Though doing so revealed relatively little that those who have been paying attention didn’t already know, the Washington Post’s publication earlier this month of “Lessons Learned” documents produced by the Special Inspector General for Afghanistan Reconstruction—the so-called Afghanistan Papers—has led to some stock-taking about the long war in Afghanistan. Whatever lessons have been learned have come at a high price: over one hundred thousand Afghans killed; thousands of American and NATO soldiers, contractors, and civilians’ lives lost; many tens of thousands more injured; and over a trillion dollars in U.S. treasure.

Underneath the strategic misrepresentations of the Afghanistan Papers, there lurks a genuine but deeply misplaced confidence in America’s ability to get counterinsurgency right.

The initial invasion of Afghanistan was popular, had clarity of purpose, and achieved reasonable success. In 2001 a few thousand U.S. troops, supported by air power and in partnership with the Northern Alliance, rapidly displaced the Taliban. And perhaps most importantly, the counterterrorism mission was largely achieved. While Mullah Omar and Osama bin Laden remained beyond our reach, al-Qaeda was decimated. Through 2007 the U.S. and NATO maintained a manageable presence—rising to roughly 20,000 troops in 2004—charged primarily with securing the capital and insuring that counterterrorism gains were not reversed.

But as the war dragged on, mission creep set in. Led first by neoconservatives and later by liberal interventionists, the Afghan conflict seemed to be reimagined on a nearly yearly basis. Each time the mission grew larger, more complex, and less winnable. Troop numbers swelled as we pivoted from counterterrorism to poppy eradication, counterinsurgency, and state building.

War planners, having learned the most superficial lessons of the Vietnam War, knew they needed to hold on to public support. And so a terrorist outfit and a backward gang of thugs were portrayed as an existential threat. Statistics were spun. Talking points informed minimally, if at all.

We could have succeeded in Afghanistan, experts suggest, if only we had been a little more competent, a little less distracted, a little more technocratic.

The unwarranted optimism may or may not have been the result of deceit. As Leslie Gelb and Richard Betts show in their postmortem analysis of the war in Vietnam, The Irony of Vietnam: The System Worked (2016), “Optimism is psychologically necessary for dedicated and energetic performance; analytical defeatism becomes operationally counterproductive.” Whether the result of lies or motivated reasoning, the outcome was the same. No one would admit what should have been a transparent truth: no goal beyond the initial counterterrorism mission was ever achievable in Afghanistan.

Dissembling has gotten most of the attention in discussion of the Afghanistan Papers. But if we dig a little deeper, we find something that should trouble us more. Underneath the strategic misrepresentations, there lurks a genuine but deeply misplaced confidence and optimism in America’s ability to get counterinsurgency right.


Within the documents, and in the reactions to them, we read experts lamenting failures of planning and implementation. But these laments are tinged with the same dangerous hubris that got us into our current situation in the first place. We could have succeeded in Afghanistan, they suggest, if only we had been a little more competent, a little less distracted, a little more technocratic. If we hadn’t overloaded the country with aid, surely we would have succeeded at building up local and national institutions. If only we hadn’t thoughtlessly prevented the poppy harvest, maybe the Taliban wouldn’t have reestablished a toe-hold. Perhaps if we’d focused on local, rather than national, politics, corruption would not have festered in the ranks of the elite.

No one would admit what should have been a transparent truth: no goal beyond the initial counterterrorism mission was ever achievable in Afghanistan.

“We went too fast, and that’s why we wound up with corruption. Next time we should provide more oversight and not rush to democracy,” argues senior International Security Assistance Force (ISAF) advisor Colonel Bob Crowley. “Securing the support of the neighboring powers and co-opting the competing patronage networks,” are the key according to Ambassador James Dobbins. If only we hadn’t “taken our eye off the ball” to focus on Iraq, argues James Mattis. The difference between failure and success, such arguments seem to suggest, was just a matter of making better decisions.

Such views are also to be found in the scholarly debate surrounding the war in Afghanistan.

