About this Issue

After years of delay, the United States is in the process of withdrawing troops from Afghanistan, thus ending the longest military engagement in U.S. history. Was it worth it? What should have been done differently? What lessons have we learned from the conflict? Is leaving a good idea or a bad one? Will we be back?

Foreign policy experts hold a range of views, and we have gathered several of them here to discuss the conflict and its long-awaited conclusion. Emma Ashford of the Atlantic Council’s Scowcroft Center for Strategy and Security writes the lead essay this month; she will be joined by Laurel Miller of the International Crisis Group, Barnett Rubin of the New York University Center on International Cooperation, and Hal Brands, the Henry A. Kissinger Distinguished Professor of Global Affairs at the Johns Hopkins School of Advanced International Studies.

Comments are open through the end of the month, and readers are invited to join the discussion.

Lead Essay

An End to Endless War?

In truth, there is little that one can say about Afghanistan that has not already been said in twenty years. In a nutshell: America’s longest war was lost long ago–perhaps as much as seventeen or eighteen years ago–though reality is only now catching up with policymakers. President Biden’s decision to largely honor the deal negotiated by former President Trump means that American troops will finally leave Afghanistan by September 11th, 2021, effectively bookending a two-decade war in the highest profile theatre of the global war on terror.

The costs have been high. According to Brown University’s Costs of War Project, the war in Afghanistan has cost a stunning $2.26 trillion. Those costs will not end when the United States withdraws; care for veterans and interest on the debt will likely add billions more in coming decades.[i] More importantly, the war cost the lives of at least 2,400 U.S. servicemembers and several thousand contractors. From the war’s earliest successes, through the Obama-era troop surge, to the slow decline of U.S. interest and presence in the country over the last decade, Afghanistan has represented a largely insoluble political problem. For twenty years, the United States and its partners struggled to impose western democratic institutions and liberal norms in a minimally developed and largely unconsolidated nation state, despite copious political science research suggesting it was likely impossible. Even the Obama-era troop surge proved incapable of turning the corner of an intractable conflict. Ultimately, it was the hackneyed tropes about Afghanistan as the “graveyard of empires” that ended up predicting the outcome correctly.

Of course, it is in many ways misleading to portray America’s withdrawal as a total defeat. Afghanistan’s civil war has lasted four decades; it started long before the American invasion, and it will likely last beyond America’s retreat. That America failed to end the war, or to reshape Afghan society, does not mean our intervention was an entirely pointless endeavor. As President Biden was at pains to point out in his withdrawal speech, U.S. troops actually succeeded in the more modest goals they went to Afghanistan to achieve in 2001, toppling the Taliban from power and breaking up the terrorist strongholds in that country. “War in Afghanistan was never meant to be a multi-generational undertaking,” Biden argued. “We were attacked. We went to war with clear goals. We achieved those objectives. Bin Laden is dead, and al Qaeda is degraded in Iraq — in Afghanistan. And it’s time to end the forever war.”

His last point, however, is more questionable. Afghanistan may not have started as a forever war, but it has long been the poster child for mission creep even in a well-justified conflict. It is also just the tip of the iceberg. Withdrawal has been rightly hailed as a step in the right direction, finally beginning the process of ending the global war on terror. And it is certainly a win for antiwar activists, proving that it is indeed possible to end one of America’s interminable counterterrorism deployments. Indeed, the withdrawal from Afghanistan puts the nail in the coffin of at least one part of the Bush Freedom Agenda, the idea that the world could be reshaped into a pro-American, democratic paradise by force. As Bush put it in his second inaugural address, U.S. policy was nothing less than “the growth of democratic movements and institutions in every nation and culture, with the ultimate goal of ending tyranny in our world.”[ii] Public opinion has increasingly turned against that notion. The Biden administration’s choice to withdraw all U.S. forces from Afghanistan is broadly popular; recent polls show that over half of Americans now support either downsizing or ending the U.S. presence in Afghanistan.[iii] A supermajority of the U.S. public–as well as veterans who fought in those wars–now say that the Iraq and Afghanistan wars were not worth fighting.[iv] Just twenty percent of Americans now consider promoting democracy abroad to be a top foreign policy priority.[v]

Perhaps more importantly, the Afghanistan withdrawal is a sign that Washington’s foreign policy community has soured on the notion of nation-building as a core tenet of foreign policy–or at least, on the notion that such things can be achieved through military means. As one report from the Center for American Progress put it, “The United States has numerous tools to vigorously defend its values and advance democracy without seeking to impose it using force.”[vi] Martin Indyk, a scholar at the Council on Foreign Relations, recently argued that “it is time for the U.S. to find a way to escape the costly, demoralizing cycle of crusades and retreats….never-ending wars and grandiose objectives” in the Middle East.[vii] Meanwhile, Jake Sullivan, now President Biden’s National Security Advisor, argued in 2019 that Iraq was “one of the most catastrophic decisions in American history,” resulting in “an open-ended military commitment that spans multiple countries.”[viii] In short, though regime-change fervor still exists in some quarters, Washington has, for the most part, thankfully abandoned its post-9/11 fervor for regime change and nation-building.

