It Is Folly

19. “It Is Folly”

David Kilcullen had cause to be pleased with himself at the end of 2007. The Iraq War had taken a turn for the better, and the COIN doctrine he’d long pushed had played a substantial role in that turn. The secretary of defense was force-feeding its lessons to the Army, which as a result was changing its ways, in some cases dramatically.

In fact, though, Kilcullen’s mood was apprehensive. He was concerned that too many people—officers, journalists, politicians, and policy makers—would see Iraq as a template for future wars and how to fight them, when, to his mind, it was a botch from the get-go: a terrible mistake that was steered away from disaster at the last minute, and, even then, no one should call the save a “victory.” It was unclear whether the Iraqi government had really adopted the same goals as its American rescuers, whether the country’s sectarian factions were capable of reconciling with one another, or whether Prime Minister Nouri al-Maliki and his top ministers were interested in reconciliation. Nor was it clear, even if all did end well, whether the crushing costs—in lives, money, and strategic position, for Iraq and the United States—had been commensurate with the gains.

Until recently, these issues wouldn’t have disturbed Kilcullen. He’d built up a reputation for speaking his mind, often colorfully; but he’d also come up in the school of thought, common among soldiers, even intellectual ones, that his job wasn’t to make policy but merely to carry it out in a way that minimized the pain and suffering.

His thinking on this point changed over the summer during a series of emails with a few anthropologists who had criticized some members of their profession for helping the US military in Iraq and other theaters in the global war on terror, which they saw as a cover for “indirect colonial rule.” Toward the end of 2006, a group calling itself the Network of Concerned Anthropologists sent around a petition titled “Pledge of Nonparticipation in Counterinsurgency,” which blasted those who engaged in such work for abetting “a brutal war of occupation which has entailed massive casualties.” The executive board of the American Anthropological Association went so far as to condemn such work as “an unacceptable application of anthropological expertise.”

The group’s main target was Montgomery McFate, who had coauthored the intelligence chapter of David Petraeus’s COIN field manual and had also helped design the Army’s program of human terrain teams, which instructed brigades on the tribal cultures of the people they were both fighting and defending. But some in the group also went after Kilcullen, who had a doctorate in political anthropology and who had written articles stressing cultural understanding as a component of COIN.

McFate rattled off several spirited rebuttals, noting that such eminent anthropologists as Margaret Mead and Gregory Bateson had helped the Allied armies understand the cultures of the Pacific Islanders during World War II, and arguing that her own work in Iraq and Afghanistan had persuaded many American soldiers to use less force and kill fewer civilians.

Kilcullen went further, writing a lengthy response to an article titled “Toward Mercenary Anthropology?” in the June 2007 issue of Anthropology Today, a British journal that served as a forum for these charges. Afterward, he extended the dialogue in email correspondence with the magazine’s editor, Gustaaf Houtman, and with Hugh Gusterson, an anthropologist at George Mason University who had helped launch the anti-COIN petition.

In one email addressed to both of them in late July, Kilcullen defended the “humanitarian effects of applying sound social science in Iraq,” citing “measurable data” on the decline in civilian deaths, car bombings, and sectarian murders, as well as the proliferation of peace treaties signed among local tribal leaders—all the result, at least in part, of counterinsurgency methods. As for the war itself, he wrote, “I am not political in any way.” Yes, he was working in the Bush administration, but merely as a “technical specialist.” He’d opposed the war from the beginning, because it was a strategic blunder and because he figured it would take much longer and cause much greater suffering than the policy makers realized. When the administration switched to a COIN strategy at the start of 2007, and when General Petraeus asked him to come to Iraq to help implement it, he agreed, “not because I’m a Bush administration supporter,” he wrote, but because “I felt a sense of obligation to put my money where my mouth was.”

Houtman replied with a gracious note, beginning, “I am very impressed with your frankness and vision.” In an earlier email, Gusterson had drawn comparisons between the COIN anthropologists and the German scientists in World War II. Houtman assured Kilcullen that this wasn’t meant as a personal attack on him but rather as a restatement of the well-known fact “that an academic establishment can sometimes collaborate in the wrong way with their governments, with disastrous consequences.” However, the editor added, not wishing to inflame his correspondence even to that degree, “the vision you sketch here bears no comparison to that at all.”

Over the next few weeks and months, Kilcullen wondered whether Gusterson’s comparison might have some validity. Did his role as a “technical specialist” really let him off the hook? Was joining an administration to help set right one of its own foolish decisions, and thereby make the military occupation of a foreign country go down more easily, a politically neutral act? Was he right to have excluded his judgments on the larger issue—the wisdom of the invasion itself—from his consultations?