While there is scant evidence that their decisions were much influenced by the research, the leaders of the war in Afghanistan afforded academic researchers unprecedented access to data and to the field of battle to study effective war fighting and state building. In the midst of the war, development economists, political scientists, and others fanned out across the country to field surveys and conduct randomized control trials on the effects of public service delivery, local governance institutions, mobile money, election administration, and a host of other state-building strategies. The military shared unprecedented incident-level conflict data with scholars, who used cutting-edge research designs to learn about how development assistance, communications technology, indiscriminate acts of violence, local elections, and many other events and interventions affected outcomes relevant for the counterinsurgency mission, from public opinion to insurgent violence.

All this data sharing and scholarly effort led to some intellectual contributions of real creativity and depth. Much was learned. But none of it strategic.

Politicians, war planners, and scholars alike appear to learn, and forget, the lessons of counterinsurgency with each new generation.

The scholarly approach to Afghanistan in large part matched the social scientific epistemology of the age. As reflected perhaps most clearly in this year’s Nobel prize for economics, awarded to developmental economists for taking an “experimental approach to alleviating global poverty,” we are in the midst of what social scientists call a credibility revolution. Greater attention to rigorous program evaluation and the challenges of teasing causation from correlation is meant to provide us with newly reliable evidence on policy efficacy.

The opportunity to rigorously evaluate counterinsurgency tactics thus led scholars down a path they already wanted to go down. Many were interested in asking tough empirical questions about whether the United States was implementing its counterinsurgency doctrine efficiently and how it could do better. But in so doing, the scholarly community, largely without noticing, adopted the assumptions underlying that doctrine. We asked hard tactical questions: are we doing the doctrine well? But few were able to step back to ask the bigger strategic question: does the doctrine itself make any sense?

To be sure, as Jason Lyall recently argued, there is an important sense in which scholars “have been sounding the alarm for years” about Afghanistan. But that alarm has largely been rung on the military’s terms. Lyall himself, for instance, points to some first-rate studies showing how hearts-and-minds could have been done better—the United States could have had a clearer understanding of the organizational structure of the Taliban, used aid more effectively by focusing on smaller, more local projects, or curtailed corruption more effectively.

We asked hard tactical questions: are we doing the doctrine well? But few asked the bigger strategic question: does the doctrine itself make any sense?

But this focus—by policymakers and scholars alike—on how the U.S. strategy in Afghanistan could have been implemented a little better misses the big picture. We’ve lost sight of a fundamental strategic fact of asymmetric warfare, hard-learned over the course of generations of failed Western counterinsurgencies, from Vietnam to Algeria to Cyprus. That fact is this: counterinsurgencies are wars of attrition. Democratic countries cannot win them in an adversary’s homeland when that adversary does not pose an existential threat.

We were always going to lose a counterinsurgency in Afghanistan.


Hearts and minds formed the foundation of the U.S. counterinsurgency strategy in Afghanistan. As General David Petraeus advised in his formal guidance to ISAF forces upon taking command in Afghanistan:

Secure and serve the population. . . . The people are the center of gravity. . . . Live among the people. We can’t commute to the fight. . . . Be a good guest. . . . Consult and build relationships.

This guidance was in keeping with Petraeus’s earlier approach in Iraq, informed by the counterinsurgency doctrine he laid out in the Counterinsurgency Field Manual. The theory starts with the assumption that, due to overwhelming capability, if the U.S. military knew where and with whom the threat lay, they could surely eliminate it. But they don’t. So asymmetric warfare is information-centric warfare. The struggle, therefore, is not over territory. It is over the hearts and minds of the people.

The thought was that a hearts-and-minds strategy of winning the war by winning the village provided a new, less bloody, more palatable path to counterinsurgency success.

From this flows the much-discussed philosophy of “clear, hold, and build.” Counterinsurgency, on this view, opens with security operations to remove the proximate insurgent threat. But it must move rapidly to the next step, where territory is held to create a sense of security among the population. “Prioritize population security over short-duration disruption,” Petraeus instructed. The final step is to build, using economic and political development to legitimize the counterinsurgents and delegitimize the insurgents. Fulfilling this goal requires minimizing harm to civilians, working with local partners, and building state capacity. In so doing, the theory goes, a successful counterinsurgency creates a virtuous cycle. If the counterinsurgents provide security and services, the tide of public opinion turns. The population comes to believe their long-run security interests lie with the counterinsurgents, rather than the insurgents. Pass this tipping point and the population will begin to provide information to and support for the counterinsurgents. The insurgents—fish swimming in the sea of the people—cannot survive this turn.

