much of the next twenty years preparing to tear down the old edifice and construct a new one that he saw as better suited for the age.

But these uncomfortable wars turned unacceptable once the stakes seemed minor, the duration seemed endless, or the chance of “winning” (however the word was defined) seemed dim. Comfortable wars—small, cheap, and, for all practical purposes, invisible to those not directly fighting them—were back in fashion.

•  •  •

Two or three times a year, Conrad Crane goes back to West Point to attend a conference or give a lecture, and when he does, he always makes a point of strolling through the cemetery. It’s laid out on flat, grassy land overlooking the Hudson, an old country churchyard, modest and intimate. Crane regards it as the most hallowed plot of ground on the planet, “the soul of West Point,” a sanctuary of its values: honor, sacrifice, service, and, too often, tragedy.

Six thousand former cadets are buried there, dating back to the American Revolution. The carvings on the older headstones are rich with the lore of history. But Crane usually heads straight to the back rows—Section 36, as the area is known—where the fallen from the Iraq and Afghanistan Wars are laid to rest.

Fifty-nine West Point alumni were killed in Iraq, and nineteen of their bodies lie here. They include nine lieutenants, eight captains, a major, and a colonel. All but four were younger than thirty years of age; one was over forty. They left behind a total of seventeen children (none older than twelve years of age), eleven widows, and four fiancées.

At least three of them were once students of Crane’s, a fact that pains him when he stands over their gravestones. He is intensely aware that some of them died while following the instructions he’d helped write in the counterinsurgency field manual. He hopes—he has to believe—that the manual, and the new strategy it brought forth, saved more soldiers’ lives than it cost and that the outcome of the wars, still unclear, will at some point prove worthy of their sacrifice.

But he knows he will never be sure.

•  •  •

David Petraeus and his fellow plotters succeeded in fomenting great change. By the time he hung up his uniform, not quite five years after signing his counterinsurgency manual, the American Army had evolved into a different institution. It was more flexible, more adaptive; it was, in John Nagl’s phrase, a “learning organization.”

In the aftermath of wars, especially unpopular ones, armies tend to revert to traditional practices. But this was less likely to happen after the wars in Iraq and Afghanistan. There would be no going back to a frame of mind that defined “war” strictly as a titanic clash between uniformed foes of comparable strength—and not just because the prospective foe in that clash, the Soviet Union, had in the meantime imploded. Another factor at play was that an entire generation of American officers had risen through the ranks fighting what were once called small wars, waged among the people in villages and cities, wars in which lieutenants often took as much initiative as commanders, and soldiers of all rank were as attentive to the local culture as to the enemy’s order of battle.

It was extremely unlikely that official Army doctrine would ever again refer to these sorts of battles as “low-intensity conflicts,” much less as “military operations other than war.” The colonels and generals of the post-Petraeus era had spent what seemed like a lifetime fighting in these sorts of battles; they were not low intensity, and they certainly felt like wars.

Petraeus, Nagl, and the others fulfilled their main ambition of wresting away the Army from Cold War habits and adapting it to the new era: an era of what John F. Kennedy had called, a half century earlier, “another type of warfare, new in its intensity, ancient in its origins—war by guerrillas, subversives, insurgents, assassins.” These were the sorts of threats and modes of warfare that the world was once again facing. And to the extent that the regular US Army, not just the Special Forces, had learned how to fight them, it was the result of the plot that the West Point Sosh mafia had sired.

But knowing how to fight these wars didn’t necessarily mean winning them. There’s an old military adage: “The enemy has a vote.” You can go into battle with a brilliant plan, but if the enemy adapts and shifts gears, the plan is rendered worthless after the first shots are fired. In counterinsurgency wars, it’s not just the enemy that has a vote; the ally does, too. If you send troops overseas to bolster a regime whose leaders lack legitimacy or the will to reform, the most brilliant strategy—and strategist—will have little chance of prevailing.

Robert “Blowtorch” Komer realized this, too late, in Vietnam. Peter Chiarelli and Celeste Ward experienced their own Komer moment in Iraq, as did Steve Biddle in Afghanistan and Dave Kilcullen while reassessing the entire COIN enterprise.

Petraeus knew all this, too. He’d read Komer and the other historians and chroniclers; he’d absorbed every nuance and checklist of caveats in the classic theoretical tracts. Yet, in part from overconfidence, in part from inertia, he came to view the doctrine as a set of universal principles: “the laws of counterrevolutionary warfare,” as Galula had put it.

As a commander, Petraeus had stressed the importance of getting “the big ideas” right, but the ideas in COIN theory weren’t as big as he seemed to believe. Counterinsurgency is a technique, not a grand strategy. Field manuals are guides for officers preparing to fight in a specific setting; and in that sense, a COIN field manual isn’t so different from a field manual on mountain warfare, amphibious operations, or armored maneuver combat. If the setting is appropriate, and the conditions seem ripe for success, a good field manual is essential. But if a mountain is too steep to climb, if a beach is too turbulent to storm, or if a field is too cluttered for tanks to maneuver across, the manual won’t be worth much—and it’s a commander’s responsibility to say so.

In assessing the prospects of a COIN campaign, if the insurgents are out of reach, or if the government being challenged is too corrupt to reform, or if the war is likely to take longer and cost more than a president or a nation is willing to commit, in these cases, too, it’s the commander’s responsibility to say so. David Petraeus knew that all these things were true of the war in Afghanistan, but he stopped short of saying so; he thought he could overcome the odds, as he had so many times before.

