The cigarette glowed red as he took a drag, and the smoke rose rapidly as he exhaled. It had been a long afternoon. It had been a long war.

It was February 2010, and after months establishing a relationship, Pakistan’s chief of army staff, Ashfaq Parvez Kayani, and one of us, Stan McChrystal, were having the kind of conversation senior military commanders are supposed to have, discussing the role of the NATO-led coalition’s efforts in Afghanistan and northwestern Pakistan. We’d spent hours alone, each laying out in detail a strategy for the conflict. While not quite my second home, the Pakistani army’s headquarters in Rawalpindi was now familiar ground, and Kayani, a colleague with whom I spoke easily. Nothing, however, could soften the blow of his message to me. “For the mission you’ve been given, you have the right strategy,” he told me. “But it won’t work, because you don’t have enough time.”

There was nothing revelatory in the general’s assessment, because like many others, I had already reluctantly concluded that it was likely correct. It may seem laughable that back in 2010, nine years after the war had begun and eight since I had first started serving there, we felt pressed for time. But for most of those years, the coalition’s efforts had been underresourced and poorly coordinated. And in December 2009, U.S. President Barack Obama had announced a commitment to begin reducing the United States’ role in 18 months. The clock was ticking. Still, the president had also decided to reinforce the U.S. effort so that it would comprise 150,000 U.S. and coalition forces and include an ongoing effort to train, equip, and advise 350,000 Afghan forces. If ever the United States had a realistic shot of success in its post-9/11 involvement in Afghanistan, it was then.

That was seven years of hope, effort, blood, and frustration ago. Today, anything that feels like success looks more distant than ever. The U.S.-backed government in Kabul remains plagued by political infighting and corruption, and the Afghan security forces cannot control significant parts of the country. The Taliban, while no longer the idealistic young fighters that swept north in 1994 and not particularly popular with the Afghan people, have leveraged Kabul’s weaknesses to make gains in recent years.

Against this backdrop, U.S. President Donald Trump has outlined a new strategy. As he detailed in a speech in August, the United States will continue its commitment in Afghanistan, modestly increase the number of troops to boost the capacity of the Afghan security forces, and redouble counterterrorist operations against the Islamic State, or ISIS, and other groups. It is largely more of the same.

The announcement represented a major reversal: as a candidate, Trump unequivocally declared his intention to end the U.S. military’s involvement in Afghanistan, but as president, he has pledged to extend it. In truth, however, there wasn’t much room for a different decision: withdrawing would risk turning the country back into the terrorist safe haven it was before 9/11, and drastically ramping up the U.S. presence would be a political nonstarter. That leaves something resembling the current approach as the only real option. Stuck with doing more of the same, Washington must try to do it better. 


The United States invaded Afghanistan in 2001 to destroy al Qaeda and overthrow the Taliban regime that was hosting it. The overarching goal was always to protect the United States by denying terrorists a safe haven in which to plan and train, but over time, the mission grew. Eventually, it came to include the establishment of an Afghan nation that defended its own sovereignty, embraced democracy, educated women, and cracked down on opium production.

Although the initial operations appeared to work, complexities on the ground, plus the distraction of the war in Iraq, sidetracked the effort, and the Taliban’s presence expanded. When Obama came into office, in 2009, he took a hard look at the Afghan campaign and announced a surge of U.S. troops and a reinvigorated counterinsurgency strategy. But by the middle of 2015, the troop surge was complete, and a subsequent drawdown left only 9,800 coalition troops in the country, most of whom were focused on training and advising the Afghans. Progress had been made, but it was limited.

Today, Afghanistan is struggling to survive. Although the Taliban have de facto control over only limited areas of the country, their presence and influence are likely at their highest levels since the group lost power in 2001. Remnants of the al Qaeda network and one of its branches, al Qaeda in the Indian Subcontinent, are also active, having been pushed out of Pakistan’s tribal areas in late 2014 by the Pakistani military. The Islamic State in Khorasan, as the branch of ISIS in Afghanistan and Pakistan is known, enjoys free rein on both sides of the two countries’ border. Although each of these groups has its own transnational agenda, all have made common cause with the Taliban to overthrow the Afghan government. 

The fragility of Afghanistan’s security sector is making their job easier. The 180,000 soldiers of the Afghan National Army, trained and equipped largely by the United States, are employed primarily at static checkpoints around the country that are vulnerable to Taliban attacks. The Afghan National Police, which is riddled with corruption and poor leadership, is used more for the protection of members of parliament and other officials than for its intended purpose of enforcing law and order. Afghanistan’s premier intelligence agency, the National Directorate of Security, is increasingly involved in military operations against terrorist groups instead of providing essential intelligence. 

