A U.S. soldier points his rifle at a doorway after coming under fire from the Taliban while on patrol in Zharay district in Kandahar province, southern Afghanistan, April 26, 2012.
Baz Ratner / Reuters

There is no end in sight to the U.S. war in Afghanistan, which, as it enters its 16th year, is the longest conflict in U.S. history. Americans have increasingly questioned the war’s value, and before he became president, Donald Trump was among them, criticizing the country’s continued involvement in Afghanistan. In 2013, he tweeted, “Let’s get out!,” citing the “waste of blood and treasure.” During the election campaign, he routinely promised not to involve the United States in wars that it could not win. But as president, Trump has reversed course, following instead the path of his predecessors, Presidents George W. Bush and Barack Obama.

In August, Trump approved the deployment of several thousand more troops to Afghanistan, adding to the 8,500 already stationed in-country. “With our resolve, we will ensure that your service and that your family’s will bring about the defeat of our enemies and the arrival of peace,” he said. “We will push onward to victory with power in our hearts, courage in our souls, and everlasting pride in each and every one of you.”

There is, of course, no assurance of peace, at least not of the sort Trump has envisioned. Victory is simply not in the cards. The reality is that the United States is best off keeping a residual force in Afghanistan—but not escalating, as Trump plans to do. The small, on-the-ground force should focus on counterterrorism and on training the Afghan military. Such efforts will not stop the Taliban from gaining de facto control over much of the country, but they will manage to keep terrorist groups weak and help the Afghan government avoid total defeat.


In the debate about U.S. involvement in Afghanistan, those who are pro-involvement tend to prioritize the future risk of not doing anything while those against involvement focus on the considerable cost of past U.S. engagement and the seeming futility of the situation. Fear of terrorism is the primary driver keeping the United States in Afghanistan. Because the Taliban hosted al Qaeda in the years before 9/11, the concern is that they would do it again (or with other terrorist organizations), should they gain power in all or part of the country. As Trump warned, “A hasty withdrawal would create a vacuum for terrorists, including ISIS [the Islamic State] and al Qaeda, would instantly fill just as happened before Sept. 11.”

The U.S. Defense Department counts as many as 20 active terrorist groups in Afghanistan. Moreover, the rise of the Haqqani network within the Taliban (their number two is now Sirajuddin Haqqani) is troubling given the network’s ties to international terrorist groups. In addition, the Islamic State in Khorasan Province has emerged in recent years and has executed brutal terrorist attacks. ISKP is a rival to the Taliban, and it could possibly conduct more attacks if it had greater freedom of operation in Afghanistan.

Although the Afghan government controls only part of the country, its presence and the U.S. drone program impede the Taliban’s ability to host the large-scale trainings that enabled al Qaeda and other groups to flourish in the pre-9/11 era. Government forces can raid Taliban-controlled areas and disrupt camps, and the United States can send special operations forces on raids or otherwise strike any facilities with relative ease. Setting aside the issue of Taliban or ISKP control, the United States also uses Afghanistan as a base for drone attacks in Pakistan. Much of the drone program depends on human assets for reporting—it cannot be done entirely “over the horizon,” with no allies on the ground. 

Another concern is that a total U.S. departure, which would essentially signal an open defeat, would boost the morale of the jihadists. Afghanistan is where they beat the Soviet Union during the war in 1979–89, and a victory over yet another superpower would provide them with a tremendous psychological boost; this could, in turn, help them enhance their recruiting and fundraising efforts. U.S. allies would be equally dispirited, seeing the United States abandon the region despite years of sacrifice. Finally, a U.S. withdrawal would embolden Pakistan’s generals, convincing them (correctly) that covertly supporting militants against their enemies is a winning strategy and one they should continue to use against India.

The is also a humanitarian case for remaining in Afghanistan. Since the fall of the Taliban in 2001, elementary and secondary education enrollment has increased steadily throughout the country. Life expectancy rose from 56 years in 2004 to 60 years in 2010. This progress could be reversed if the Taliban comes back to power. The group opposes women’s rights, religious tolerance, education for girls, and general liberal democratic values. 