Yet in other ways, withdrawing from Afghanistan will not resolve America’s broader problem with endless wars. Though it may not have seemed it, Afghanistan was in reality one of the easier legacies of the war on terror to resolve. It was a clearly delineated conflict, in one specific country. There was widespread public awareness of the conflict (if not actual interest in the details), and public opinion had turned against it long ago. There has been a growing antiwar movement within the Democratic Party, and a sense among the Republican base–fueled by Donald Trump’s pronouncements–that nation-building was not in American interests. And perhaps most importantly, there was a strong sense of “mission creep.” Because our initial goals in Afghanistan were more circumscribed–the toppling of the Taliban rather than the end of terrorism, for example–it was possible to argue that the mission in Afghanistan today was not what had originally been intended.

Even with that, it took a decade or more for American leaders to overcome the political disincentives to withdrawal. No president, after all, wanted to be the man who lost Afghanistan. If not for Donald Trump’s personal idiosyncrasies–and his willingness to ignore criticism–it is possible the Afghanistan withdrawal would not have happened at all. They say that only Nixon could have gone to China; one might say only Trump could have opened negotiations with the Taliban. Biden is the beneficiary of that unusual circumstance. Afghanistan thus serves as a cautionary tale for those who seek to end the broader war on terror, which in some ways poses bigger operational, political, and legislative challenges. In 2019, there were U.S. troops in approximately forty percent of the world’s countries; some of these countries host bases; some have been targeted with drone strikes; and others have U.S. troops on the ground in combat, engaged in various forms of partner advising, or on train-and-equip missions.[ix] All of these will continue in the aftermath of an Afghan withdrawal.

Indeed, it is often challenging to ascertain where U.S. troops are actually fighting; the line between “training” missions and “combat” missions can shift in minutes under battlefield conditions. Consider the 2017 deaths of four American soldiers in Niger, which came as a surprise not only to the public, but to members of Congress, many of whom were unaware that troops might be fighting in that country–or even on that continent.[x] The soldiers were technically tasked to a partner support mission, which nonetheless brought them close enough to the frontlines for an ambush to turn deadly. And this was not an isolated case. The overlapping web of authorities used for foreign military deployments–Titles 10 and 50 of the U.S. Code (which authorize covert actions and espionage), Article 2 of the U.S. Constitution (which gives the president his Commander-in-Chief role over the U.S. military), various stand-alone Authorizations for the Use of Military Force (AUMFs) in specific conflicts–often differentiate military missions far less than one might think. And the nebulous nature of these commitments makes it challenging for Congress to perform oversight of America’s military campaigns. It also makes it harder to end those deployments. How can one end endless war, after all, if one doesn’t know where or how it’s being fought?

It short, it’s a situation well-suited to legislative inertia. There is a growing level of agreement within Congress–and in Washington more generally–that the 2001 AUMF, which authorized the original invasion of Afghanistan, needs to be repealed. It is clearly outdated; few sitting members of Congress voted for it. It is also excessively broad, having been used to justify not only the fight against al Qaeda, but also against offshoots like ISIS, and associated groups like Boko Haram. But there is little agreement on how to proceed after repeal. Some in congress want to repeal the AUMF entirely; many others want to repeal and replace it with newer authorities, whether broader or more circumscribed. Some want to insert sunset clauses that require regular congressional debate and re-authorization.[xi] This lack of agreement on what to do next means that basic repeal of even lapsed AUMFs like the 2002 Iraq War AUMF has so far been stymied in Congress. There are considerable doubts about whether any administration will consent to having its power over questions of war and peace constrained in this way.

The continuation of the war on terror thus remains largely a political problem. Many of the same dilemmas exposed by the debate over Afghanistan are in play in the broader debate about America’s post-9/11 wars. The notion of failed states as a breeding ground for extremism persists in the popular imagination, though much of the research on counterterrorism suggests that this is far less of a problem than commonly assumed. And though the overall level of terror threat is low, politicians remain highly risk averse. Indeed, the argument against withdrawal that persisted longest on Afghanistan was the notion that maintaining a small troop presence there could act as a relatively low-cost “insurance policy” against a potentially catastrophic future terrorist attack, reflecting the long-running notion that it’s better to fight terrorists over there than over here. It’s a relatively misleading argument, yet in the absence of the domestic pressure we saw on Afghanistan, this risk aversion may carry the day. The Afghanistan withdrawal represents the beginning of the end of the war on terror, but it is certainly not the inevitable end.