Compounding the challenges, the Afghan legal system struggles to deal with corruption and criminality. Knowing little about the law and the rights of citizenship, Afghan security forces often make critical mistakes, for instance, detaining innocent civilians. By contrast, Taliban fighters—especially those in the lethal Haqqani network, an offshoot of the Taliban based in Pakistan—often have a thorough understanding of the law. When captured, they have proved adept at minimizing their sentences or avoiding conviction altogether. 

In Kabul, meanwhile, politics have reached a standstill. Despite its name, the National Unity Government—a power-sharing deal brokered by the United States in 2014 that made Ashraf Ghani president and Abdullah Abdullah chief executive—is deeply divided. 

Whatever progress the United States has made after 16 years, it is inarguably incomplete. To some Americans, the effort has succeeded in building a shaky foundation on which more can and should be constructed. To others, it represents a fruitless waste of blood and treasure. For the ordinary Afghan, however, the U.S. campaign has led to frightening uncertainty about the future.

A soldier during a night mission in the Pesh valley of Kunar Province, February 2015
Carlos Barria / Reuters


In 1902, Vladimir Lenin published a now famous pamphlet titled What Is to Be Done?, in which he prescribed a strategy for what later became the Bolsheviks’ successful takeover of Russia’s 1917 revolution. Lenin argued that Russia’s working classes required the leadership of dedicated cadres before they would become sufficiently politicized to demand change in tsarist Russia. It was a clear-eyed assessment of reality. The same is needed for Afghanistan now.

The United States has three basic options in Afghanistan: do less, continue on the current path, or do more. There is material for endless debates about the merits of each, but it helps to begin by remembering what the United States’ objectives in Afghanistan were and still are. As Obama said in his 2009 West Point speech about Afghanistan, “We must deny al Qaeda a safe haven. We must reverse the Taliban’s momentum and deny it the ability to overthrow the government. And we must strengthen the capacity of Afghanistan’s security forces and government, so that they can take lead responsibility for Afghanistan’s future.”

Whatever progress the United States has made after 16 years, it is inarguably incomplete.

If those objectives, or anything close to them, remain valid, it is hard to view doing less as an acceptable course of action. Although the government of Afghan President Mohammad Najibullah survived for three years after the Soviets withdrew, before falling to opposition forces, it took muscular logistical support and infighting among the opposition warlords to keep it in the fight for so long. Many observers believe that absent at least the current level of support, Ghani’s government could last only a small fraction of that time.

As for the doing-more option, why couldn’t the United States consider a version of the 2009 troop surge again? That strategy, while flawed due to ambitious timelines and the failure to execute a truly whole-of-government approach, could have succeeded had Washington demonstrated the necessary patience and commitment. But executing a counterinsurgency campaign over an extended period, always difficult for the American psyche, was a particularly tall order after the recent experience in Iraq. Today, gathering the popular and political support for a major increase in U.S. troop levels and a renewed commitment of many years is even more unlikely. Unless conditions on the ground changed drastically, it would be unrealistic to propose such a strategy. Besides, Afghans across the nation appreciate that a stepped-up U.S. presence would not be politically sustainable for long, thus increasing their concerns about what would happen after the Americans left.

That leaves the current approach as the only viable option. Under this strategy, Washington would have to lower its ambitions in Afghanistan, with the goal being merely a long-term relationship with and a limited military presence in a troubled but functioning country. As they shed some of the loftier goals of the past, policymakers will have to make it clear that the United States is unequivocally committed to its core goals. It would still promote regional stability, encourage modest but steady economic development, and maintain a platform from which to collect intelligence and carry out counterterrorism operations. Although this strategy would indeed come at a cost, its advantages—namely, ensuring the survival of a non-Taliban government—would be worth the price.

Critics may charge that following this course would meet the definition of insanity—which, as that old adage has it, is doing the same thing over and over again and expecting a different result. But as with everything in Afghanistan, the truth is more complicated. The United States has no better choice at hand, and in fact, this one is not all that bad. What’s more, within the confines of this strategy, there is room for improvement—in terms of fighting the Taliban in Afghanistan, dealing with them and their allies in Pakistan, and building a more responsible government in Kabul. 