Petraeus’s COIN field manual caused such a stir when it was published at the end of 2006 because the Army brass had deliberately avoided writing such a guide—had avoided even thinking about the subject—for two decades, and now here they were, in Iraq, caught in a war that had begun as a blitzkrieg invasion but had quickly devolved into an insurgency revolt.

Ideally in a war, civilian authorities set the strategic goals, and military commanders devise the plans and tactics to achieve those objectives. But the Bush administration had no strategy in Iraq beyond toppling Saddam; and in Afghanistan, it had no strategy beyond ousting the Taliban. Petraeus’s manual seemed to offer a way out, a recipe for action, a set of plausible criteria for a victory of sorts.

The manual’s authors, including Petraeus, wove this perception into being. They set out explicitly to write something much more than the usual field manual; they meant it to serve as a manifesto—and, in its official acceptance, the spearhead—of a revolution within the Army as an institution and a culture.

This was asking too much of mere doctrine. COIN—the field manual and the long history of ideas it embodied—was like a set of instructions on how to drill an oil well: it didn’t guarantee that there was oil in the ground or that drilling for oil was the wisest energy policy.

It was a useful guide for conflicts where the insurgents could be contained within a clear-cut area and where the intervening power and the foreign government had interests that aligned. The COIN approach helped produce stunning results in parts of Iraq and Afghanistan: Mosul, Tal Afar, Anbar Province (whose Awakening then spread to many other Sunni districts), and, briefly, the Shinwari pact in Nangarhar Province.

The distinctive thing about all those areas was that the Americans and the local powers—the mayor, provincial council, or tribal elders—shared common interests or, at least, common enemies.

But the central governments, which COIN was ultimately supposed to strengthen, were another matter. A much larger, more pertinent “big idea”—which neither Petraeus nor any other outsider could ultimately control—was the identity and interests of the foreign country’s ruling elite. If that identity and those interests obstructed the regime’s willingness or ability to govern its people with legitimacy, and if the intervening power had little leverage to alter this fact, then—as Dave Kilcullen concluded in the end—it was “folly” to embark on a counterinsurgency campaign in the first place.

In that respect, the modern age itself has reduced much of the whole COIN concept to folly. The most often-cited models of successful counterinsurgencies—Malaya, Kenya, the Philippines, Algeria, and Northern Ireland—were colonial wars. There was no divergence of interests between the country sending the troops and the local authorities welcoming them, because they were one and the same or at least fragments of the same empire. When Field Marshal Sir Gerald Templer commanded the British colonial army in Malaya, he didn’t have to persuade the local government to do what he wanted; for all practical purposes, he was the local government.

The successful COIN campaigns of lore were also wars of stunning brutality: in Malaya, four hundred thousand civilians forcibly relocated and thousands of insurgents starved to death; in the Philippines, sixteen thousand guerrillas killed in battle, and two hundred thousand civilians killed by disease and starvation spreading through the relocation centers (formally called “concentration camps”); in Algeria, the routine use of torture in the “pacification” campaigns.

This was the dark side of counterinsurgency, but it had also been seen by its practitioners as an essential side. And it was far too dark an alleyway for a democracy to roam in the global-media age—too savage, and prone to drag on far too long.

Even John Nagl allowed as much. He would often tell friends that, judging from the lessons of Malaya, the ideal counterinsurgency campaign was one fought “on a peninsula against a visibly obvious ethnic minority before CNN is invented.” Eating soup with a knife, as he’d quoted Lawrence cautioning, was “messy and slow” indeed.

Modern-day America does not like messy and slow, especially slow. David Petraeus had written in his PhD dissertation, nearly a quarter century before retiring from the Army, “Vietnam was an extremely painful reminder that when it comes to intervention, time and patience are not American virtues in abundant supply.”

Were he to update his thesis, he might note that Iraq and Afghanistan are reminders as well.

Yet Nagl did not step back from his standing as COIN’s most ardent public enthusiast, nor did Petraeus repudiate the doctrine he’d done more than anyone to revive. As they saw it, and still do see it, American armed forces might once again find themselves in an insurgency war. It had happened at least once each generation, and rarely by design. Such wars, like any wars, went better if the Army sent to fight it knew how: if its commanders knew that firepower alone wouldn’t do the trick and might make things worse; and if the soldiers had some understanding not only of battlefield tactics but also of the local culture of the people they were protecting.

Petraeus, Nagl, and their compatriots reshaped the American Army so that a core group of officers knew how to do those things—and a body of doctrine existed so they could train their successors to do them, too.

But in the aftermath of the Iraq War and as American involvement in Afghanistan was winding down, it seemed unlikely that any president would soon get involved in this sort of war again, not on a scale as large or long-lasting as those wars had proved to be. COIN may have made for a “smarter” or “better” war than the bash-down-doors, shoot-first alternative, but that didn’t make it a smart war or a good war. The COIN strategists made the American military more adept at fighting this kind of war, but they didn’t—they couldn’t—succeed at making this kind of war acceptable, either to the American public or to the people in the lands where it was fought. They didn’t, they couldn’t, instill the broader message that these kinds of wars—protracted, large-scale battles in the “long war” on global terror—were now permanent fixtures on the landscape and that the COIN way was the right way, the only way, to fight them.

In the end, they didn’t, they couldn’t, change—at least in the way they intended to change—the American way of war.