However, there is a glimmer of hope in Washington’s increasing obsession with so-called “Great Power Competition.” The concept originated in the Trump administration’s 2017 National Security Strategy, which argued that American power is being challenged by states like China and Russia in a competition that “requires the United States to rethink the policies of the past two decades.”[xii] Washington has wholeheartedly embraced the concept, as has the Biden administration; indeed, Biden’s first address to a joint session of congress was largely centered around the notion of competing with China. And though it has not immediately led to a drawdown in America’s war on terror conflicts, it is likely that a renewed focus on peer competitors like China and Russia might lead these counterterror missions to atrophy. It will not necessarily be a clean exit from the post-9/11 wars, which are likely to continue in some diminished form. Nor is Great Power Competition necessarily something to be welcomed. But as budgetary dollars shift toward great power competition, bureaucratic incentives to prioritize these counterterrorism missions will decline.

Ultimately, the withdrawal from Afghanistan should be seen less–as the president described it–as an end to the “forever war,” and more as the start of a long process of ending a twenty-year war on terror. Afghanistan itself represents a microcosm of America’s foreign policy during those two decades: from righteous struggle to overconfidence and overstretch, followed by disillusionment, and eventually, retreat. One can only hope that America’s next foreign policy crusade is less ambitious and more realistic.


[i] https://watson.brown.edu/costsofwar/files/cow/imce/figures/2021/Human%20and%20Budgetary%20Costs%20of%20Afghan%20War%2C%202001-2021.pdf

[ii] https://www.npr.org/templates/story/story.php?storyId=4460172

[iii] https://www.brookings.edu/blog/order-from-chaos/2021/03/19/americans-are-not-unanimously-war-weary-on-afghanistan/

[iv] https://www.pewresearch.org/fact-tank/2019/07/10/majorities-of-u-s-veterans-public-say-the-wars-in-iraq-and-afghanistan-were-not-worth-fighting/

[v] https://www.pewresearch.org/fact-tank/2021/03/02/americans-put-low-priority-on-promoting-democracy-abroad-2/

[vi] https://www.americanprogress.org/issues/security/reports/2018/09/05/457451/securing-democratic-world/

[vii] https://www.wsj.com/articles/the-middle-east-isnt-worth-it-anymore-11579277317

[viii] https://www.theatlantic.com/magazine/archive/2019/01/yes-america-can-still-lead-the-world/576427/

[ix] https://www.smithsonianmag.com/history/map-shows-places-world-where-us-military-operates-180970997/

[x] https://www.cnn.com/2017/10/23/politics/niger-troops-lawmakers/index.html

[xi] https://www.politico.com/news/2021/03/03/bipartisan-bill-strip-biden-war-powers-473312

[xii] https://trumpwhitehouse.archives.gov/wp-content/uploads/2017/12/NSS-Final-12-18-2017-0905.pdf

Response Essays

Balancing the Risks and Rewards of Withdrawal

There are plenty of good reasons—or defensible ones, at least—for the United States to withdraw from Afghanistan. The war isn’t going well for the Afghan government, and there is little hope of decisively turning the tide. The pull of competing priorities such as an ever-more dangerous rivalry with China is strong and getting stronger. The war often feels like the residue of an American preoccupation with the threats of a receding era rather than the threats of the emerging era. The human and financial costs of the war would probably go up if the Taliban began attacking American forces after a U.S. refusal to withdraw.

Yet most of these assertions, or variants of them, have been true for a long time. When Barack Obama announced his Afghan surge in 2009, he acknowledged that conditions in the country were grim: “The Taliban has gained momentum” and the Afghan government was losing ground. It has been clear, since the end of the George W. Bush years, that the primary focus of American national security policy should be the dangers posed not by Afghanistan but by the country next door—a rising and increasingly assertive China. The costs of the war were much higher in 2005, or 2010, or even 2015 than they are today.

So why is Washington undertaking a full military withdrawal from Afghanistan only now?

The easy answer is politics: Presidents Obama and Trump surely feared paying some electoral cost for admitting defeat in that war. But the obvious follow-up question is, “Why would there be a political cost associated with ending a long, frustrating war?” The reason was fear of the real security consequences that might follow.

For years, it has seemed unlikely that the Afghan government would survive an American withdrawal. As Carter Malkasian, America’s sharpest observer of the war (and my colleague at Johns Hopkins SAIS), has written, the upshot of retrenchment would probably be a Taliban takeover of most of the country—or, perhaps, a return to the all-out civil war of the 1990s. The Taliban, which maintains very close ties to al Qaeda, might not actively encourage terrorist attacks against the United States, but it might not do very much to prevent them, either. Or, if Afghanistan simply collapsed into a Syria or Libya-style civil war, extremists—even those the Taliban sees as rivals, such as ISIS—might fund opportunity in upheaval. As Malkasian argues, “The United States should recognize that the most direct route out of Afghanistan is to live with the threat of terrorism.”

Experts argue about how to measure the risk of major terrorist attacks following an American withdrawal, but that risk seemed sufficient that both Obama and Trump—two very different presidents who nonetheless converged in their evident desire to leave Afghanistan—ultimately chose not to.