The more he thought about it, the more Kilcullen realized that the anthropologists had a point. He didn’t think that they were right. He was proud of the work that he and his colleagues had done; they’d helped clean up a mess left by others, and that didn’t make them aligned with the culprits. To the extent they did share culpability, they had a moral obligation, having invaded the country, to keep it from crumbling into chaos.

Still, he was coming to the view that as a specialist and a professional, he also had an obligation to state his views on the whole picture: not just on how to improve the conduct of the war but also on whether going to war—this war or some future war—was a good idea in the first place. Questions of policy, not merely the execution of it, should very much be within his purview.

At the moment, Kilcullen was working as the special adviser on counterinsurgency for Secretary of State Condoleezza Rice, and he was spearheading a project that might offer a good platform for this new take on his responsibilities.

The project was a counterinsurgency guide for the civilian side of the US government. The project stemmed from Galula’s dictum that insurgency wars were “80 percent political, 20 percent military” and from the line in Petraeus’s field manual that counterinsurgents should use “all instruments of national power.” Contrary to these basic principles, COIN in Iraq was in fact performed almost entirely by the military; even the few civilian field specialists from State and AID were attached to Army brigades, and mainly in advisory roles. There must be plenty of officials in the Departments of Justice, Agriculture, Commerce, and Transportation who knew more than any Army officer about setting up courts, growing crops, financing businesses, and building roads—the elements of the civil society that counterinsurgents were supposed to be creating. But where were they? The premise of the project was that these agencies’ leaders didn’t understand the COIN concept. They needed to be brought in; they needed, to use one of Petraeus’s favorite phrases, to buy in.

The seeds for such a guide had been planted two years earlier, in the spring of 2006, when Petraeus, soon after his COIN conference at Fort Leavenworth, approached John Hillen, the assistant secretary of state for politico-military affairs, and asked him when State and all the other civilian agencies were going to get behind the policy.

In response, Hillen and his assistant, a Foreign Service Officer named Donna Hopkins, organized a similar sort of conference. It took place on September 28 and 29 at the International Trade Center in downtown Washington, a few blocks from the White House. Petraeus’s field manual was about to be published; the interim draft had leaked to the press a few months earlier. A companion volume for the rest of the government might form the basis for a truly integrated civil-military policy. And this conference, Hillen thought, might do for the interagency COIN guide what Petraeus’s Leavenworth conference, seven months earlier, had done for the Army field manual. Hillen, too, it seemed, knew something about buy-in.

The conference—called the Interagency Counterinsurgency Initiative—was a big event. More than a hundred officials attended from seventeen federal departments and agencies. Hillen’s cosponsor was Jeb Nadaner, the Pentagon official who had shepherded the DoD directive declaring stability operations to have the same priority as combat operations. And there were speeches by the usual suspects, including Petraeus himself, Kilcullen, John Nagl, Conrad Crane, Eliot Cohen, Eric Edelman, Janine Davidson, and Sarah Sewall, whose Carr Center for Human Rights Policy at Harvard had cosponsored the Leavenworth conference.

But over the next year, the initiative fell apart. Kilcullen had been assigned to take the lead in drafting a report, but one chapter into it, Petraeus was promoted to commander of US forces in Iraq, and he asked Kilcullen to come along as an aide. Midlevel officials from several different agencies formed a team to write a new draft. After reading it, Conrad Crane told them that it was a mere laundry list of little use. From there, the project deteriorated further. An AID official wrote one section proposing that US law-enforcement officers train local police in a COIN campaign. Crane told him that this was a bad idea; they’d tried it in Iraq, and it failed miserably. The official replied that he couldn’t delete the passage because his bosses were planning to invest a big chunk of their budget in just such a project. As the writing team progressed, the draft devolved, with officials from the disparate agencies disagreeing over clauses large and small. Nothing materialized.

The fact was, the US government wasn’t set up as a colonial enterprise. The Justice Department ran some intelligence and counter-narcotics operations in dozens of unstable and dangerous countries; Agriculture administered food programs abroad; Commerce promoted exports. But these activities followed their own logic: their managers wanted nothing to do with counterinsurgency; any formal ties to the military would hinder their effectiveness.

By the time Kilcullen returned from Iraq in the fall of 2007, the project had dwindled to three core agencies: State, Defense, and AID, with some input from the CIA and the White House. The original idea had been to produce a how-to guide, similar to the Army field manual but for civilian experts assigned to counterinsurgency campaigns. This notion was moot, now that so many agencies had dropped out. So Kilcullen, who took over the writing job on his own, turned it into a guide for senior policy makers, and he decided to give them the advice he now wished he’d dispensed long ago.