This theory of counterinsurgency was not new. It was, for instance, the approach advocated by the French in their failed efforts in Algeria. But it animated U.S. war planners because it constituted a rejection of the Vietnam-era strategy of attrition. No longer would U.S. counterinsurgents focus on relative body counts. Commanders on the ground had a new, practical, and evaluatable objective. They had to capture hearts and minds by clearing, building, and holding. The war would be won one village at a time.


Unfortunately, the Vietnam-era mistake was not conceptualizing the conflict as a war of attrition. Like every counterinsurgency, Vietnam was indeed a war of attrition. The mistake was believing that the war of attrition could be won.

Wars of attrition are not strategically complicated. They are won by the side with greater resolve. The greater a side’s perceived rewards to victory and the lower its costs of continuing the fight, the greater its resolve.

The idea underlying the Vietnam body-count strategy was to focus on the cost side of this equation. If, at each encounter, the United States could kill more enemy soldiers than it lost, the thinking went, eventually the mounting costs would wear down North Vietnamese resolve. This is why the United States was willing to amass horrendous death tolls to take seemingly strategically meaningless hills that it then abandoned in short order. The territory wasn’t the point in that war either. The death toll, or more accurately the relative death toll, was the point.

Without resolve equal to the Taliban’s, there was no path from winning the village to winning the war.

Vietnam exposed gaping flaws in this strategic calculus. To be sure, the United States imposed more terrible costs than it absorbed. But that didn’t matter once American domestic support for the war crumbled, while North Vietnamese support held firm. “The final outcome of wars,” wrote Fred Charles Ikle, a key American strategist in Vietnam, depends not on “self-contained contests of military power” but on, “a much wider range of factors, many of them highly elusive—such as the war’s impact on domestic politics.”

If the Afghanistan Papers are to be believed, U.S. policymakers and war planners in Afghanistan internalized this lesson and were much concerned with public opinion. This helps explain why the former White House Afghan war czar and NATO Ambassador, General Douglas Lute, worried, as things went south, what would happen to the war effort in Afghanistan “if the American people new the magnitude of this dysfunction.” Or why, as Colonel Crowley felt, “truth was rarely welcome.”

Of course, the strategy in Afghanistan was not to win the war of attrition through body-count ratios. The American public would certainly not have stood for that. But the thought was that a hearts-and-minds strategy of winning the war by winning the village provided a new, less bloody, more palatable path to counterinsurgency success. The key was to do it quickly enough that domestic support held. As General Petraeus put it, “this was a big time race against the clock.”

But, while the importance of domestic opinion is certainly one lesson of Vietnam, it is hardly the only one. The most fundamental problem for the counterinsurgent, as Andrew Mack explained almost half a century ago, has to do with the differential stakes of the conflict for the two sides: “For the insurgents the war is ‘total,’ while for the external power it is necessarily ‘limited’.”

That is, when all is said in done, for the insurgents, losing the war means losing everything. But, unless the insurgents pose an existential threat to the homeland, the same simply cannot be said for the counterinsurgents. And, try as American leaders might, within a few of years of the attacks of 9/11, selling al-Qaeda as an existential threat to the United States was even less tenable even than selling the domino theory during Vietnam.


This is the core problem for American counterinsurgency doctrine. The United States was never going to muster the resolve to win a war of attrition in the Afghanistan. Grand attempts were made—by the military, policy makers, development professionals, and academics—to understand how to win the village. But, without resolve equal to the Taliban’s, there was no path from winning the village to winning the war. No amount of optimizing the implementation of our counterinsurgency strategy—no technocratic tweak or marginally improved decision-making process—could reverse this foundational, strategic fact.

What could have been a successful counterterrorism mission became, instead, yet another failed counterinsurgency.