They were able to make that decision because of a second factor: The costs of U.S. involvement in Afghanistan have fallen quite dramatically over the past decade. American fatalities in Afghanistan never exceeded 22 in the years between 2015 and 2019, compared to nearly 500 in 2010. American commanders believed that they could shore up the Afghan government and keep the pressure on terrorist groups with a presence of a few thousand troops (complemented by a similar allied presence), not the 100,000 that were present at the height of the American surge. In the same vein, the other major U.S. counter-terrorism intervention of recent years—the counter-ISIS campaign, centered in Iraq and Syria—cost a few billion dollars per year and took a mercifully small toll in American lives.

Lost amid all the talk of “forever wars” is that the United States had, finally, arrived at a sustainable counter-terrorism model featuring airpower, special operations forces, and other unique enablers deployed in support of local forces that made the vast majority of the sacrifices. That’s why the politics of the Afghanistan war were actually so permissive—the costs simply weren’t high enough for most Americans to care very much about whether the United States stayed or went.

Indeed, it is misleading to argue that the United States should withdraw from Afghanistan because the costs of the Global War on Terror have been too high. Of course, the costs of the GWOT have been too high. If American policymakers had foreseen, in late 2001, the price the country would have to pay in lives, dollars, and distraction for fighting two major wars in Iraq and Afghanistan, they might well have made different choices. (Then again, given the shock of 9/11 and the fact that the American homeland has mostly escaped significant Salafi Jihadist attacks in the twenty years since, they might not have.) But the costs of 2007 or 2011 are not the costs of 2021. The question is whether the United States should continue paying the far more modest price of sustaining its present CT model.

That question is difficult to answer, because it hinges on something not yet knowable—what the costs of withdrawal will be. In one relatively optimistic scenario (with the emphasis on relatively), Afghanistan becomes a humanitarian nightmare but the strategic costs to the United States are modest. The threat of international sanctions or even renewed U.S. military intervention prevents the Taliban from reverting to its worst old habits. The United States funds proxies, such as a renewed Northern Alliance, to prevent an entirely hostile Afghanistan. A layered approach to counter-terrorism—support for friendly Afghan forces, backed by the prospect of over-the-horizon military strikes, supported by all the diplomatic, intelligence, and financial measures the U.S. government uses to keep extremist groups off-balance—limits any resurgent threat.

Yet there are severe uncertainties about this scenario, involving the durability of the Afghan government, the feasibility of even knowing what to hit once the United States has lost situational awareness in a remote, landlocked country, the Taliban’s future intentions, and other issues. So there is also some real possibility of a much darker scenario, roughly comparable to what happened after the United States left Iraq in 2011. In this scenario, a chaotic or malignly governed Afghanistan generates instability and threats that require Washington either to live with an undesirable level of danger or re-intervene at a higher cost in lives, treasure, and strategic distraction.

This is why President Biden’s decision to withdraw is best characterized as a gamble. It is a gamble politically, in the sense that the politics of the Afghanistan issue could change if the result of withdrawal is a horrifying refugee crisis, systematic human rights violations, or a humiliating withdrawal of the American embassy under fire. And it is a gamble strategically, in the sense that Biden is running a real-time test of the proposition that America can reduce the military pressure it has exerted on the world’s most dangerous Salafi-Jihadist groups without allowing the threat they pose to return to intolerable levels.

If Biden’s wager pays off, he’ll be vindicated and critics of the “forever wars” will surely ask, not without justification, why America didn’t pull the plug sooner. If it doesn’t pay off, then America’s long stay in Afghanistan might not seem, in sad hindsight, like such an absurdity after all.

A Critique of U.S. Involvement in Afghanistan Focused on “Nation-Building” Won’t Help in Learning Lessons

Emma Ashford’s critique of America’s last 20 years of involvement in Afghanistan focuses on the purported error of embarking on “nation-building.” She casts the U.S. military engagement–at a cost of the lives of over 2,400 U.S. service members and over $2 trillion–as being primarily for this purpose, suggesting that the United States was trying to “impose” democracy, and that nation-building in this time period had been “a core tenet of foreign policy.” This is a misinterpretation of what the American war in Afghanistan was about.

The root of the misinterpretation lies in the characterization of the original goals of the U.S. invasion. She expresses these as the “more modest” goals (as compared to nation-building) of “toppling the Taliban from power and breaking up the terrorist strongholds.” The difficulty with this characterization is that regime change–which is what “toppling the Taliban” means–was a distinctly immodest goal. Ousting the regime was quick work for the U.S. and allied militaries, but the implication of ousting it was that something had to be put in place instead. It would have been–and would have widely been seen to be–the height of irresponsibility to accomplish only half a regime change, out with the old and in with … nothing?