This guide, he wrote up front in his draft, would focus above all on the “formulation of policy,” the factors that policy makers should consider when “determining whether the US should become engaged in a COIN campaign overseas”—precisely the question that he’d once thought he had no business addressing. The decision to intervene, he went on, “should not be taken lightly,” as “historically, COIN campaigns have almost always been more costly, more protracted, and more difficult than first anticipated.”

The authors of the Army’s field manual had dropped hints along these lines: brief, elbow-nudging passages noting that insurgency wars were “protracted by nature,” requiring “firm political will and substantial patience” as well as “considerable expenditures of time and resources.” Some of its authors had wanted to insert an explicit warning: if you’re not willing to spend the time and the money, then don’t start a COIN war in the first place. But an Army field manual was no place for policy advice: by nature, it assumed that the political leaders had decided to fight this kind of war; the field manual’s purpose was to instruct soldiers on how to proceed.

Kilcullen had long thought—and still believed—that it was important to study insurgency wars because big powers, including the United States, tend to get involved in these kinds of wars at least once every generation, whether willingly or incrementally, and they can turn disastrous if their armies didn’t know how to fight them. If you were going to engage in a COIN campaign, there were certain things you had to do. Those things were what he’d spent his entire professional life learning and setting down. But he also thought—he’d always thought—that it was usually best not to undertake COIN campaigns in the first place. They took a long time, cost a lot of money, got a lot of people killed, and often didn’t work. This side of his thinking, though, he’d discussed only with his friends.

But now Kilcullen had retired from the military; he was writing a guide for the political leaders who help make the decisions on going, or not going, to war; so he flashed the warning lights brightly. The theme throughout his fifty-page guide, sometimes implicit, quite often unabashed, was basically: don’t do this!

On its first page, echoing the themes of Robert Komer’s Vietnam-era Bureaucracy Does Its Thing and Celeste Ward’s briefing from the previous year on diverging American and Iraqi interests, Kilcullen emphasized that success in COIN often depends on whether the host government is willing to make the sorts of reforms that earn its people’s loyalty. “However great its know-how and enthusiasm,” he wrote, “an outside actor can never fully compensate for lack of will, incapacity, or counterproductive behavior on the part of the supported government.”

This was a greater obstacle than some might think, he went on, because the fact is, a government needing outside help to stave off insurgents probably does lack the will or capacity to fight on its own; and, since insurgencies thrive on real grievances, it probably is resistant to reform. In these cases, a country that intervenes to wage a COIN campaign “will almost always need to co-opt, persuade, or occasionally pressure the local government to give up counterproductive behavior, take genuine steps to reform its actions, win the support of its people, and demonstrate effectiveness and legitimacy.”

Kilcullen then hammered home the main point:

It is folly to become engaged with counterinsurgency in a foreign country unless there is a reasonable likelihood that the affected government will introduce necessary reforms and will demonstrate adequate willpower and capacity to defeat insurgents (or at least be willing to accept advice as well as assistance). Before deciding to provide overseas COIN assistance, US officials must determine how likely it is that the local government will cooperate and how willing it is to undertake necessary reforms.

Reduced to a syllogism, his argument went like this: we shouldn’t engage in counterinsurgency unless the government we’re helping is effective and legitimate; a government that needs foreign help to fight an insurgency generally isn’t effective or legitimate; therefore, we generally shouldn’t engage in counterinsurgency.

Few of these ideas were original; similar points about favorable and unfavorable conditions for insurgents had been made by Galula and the Army field manual. But for Kilcullen, this was a departure from the scope and nature of the arguments that he’d been accustomed to making in a public, political forum.

On July 9, 2008, Kilcullen hosted an all-day workshop in a large classroom at the Foreign Service Institute in Arlington, Virginia, to vet his draft of the civilian COIN guide. It was Kilcullen’s show: he made the opening remarks at 8:30 a.m., as well as the closing remarks at 4:30 p.m., and he led several of the discussions in between.

The most heated arguments took place over what he’d figured would be the eyebrow-raiser: the passage about the “folly” of counterinsurgency and the edict that American officials “must” determine if a foreign government will make reforms before they decide whether to intervene.

Conrad Crane protested that Kilcullen had gone too far. Certainly, Crane said, we should try to use our leverage to persuade the foreign government’s leaders to reform. But how can we make them do that? And besides, if they did everything we demanded, wouldn’t the local population view them as American puppets? The war in El Salvador took more than ten years to settle. Should we have decided not to intervene just because, especially at the beginning, the Salvadoran government was corrupt and oppressive?