Continuing to dismantle the Taliban in Afghanistan is easier said than done, of course, but it is probably essential to the survival of Afghanistan as a nation. No other opposition group in the country has been as successful in building a movement as the Taliban have. Portraying themselves as the more legitimate alternative to the current regime, the Taliban threaten the state and continue to offer sanctuary to ISIS and other transnational threats.

The United States should continue to squeeze the Taliban with a steady campaign of targeted strikes against their leadership, training camps, and other facilities. But Washington also needs to look outside Afghanistan and seek to increase international pressure on the group. Getting a UN resolution designating the Taliban as a global terrorist group would be a powerful move—it would severely undercut their legitimacy and reduce their access to external support—but an admittedly heavy lift. More likely to bear fruit would be the application of diplomatic pressure on countries offering support and sanctuary to the Taliban, especially the Arab Gulf states, where to this day, the Taliban freely collect donations and run businesses. 

A marine stands in a sandstorm in Afghanistan's Helmand province, October 2010.
A marine stands in a sandstorm in Afghanistan's Helmand Province, October 2010
Finbarr O'Reilly / Reuters

That would also mean putting pressure on Pakistan, of course, a tactic that has proved difficult and largely ineffective. Although the Pakistanis have taken action against some threats, the leaders of the Taliban, the Haqqani network, and other terrorist groups continue to operate relatively freely in major Pakistani cities, such as Peshawar, Quetta, and even the capital, Islamabad. It would be nice if it were possible to secure Afghanistan without reorienting the U.S. relationship with Pakistan, but experience proves that it is not. 

Disappointingly, pressuring Pakistan to take more effective actions to deny the Taliban sanctuary is not the silver bullet some hope for. Pressure could come in the form of reduced military assistance, but Washington’s leverage is relatively limited and could threaten U.S. supply lines that run through Pakistan, as well as add further friction to an already strained relationship. Still, wherever possible, pressure is appropriate. 

A political solution to the problem of the Taliban would be preferable, and it’s possible that renewed military pressure could drive the group to the negotiating table. But it would be a mistake to overestimate the Taliban’s sensitivity to such efforts. As long as the group believes there is any probability of success, even over a long time horizon, it is likely to stay in the fight, so a peace deal remains a distant prospect. It’s worth remembering that the efforts of Afghanistan’s High Peace Council, a body designed to negotiate a deal with the Taliban, came to a halt in 2011, when its leader, Burhanuddin Rabbani, was assassinated. (Rabbani was killed when someone claiming to be a member of the Taliban who wanted to discuss peace detonated a bomb hidden in his turban.) And the Taliban’s steady drumbeat of high-profile attacks in Afghanistan, resulting in scores of civilian deaths, makes negotiations nearly impossible in the current environment. The best the United States can do is to put unrelenting pressure on the Taliban while helping build the capacity of the Afghan state—so that the Afghans can eventually assume full responsibility for maintaining their sovereignty and preventing the reemergence of terrorist sanctuaries.


There is a common Afghan saying that roughly translates as “If water is muddied downstream, don’t waste your time filtering it; better to go upstream.” Likewise, no U.S. military campaign in Afghanistan can succeed if the enemy enjoys a safe haven in Pakistan. The United States must therefore refine and focus its operations there.

Filtering the water upstream, so to speak, has proved politically difficult across national borders. The U.S. military’s 1916–17 incursion into Mexico to hunt the guerilla leader Pancho Villa was famously controversial, as were its campaigns against North Vietnamese sanctuaries in Cambodia and Laos during the Vietnam War. British forces acting on behalf of Malaysia conducted cross-border operations in Indonesia in the 1960s; the Soviets threatened to attack mujahideen safe havens in Pakistan in the 1980s; and during the Iraq war, U.S. Special Forces reached into Syria in pursuit of al Qaeda in Iraq operatives. In each case, the complexities were huge.

No U.S. military campaign in Afghanistan can succeed if the enemy enjoys a safe haven in Pakistan.

Still, it would be a mistake to rule out U.S. operations in Pakistan. Like the mujahideen in the 1980s, the Taliban today are organized around three main hubs in the country—the province of Baluchistan, the Waziristan region of the Federally Administrated Tribal Areas, and the province of Khyber Pakhtunkhwa. From all three places, the Taliban launch attacks across the border into Afghanistan with impunity. Teams of Haqqani network operatives sent to conduct high-profile attacks have even managed to pass through the Torkham border crossing, in the famous Khyber Pass, using legitimate documents. Specially selected and trained inside Pakistan, they conduct meticulously planned and rehearsed lethal attacks against foreign embassies, Afghan government offices, and U.S. and NATO military installations.