The toppling was the easy part, the replacing not so much. That’s where the nation-building (or state-building, or any of a variety of less hot-button terms) comes in. It was a means, not a goal in and of itself, the Bush-era “Freedom Agenda” that Ashford cites notwithstanding. And it was a means that was directly tied to the primary goal of the invasion: counter-terrorism. Nation-building was seen as instrumental for avoiding a repeat of the Taliban-era coziness with and hosting of al Qaeda by supporting the installation of a new system of government that would be sympathetic with U.S. national security interests, embraced by most Afghans and therefore durable, and, ultimately, able itself to protect itself and the population.

U.S., European, and other supporting governments’ nation-building activities included constructing and equipping security forces, funding elections, providing technical assistance for creating state institutions, and delivering aid for expanding education and health care. On the civilian side, this was in many respects normal development assistance (though on steroids) for a poor country that donors had hoped to ease out of many years of conflict and harsh authoritarian rule. Once the Taliban regrouped as an insurgency and gained strength around 2005-2006, these kinds of activities also became part and parcel of counter-insurgency strategy.

The assistance that was provided for establishing–from scratch–and strengthening Afghanistan’s security forces was meant to be the U.S. and NATO military exit strategy. The idea was to enable the Afghan government to secure the country by itself and to become a long-term counter-terrorism partner in the region, and to enable the foreign military forces deployed in the country to be replaced by indigenous forces. This was the so-called “transition” strategy–which was implemented to a considerable extent, but not completed, by the original deadline, the end of 2014.

It has become commonplace to say, as Ashford suggests, that there was “mission creep,” because the scale of these activities, their costs, and the length of time spent grew beyond what most U.S. policymakers and the public expected in the early years. However, the expansion of the mission was not fundamentally because U.S. goals expanded over time; it was because one of the initial goals–eliminating Taliban rule in Afghanistan, and ensuring it remained eliminated–was much more difficult than originally anticipated. The problem was unrealistic expectations of what could be achieved, especially in a constrained timeframe. What is now seen by many as an expansion of war aims into nation-building terrain was actually the logical consequence of engaging in regime change in the first place, and then attempting to follow through in getting the change to stick.

Certainly, arguments can be made against the practice of nation-building, particularly at such large scale. A wealth of data supports criticism that U.S. security and civilian assistance to Afghanistan did not have its full intended impact and was too often poorly planned and implemented. No doubt there is plenty of room for tactical improvement. But to argue that the United States should have made a different strategic choice not to entangle itself in nation-building requires examining the alternative–which would have been choosing, at the start, to go after al Qaeda but not to change the regime in Afghanistan. Going in a different direction at that fork in the road would have produced a very different history of the last 20 years.

Ashford correctly points out that risk aversion on the part of U.S. leaders played a strong role in hesitance to withdraw U.S. forces even as conditions in Afghanistan deteriorated, and even the continued presence of terrorist groups there came to be seen as less important to U.S. security over time. She inaccurately credits Donald Trump, however, for breaking the spell. Trump was not, in fact, the president who “opened negotiations with the Taliban”; that started during the Obama Administration, with tentative steps first taken in 2009 and then intermittent efforts through to the end of that presidency. The Obama-era initiative was unsuccessful, however.

After Trump initially approved a strategy early in his presidency that ramped up the war effort and that put off peace-making, by the end of 2018, in a bid to find a way out of Afghanistan through negotiation, his administration put talks with the Taliban at the center of its strategy. This resulted in a February 2020 U.S.-Taliban deal that called for a complete withdrawal of U.S. and NATO forces (by 1 May 2021), in exchange for Taliban promises not to allow terrorist groups to use Afghan territory for activities that could threaten U.S. security.

Ashford rightly observes that President Biden benefitted from Trump having done this deal, given Biden’s own longstanding inclination to reduce U.S. military involvement in Afghanistan. Still, the choice was Biden’s and–if he had been as risk-averse about withdrawal on his watch as his predecessors, including Trump–he could have set aside the deal, arguing that the Taliban have not fully complied with their end of the bargain.

Ashford regards Biden’s decision to withdraw from Afghanistan as “the start of a long process of ending a twenty-year war on terror.” The decision might turn out to be part of a broader reconsideration of the role and nature of counter-terrorism operations overseas. But a full appraisal of what the United States did over 20 years in Afghanistan would need to extend beyond counter-terrorism and avoid being too U.S.-centric. The invasion was not just a sideshow in a “civil war [that] has lasted for four decades … and [that] will likely last beyond America’s retreat.” That characterization brushes past the ways that the intervention–undertaken for the United States’ own security purposes–greatly intensified conflict in Afghanistan, leading to far greater loss of life among Afghans than the losses suffered by Americans.

It should not be forgotten that the U.S. exit from Afghanistan in no respect will “end” an “endless war.” Whatever one’s view on the merits of exiting now, there should be room to spare a thought for the costs that the U.S. invasion visited on many Afghans, and the violence ahead that they will suffer in the intensifying war the United States will leave behind. And the United States should spare no effort in trying to mitigate that suffering.