Crane lost the argument, not just with Kilcullen but with the majority of officials and consultants in the room. As a compromise, Kilcullen did insert a sentence after his “folly” line: “Unfortunately, there will inevitably be occasions when the assessment of the insurgency situation will weigh heavily against US involvement, but specific US national interests will drive policy makers towards engagement.” But, he added, even in these cases, an assessment should still be made, as it may “prompt caution over the form of engagement to be used, perhaps encouraging a more limited involvement from which a subsequent exit can be made with less political consequences . . . It is often the case that the less intrusive and more indirect the approach selected, the more likely it is to succeed.”

Kilcullen was reciting the argument made by some of the Special Forces veterans at Petraeus’s COIN conference nearly two-and-a-half years earlier: that a small footprint is often best, in that it gives us more leverage and focuses attention on the fact that the fight is ultimately the host government’s, not ours.

Dealing with the Iraqi insurgency required a lot of troops, Kilcullen wrote on, but that war wasn’t a typical case of COIN. America had sent troops in the first place “to bring about regime change,” not to help prop up a government. When the social and political order collapsed—a nearly inevitable result of the regime change itself and of the subsequent (and still mystifying) decision to disband the Iraqi army—America had an obligation to repair the damage it wreaked. But, he noted, Iraq should not be regarded as a case to emulate when thinking about counterinsurgency.

Crane wasn’t the only one in the room who had problems with Kilcullen’s language. Several others thought that “folly” was too snarky a word for this sort of document and that it seemed presumptuous to tell senior officials they “must” weigh certain factors before deciding on a policy.

But Kilcullen noticed that these critics weren’t quarreling with him on substance; they only thought he should express the points more cautiously. He fired back. As the professionals who executed this fuckup, he said, we have to lay down a marker; we have an obligation to say what we think. If our bosses think we’ve gone too far, they’ll notch it down. But let’s not censor ourselves preemptively.

By a vote, the room agreed with Kilcullen.

Over the next few weeks, he tightened the prose, circulated revisions for final comments, and sent the finished product to the officials who would sign it.

He was feeling good about having expressed his views so clearly, and completely, for once. He gave advance copies of the guide to a few journalists he knew. When one of them, Spencer Ackerman, called to ask some questions, Kilcullen opened up.

If future policy makers follow the guide’s lessons, he said, maybe they’ll avoid another war like Iraq. He then added, “The biggest fucking stupid idea was to invade Iraq in the first place.”

Ackerman’s story, written for the Washington Independent, quoted Kilcullen’s line in full. The Huffington Post, a much more widely read site, picked up the story, running it under the headline “Rice Adviser: Iraq Invasion Was ‘F*cking Stupid.’”

Within hours, Kilcullen got a phone call from Eliot Cohen. “What the fuck, Dave?” Cohen exclaimed. Condi had seen the piece and was very displeased.

Kilcullen posted a response on the Small Wars Journal site, stating that Ackerman “did not seek to clear that quote with me, and I would not have approved it if he had.” Tellingly, Kilcullen did not retract, or deny uttering, the remark; nor did he claim, as many do under similar circumstances, that he’d been speaking off the record. In fact, he restated his point, though more politely. “If [Ackerman] had sought a formal comment,” Kilcullen wrote, “I would have told him what I have said publicly before: in my view, the decision to invade Iraq in 2003 was an extremely serious strategic error.”

Cohen and Rice were hardly mollified. An “extremely serious strategic error” was only a notch or two below “fucking stupid” when it came to remarks that a special adviser to the secretary of state shouldn’t say in public about the defining decision of a presidency.

Ten days later, on August 8, a brief war broke out between Russia and the Republic of Georgia, a crisis atmosphere enveloped the administration, and the F-bomb affair vanished in the fog. The next time Rice saw Kilcullen at a meeting, she looked at him as if she should be angry but couldn’t remember why.

Meanwhile, the civilian COIN guide moved forward. Robert Gates at the Pentagon and Henrietta Fore, the administrator of AID, gave it their signatures. But Condi Rice didn’t, not right away. Cohen told Kilcullen that she had definitely received it and had probably read it. But by the end of the year, it still sat on her desk. Was she still ticked off at Kilcullen? Did she realize, reading it in the context of the Ackerman interview, that the guide was a harsh critique of her own time in office?

Rice finally signed the guide on January 13, 2009. Kilcullen’s prose was left unchanged. But the document was dead on arrival. Barack Obama would be sworn in as president in a week’s time. New teams tend to ignore what the old team leaves behind, especially something dropped on the doormat so late in the day.

In this case, though, the warnings might best have been noted.