A purely defensive strategy against these threats will never be sufficient; highly focused offensive operations, primarily in Afghanistan but, when necessary, also inside Pakistan, are required. To be sure, the United States has conducted such operations since the war in Afghanistan began, but it can do more. To maximize their effectiveness, the United States should assemble an integrated task force with Afghanistan that allows the two countries’ intelligence communities, law enforcement agencies, and militaries to collaborate. (Washington should also break down its own organizational silos that inhibit coordination and intelligence sharing when it comes to threats in Afghanistan.) In the best-case scenario, Pakistan would willingly participate in these joint efforts, but in the event that it does not show such unprecedented cooperation, they should go on regardless. 


The final element of the United States’ strategy in Afghanistan should involve convincing the Afghan government to press forward with reforms. Absent a concerted and effective campaign to reduce corruption and increase the effectiveness of key institutions, legitimacy with the Afghan people will remain elusive. Over the past 16 years, Washington has spent billions of dollars on training and equipping Afghan forces and building Afghanistan’s infrastructure, yet the country still has few properly functioning institutions. The handful of ones that do work owe their success to investments in developing leadership.

Driving reform in the Afghan government will require continuous coordination with the Afghans themselves, and many more than three cups of tea. Improving Afghanistan’s institutions will take the long-term work of building human capital and changing officials’ behavior, rather than short-term infrastructure or other projects. Accordingly, the United States needs to work closely with Afghanistan to select, train, mentor, and support the right caliber of leaders. Putting in place a “civilian surge” of large numbers of nonmilitary experts, as some have called for, is impractical. Creating and fielding such a group has proved difficult in the past, and the American public has little appetite for such an effort. But the United States could find purchase in supporting a smaller network of U.S. and international civilian advisers who would stay in Afghanistan for longer tours of duty. Driving change in any society is difficult, but Afghanistan’s complex environment is no place for well-intentioned neophytes or dilettantes. For the greatest probability of long-term success, the United States will need to create across multiple organizations a cadre of dedicated professionals who are steeped in the language, culture, and political realities of Afghanistan and who are connected by a coordinated strategy.


It’s tempting to view any further effort in Afghanistan as the ultimate example of stupidity or stubbornness. In the so-called graveyard of empires, failure may seem inevitable. But such pessimism ignores that a majority of Afghans oppose a Taliban regime and few would benefit from the Taliban’s return to power. Furthermore, the United States and its allies in post-9/11 Afghanistan have largely avoided being cast as colonialists. To be sure, Afghans have expressed their frustrations—from outrage over civilian casualties to disappointment about the lack of economic progress—but more of them wish for a better-executed effort than wish for abandonment.

Other skeptics may argue that even a limited effort could fail, and if it does, Washington could be forced into the hellish position of reluctantly increasing its commitment to an unworthy client state. The prospect brings to mind memories of the gradual, and ultimately unsuccessful, escalation in Vietnam. This is indeed a risk, but it is manageable, if Washington carefully identifies its objectives, and worth accepting in light of the alternatives. 

As satisfying as it might be to declare “game over” and move on, a post-American Afghanistan is not a pretty picture. Even though too great a Western presence in the Muslim world generates resentment, it is also true that a total absence reinforces the narrative that the United States doesn’t care about the non-Christian parts of the world. Without resurrecting the domino theory from the Cold War, one can still say that an American retreat from Afghanistan is unlikely to return the country to the tranquil place that served as the exotic setting for James Michener’s 1963 novel Caravans. More probable is a repressive and ideological regime that supports transnational terrorist groups. Among a range of unpalatable choices, the best option is to pursue some version of the current policy. The United States might as well do that as well as it can.

You are reading a free article.

Subscribe to Foreign Affairs to get unlimited access.

  • Paywall-free reading of new articles and a century of archives
  • Unlock access to iOS/Android apps to save editions for offline reading
  • Six issues a year in print, online, and audio editions
Subscribe Now
  • KOSH SADAT is a former Lieutenant Colonel in the Afghan Special Operations Forces. From 2009 to 2011, he served as Aide-de-Camp to the Commander of the International Security Assistance Force in Afghanistan.
  • STAN McCHRYSTAL is a retired U.S. Army General. He first served in Afghanistan in 2002 as Chief of Staff of Combined Joint Task Force 180, and from 2009 to 2010, he served as Commander of the International Security Assistance Force in Afghanistan.
  • More By Stanley McChrystal
  • More By Kosh Sadat