In that regard, steps the U.S. and other governments should take as the withdrawal approaches and afterward include following through on assurances that civilian and security assistance to Afghanistan will continue, and ensuring mechanisms are in place for delivery of humanitarian assistance throughout the country as the conflict probably worsens in the months ahead. In addition, the United States should provide strong political and practical support to establishment of a new UN leading role in peace diplomacy, given the diminishing U.S. role in Afghanistan.

Logic (though perhaps not politics) dictates that once there are no more American boots on the ground, the United States should do more, not less, to sustain its diplomacy and development aid in support of the Afghan government and any possibility of peace talks. Doing so would make good on the claim that military withdrawal does not mean the United States is abandoning Afghanistan and would show seriousness about the idea that diplomacy and aid can be powerful tools of U.S. foreign policy.

Endless War and the De-Politicization of National Security

Why this sudden bewilderment, this confusion?

(How serious people’s faces have become.)

Why are the streets and squares emptying so rapidly,

everyone going home lost in thought?

Because night has fallen and the barbarians haven’t come.

And some of our men just in from the border say

there are no barbarians any longer.

Now what’s going to happen to us without barbarians?

Those people were a kind of solution.

“Waiting for the Barbarians,” Constantine. P. Cavafy (1904), trans. Edmund Keeley

In Afghanistan, Emma Ashford writes, “For twenty years, the United States and its partners struggled to impose western democratic institutions and liberal norms in a minimally developed and largely unconsolidated nation state.” This is a popular reading of the decision to withdraw U.S. forces from Afghanistan: it both flatters our past good intentions and lends to the prospect of pursuing self-interest the cover of wisdom and virtue. Given the failure of supposed altruism, Ashford finds “a glimmer of hope in Washington’s increasing obsession with so-called “Great Power Competition,” though she oddly attributes this venerable concept not to Thucydides, Kautilya, or Sun Tzu but to the Trump administration’s 2017 National Security Strategy.

Had the United States been interested in a struggle to “impose democratic institutions and liberal norms” on Afghanistan, it could have started the effort much earlier. It could have refrained from helping Pakistan and Saudi Arabia to empower Gulbuddin Hekmatyar and Abdul Rab Rasul Sayyaf during the 1980s, and it might have responded to the Taliban after 9/11 by means other than toppling their regime by empowering warlords whose misgovernance and violent clashes had initially made the Taliban seem like a welcome relief. The United States could have provided some support to such efforts during the twelve years that intervened between the 1989 Soviet withdrawal and the 2001 al Qaeda attack on the United States, rather than relegating the area to the peripheries of global strategy.

Ashford recognizes that “our initial goals in Afghanistan” had nothing to do with “democratic institutions and liberal norms.” They were “the toppling of the Taliban.” She even asserts that the initial goals did not include “the end of terrorism,” and that the shift to the strategy called “nation building” resulted from mission creep. The rapid shift of resources from Afghanistan to Iraq, however, shows that for the Bush administration, the intervention was from the very beginning the opening salvo of the “War on Terrorism” aiming at the transformation of the Muslim world in the U.S. strategic interest. Afghanistan itself remained at the margins, however. The “war on terrorism,” as the primary strategic goal, always took precedence over democratization and nation-building, which were tactics inconsistently adopted in service of the counter-terrorist goal.

There are many valid reasons that a country as impoverished and with such weak state institutions as Afghanistan would find it difficult if not impossible to establish a stable democracy, but the insurgency did not derive from opposition to liberal institutions. It derived from illiberal counter-terrorism measures, including both political exclusion and violent abuses, and a failure of regional diplomacy, especially with Pakistan.

I have chronicled elsewhere how counter-terrorist priorities, defined “kinetically” (as capturing or killing “terrorists”) always took precedence in U.S. policy in Afghanistan over either democratization or peacemaking. The Taliban did not refuse to join the new “liberal” order. They were told from the beginning, on television, by U.S. Secretary of Defense Donald Rumsfeld, that they were not welcome. On December 6, 2001, the day after the signing of the Bonn Agreement, Rumsfeld publicly rejected a truce negotiated by Afghan President Hamid Karzai, which would have permitted the Taliban leadership to participate in the political processes envisioned by the Agreement. In 2002, U.S. Vice President Dick Cheney ordered a halt to efforts by the CIA to help former Taliban officials organize to participate in the new order.[1] In 2004, when the Taliban sent a delegation to Kabul to negotiate with Karzai, the United States informed the Afghan government that it would not guarantee the security of Taliban officials it considered to be linked to al Qaeda, ending the entire effort. As a result of this incident, the United States codified its position on negotiations with the Taliban. It compiled a list of members of the Taliban leadership whom it considered as “linked” to al Qaeda and told Karzai that he could not engage any of them without explicit agreement from the United States.[2]

This policy was not supported by any analysis or intelligence estimate of its effect on the political future of Afghanistan. Counter-terrorism policy as a construct systematically elided the political. The Taliban and al Qaeda, two completely distinct organizations, with different histories, ideologies, and memberships, were amalgamated along with many others into the concept of “Islamic extremism,” as if their beliefs and actions resulted from a common distinct interpretation of a religion, rather than from the political and social processes that produce other political actors.