The closest U.S. counterinsurgency doctrine came to an answer to this challenge was a halfhearted nod toward the importance of establishing a robust Afghan state. A strong local partner, one might hope, would see the conflict in the same existential terms as the Taliban. And so, a high-capacity state with real skin in the game might reverse the logic of attrition. But this last-gasp defense is beset by its own strategic contradictions. Two stand out.

The first is a toxic combination of commitment problems and corruption. As has been well-documented, the leaders of Afghanistan’s central government profited hugely by misappropriating U.S. aid. The last thing these local “partners” wanted was for the United States to turn off the spigot of money and weapons with which they were enriching themselves and their patronage networks. But the surest way to keep the resources flowing was to avoid ever being strong or prepared enough to fight the Taliban on their own. No American leader wanted to own failure in Afghanistan. And so no Afghan leader had an incentive to succeed. Predictably, a strong partner with real resolve did not emerge.

There is, it should be said, some good news in Afghanistan. Childhood mortality continues a steady decline that has been ongoing since at least the early 1990s. And GDP has increased—from $4 billion in 2001 to $19 billion in 2018—though this should perhaps be weighed against the billions of dollars spent in the country. But, overwhelmingly, the evidence suggests that little progress has been made on the state building mission. The Afghan government has failed to establish authority, capacity, or legitimacy. Earlier this year, when the United States ceased providing assessments, the Afghan government only controlled in the neighborhood of 55 to 65 percent of the country’s territory (by population). The most recent national elections were again characterized by fraud, high levels of voter intimidation and low levels of voter turnout, and violence. And corruption remains endemic. None of this is lost on the Afghan people.

Even now, as we take stock nearly two decades in, we seem reluctant to embrace the fundamental, strategic facts.

The Asia Foundation’s Survey of the Afghan People, the longest running barometer of Afghan public opinion, paints a bleak picture. In 2006 twice as many Afghans felt the country was moving in the right direct as felt it was moving in the wrong direction. Today those numbers are almost flipped, with 58 percent believing it is moving in the wrong direction while only 36 percent believe it is moving in the right direction. In 2006, 40 percent of Afghans reported fearing for their personal safety, today 75 percent do. Fewer than half of Afghans say they believe the national police is honest and fair; 97 percent report that government corruption is a problem, while 63 percent report feeling fear while voting, up from 41 percent in 2006. In light of all this, it is perhaps not surprising that the United States appears to have come to the belated conclusion that the partnership strategy is a failure, seeking what looks like a separate peace with the Taliban that might well leave the Afghan government to fend for itself.

Perhaps one could argue that commitment problems and corruption were hard-to-predict, contingent facts about American and Afghan politics. I don’t think so, but perhaps. The second contradiction, however, was even more fundamental. The U.S. counterinsurgency strategy stood on two legs: winning hearts and minds and building a strong state that could stay the course in the war of attrition. But, as Paul Staniland argued a decade into the war in Afghanistan—now almost a decade ago—building strong states and winning hearts and minds are themselves contradictory goals:

Rather than a simple, apolitical technocratic exercise in administrative efficiency, state-building is characterized historically by relentless coercion, social homogenization, and center-periphery conflict. The imperatives of creating strong governments and of ‘winning hearts and minds’ can directly clash with one another.

You can’t build a strong state by winning hearts and minds. And without a strong state, there is no actor to reverse the logic of attrition. Not surprisingly, then, counterinsurgencies are rarely won. And when they are, they are won by governments that are prepared to engage in a level of brutality that the United States would never have been willing to entertain.

Politicians, war planners, and scholars alike appear to learn, and forget, the lessons of counterinsurgency with each new generation. This forgetting—aided by ideology on the part of the politicians, hubris on the part of the war planners, and epistemological blinders on the part of the scholars—has led to tragic consequences in Afghanistan. What could have been a successful counterterrorism mission became, instead, yet another failed counterinsurgency. Even now, as we take stock nearly two decades in, we seem reluctant to embrace the fundamental, strategic facts. Counterinsurgencies are wars of attrition. Wars of attrition are won through resolve. And the side facing an existential threat will always have the greater resolve.