Such an analysis is an extreme instance of what social psychologists call “the fundamental attribution error,” which results from “a cognitive bias to assume that a person’s actions depend on what ‘kind’ of person that person is rather than on the social and environmental forces that influence the person.” Since the “social and environmental forces” that influence al Qaeda and the Taliban included U.S. policies in the middle east and South Asia, including Afghanistan, the counter-terrorist paradigm directs attention away from the United States and its interactions with the rest of the world and instead focuses on the characteristics of other actors that make them resistant to our assumed good intentions.

It is no wonder that such a cognitive framework leads to bad policy. Over two thousand five hundred years ago, Sun Tzu taught “Know yourself, and know your enemy.” That motto includes understanding the interactions that have made the two sides what they are today and made them into enemies. Understanding those interactions can provide insights on how to transform the relationship without fighting, or how to make whatever fighting is needed more effective. Ignoring one’s own role in the origin of the conflict deprives one of the ability to “subdue the enemy without fighting,” which Sun Tzu identified as “acme of skill” in warfare.

Besides ignoring one’s own role, analyzing conflict in terms of the character of the enemy also inhibits political analysis of the “social and environmental forces” in the enemy’s own political environment. U.S. policy toward Afghanistan was dominated by the designation of the Taliban as an enemy to equated with al Qaeda–as President Bush said, those who harbor terrorists will meet the same fate. That rendered understanding the actual political processes of Afghanistan and its neighbors nearly irrelevant. It defined the problem as how to defeat the enemy, rather than as how to achieve a defined political objective. In the course of this incoherent project, the U.S. government seized on many rationales–democracy and liberal norms were just one of many. The lack of self-knowledge and indifference toward the contexts that drive others would have guaranteed failure regardless of how success was defined.


[1]Steve Coll [2018], Directorate S.: The C.I.A. and America’s Secret Wars in Afghanistan and Pakistan. Penguin Press, 140.

[2] Van Linschoten, Alex Strick, and Felix Kuehn. 2012. An Enemy We Created: The Myth of the Taliban-Al Qaeda Merger in Afghanistan. Oxford University Press,

The Conversation

On Afghanistan, the Retrospective Approach Is the Right One

I greatly appreciate the detailed responses to my initial essay from Hal Brands, Barnett Rubin, and Laurel Miller. I have found their responses enlightening and challenging to some of my own assertions and understandings of America’s longest war. At the same time, I found some of the responses frustrating, not because of the authors’ excellent points, but because twenty years after the United States invaded Afghanistan, we are in many ways still having similar arguments. A couple of common themes and disagreements emerge from the responses; rather than respond individually, it seems more useful to highlight the general distinctions between the essays, which still represent some of the key disagreements within the policy community on Afghanistan policy.

The first point requires a mea culpa on my part. In focusing my essay on the implications of the war and today’s withdrawal for U.S. policy, I neglected to mention the costs of the war to those who have suffered the most. The Afghan people have suffered mightily not just over the last two decades, but ever since the Soviet invasion of that country in 1979. As Miller puts it, “Whatever one’s view on the merits of exiting now, there should be room to spare a thought for the costs that the U.S. invasion visited on many Afghans.” She is right; the total cost of the war is not only the several thousand U.S. servicemembers and contractors who lost their lives, but an estimated 240,000 people across both Afghanistan and Pakistan. In the context of the U.S. withdrawal, policymakers need to not only include some acknowledgement of these costs—which I failed to do in my initial essay—but also to reckon with the question of how we can assist the Afghan people going forward, with aid, diplomatic assistance, and immigration benefits for those who need them. Miller argues that “the United States should spare no effort in trying to mitigate that suffering.” I wholeheartedly agree.

This suggests a further point of contention between respondents. Using an Afghan-centric lens vs. a U.S.-security perspective calls into doubt many critics’ assertions that Afghanistan, far from being a theatre of the “forever war,” was a sustainable, low-cost intervention that the United States could maintain for decades. As Hal Brands describes it, however, it becomes clear that such a model is low-cost only for the United States: he argues that Afghanistan is a “sustainable counter-terrorism model featuring airpower, special operations forces, and other unique enablers deployed in support of local forces that made the vast majority of the sacrifices.” Whether this is truly sustainable in the context of Afghan casualties is thus questionable. As Barnett Rubin points out in his response essay, nation-building in Afghanistan partly failed because the United States remained focused on counterterrorism at the cost of more inclusive political or regional solutions. As he describes, “the insurgency did not derive from opposition to liberal institutions. It derived from illiberal counter-terrorism measures, including both political exclusion and violent abuses.” The less visible costs of counterterrorism—at least from Washington—undoubtedly helped to undermine the “sustainable” U.S. military presence that Brands and some other analysts have proposed for Afghanistan—and vice versa.

A third issue that arises out of the debate is the origins of U.S. strategy in Afghanistan. As both Miller and Rubin point out, I perhaps oversimplified American motives, particularly in implying that one could separate the “good war” against the Taliban and al Qaeda from the nation-building fiasco that followed it. Indeed, as Miller points out, an intrinsic part of regime change in Afghanistan—of ousting the al Qaeda-supporting Taliban and preventing their return—was backfilling the regime in Kabul with something more acceptable to the United States. She thus argues that the two missions were inseparable, rather than a case of mission creep. But while it would have been both un-strategic and irrational to simply topple the Taliban in 2001 and walk away, there were also middle options between doing nothing and the full-throated attempt to rewrite the entirety of Afghan society and politics that the Bush administration chose. The United States could simply have installed a friendly autocrat in Kabul, as unlikely as that might have seemed given the Bush administration’s strong support for the war on terror and the “transformation of the Muslim world.” Alternately, the United States could have pursued a more broad-based coalitional approach to Afghan politics. As Rubin points out in his rebuttal, the United States probably undermined its own efforts in Afghanistan, partly through the prioritization of assertive counterterrorism, but also through what he describes as a fundamental attribution error that identified the Taliban as entirely aligned with al Qaeda and refused to allow them to participate in the political process at any point. One could easily argue that the decision to pursue broad-based democratization with liberal characteristics—and to avoid a narrower power-sharing approach to Afghan politics—was a choice, not a necessity forced upon the United States.

Finally, these excellent articles make clear that twenty years after the Afghan invasion, elites are still deeply conflicted about the conflict and its necessity for U.S. foreign policy. This contrasts rather sharply with public opinion, where a sizable majority of Americans now support withdrawal, and less than a quarter now view the war as successful. Yet every essay here also framed the U.S. campaign in Afghanistan as a retrospective, rather than an argument for continued presence. I hope that Joe Biden’s decision to finally withdraw troops from Afghanistan will thus also serve to improve the quality of Washington’s debates on the war, as we finally start to consider the conflict in retrospect, rather than an ongoing and contentious military engagement. As these essays show, a retrospective approach may ultimately help us to better understand why U.S. policymakers made the choices they did, and why we failed so badly in achieving our goals.

Our Real Lessons May Still Await Us

Emma Ashford has provided a thoughtful reply to the response essays written by me, Barnett Rubin, and Hal Brands. I agree with her that there is a lot of work to do in examining retrospectively what the United States did in Afghanistan over the last twenty years, why it did those things, and why the results were so far below the expressed expectations.

Now, with completion of the U.S. and NATO withdrawal only weeks away, I have more questions than answers about the retrospective picture. The more fundamental the questions, the harder they are to answer.

One of those fundamental questions is whether Afghanistan and the United States would have been better off if the United States had never invaded in 2001—if it had chased al Qaeda, heedless of Afghan sovereignty, but had left the Taliban regime in place. That would have avoided the nation-building entanglement, the insurgency and counter-insurgency, harms done to Afghan civilians, the costs of military deployment, and any U.S. sense of responsibility for fixing anything in the country. But what exactly would this alternative history look like? Would the Taliban regime’s abuse of the population and (to put it mildly) poor governance have been just another chronic problem in the world, with no further acute manifestations like hosting the leadership of an international terrorist group? As much as many Afghans lament American errors during the last 20 years, would most of them have preferred no toppling of the Taliban?

Another of those questions is whether it is really conceivable that the United States could have fought the war differently in ways that really would have made a difference. Many analysts have pointed to ways in which U.S. military tactics in Afghanistan helped fuel the conflict (for instance, night raids on Afghan homes in rural areas and targeting mistakes, which stoked antipathy toward the U.S. and Afghan governments). Some of the points Rubin makes in his excellent response essay point to how the U.S. approach to counter-terrorism did this. But what exactly would a better fought war have looked like, and is it really imaginable that the U.S. military could have fought the war in any radically different way than it did? If there is a plausible alternative history in this regard that takes account of how the U.S. military works, in which better tactics would likely have produced a different outcome, it needs to be spelled out.

Rubin points to an alternative political history of the war, in which the United States recognizes early on that reconciliation with the Taliban could either prevent an insurgency or nip it in the bud. That is indeed the greatest error to lament. More will need to be written about what it is about American politics and bureaucratic behavior that prevented clear perception of the political dimensions of the war.

There’s a genuine risk that within the U.S. government any efforts to draw “lessons learned” from Afghanistan will focus on the ways and means employed and give short shrift to the big strategic questions and the fundamental choices that were made. Others will probably need to provide the real retrospection.