Against all these apprehensions, however, we must weigh the sorry record of the past 16 years. The United States has lost more than 2,400 soldiers and Marines—and spent a trillion dollars—during its intervention in Afghanistan. U.S. allies lost more than 1,000 men and women. Afghan casualties dwarf these: in 2016, almost 7,000 Afghan soldiers and police died. Since 2009, more than 25,000 Afghan civilians have died, and this year may mark the highest annual death toll yet, with over 1,500 dying through the end of June 2017. Not surprisingly, Afghans feel insecure, which often leads them to work with the Taliban to survive. In 2006, 40 percent of Afghans feared for their or their families’ safety; in 2016, this figure reached 70 percent.

Before the United States intervened, Afghanistan suffered a civil war for the better part of 25 years. Clearly something is deeply wrong within the country that the United States alone cannot resolve. Washington can postpone defeat, perhaps, or at least slow the Taliban’s momentum, but history suggests that a complete victory is unlikely.

Unsurprisingly, the American public feels similarly. Polls suggest that more Americans believe that the United States is losing rather than winning in Afghanistan and favor reducing U.S. involvement. The fact that so many respondents reported that they “don’t know” suggests that most Americans simply do not follow what is going on with their country’s longest war.

The human cost would be less painful if the United States and the Afghan government were on a path to victory, but they aren’t. Instead, the Taliban are ascendant. Although the group has lost many fighters, it can recruit new ones and has perhaps 30,000 men under arms. At the end of 2016, the Afghan government controlled only 57 percent of its territory, a dramatic downturn since 2015. (On the bright side, however, the Afghan government controls or influences almost two-thirds of the population because it dominates Afghan cities.) The Taliban mostly control remote areas, but they and groups like the ISKP can strike in Kabul and other government-controlled areas and terrorize population centers across the country.

An Afghan National Army soldier is seen through damaged glass as he keeps watch at the Forward Base in Nari district near the army outpost in Kunar province, February 24, 2014.
Omar Sobhani / Reuters

Efforts to defeat the Taliban entirely have failed in part because many Afghans see their government as inept, corrupt, brutal, or all three. The government’s democratic credentials are superficial: its leadership is composed of elites, bargaining to divvy up power and patronage, rather than of figures who enjoy mass support. No ideology, political party, or charismatic leader unifies Afghans. By design and necessity, local officials hold much of the power. As such, Afghanistan never truly unified under a strong government. Rather, the country’s mountainous geography—and the strong ethnic and tribal identities in many areas—hinder national unity; when Afghans unite, it is often against foreigners. Press freedom is low, as reporters face many dangers. But graft is rampant. Transparency International ranks Afghanistan as one of the world’s most corrupt countries, and roughly two-thirds of Afghans believe that the country is going in the wrong direction. Not surprisingly, then, the government in Kabul does not engender loyalty, regardless of political leadership. Finally, the government’s failures beget further failures. Its inability to rein in corruption, establish the rule of law, provide security, or otherwise perform basic governance functions leads Afghans to turn to local rulers, militias, and the Taliban, further undermining the government’s influence.

In addition, U.S. and allied forces failed to train their Afghan counterparts in large enough numbers to fight the Taliban successfully. Afghanistan’s elite special forces fight well, and plans exist to expand them. However, they represent only 17,000 of the country’s 300,000 military personnel. The Afghan military and police are taking increasingly heavy casualties, losing over 30 people per day. In 2015, the Afghan military casualty figure was more than twice that of 2011. Most of these fighters have little loyalty to their officers and political leaders and evade risking their life to fight the Taliban, and so many defect.


A political deal in Afghanistan now seems further away than ever. In 2008, Chairman of the U.S. Joint Chiefs of Staff Admiral Mike Mullen declared, “We can’t kill our way to victory.” And yet negotiations have made little progress, even during the height of U.S. troop presence in Afghanistan. The Taliban’s eagerness to kill through high-profile terrorist attacks continues to pose a threat to peacemakers. As the group grows stronger—albeit fitfully—it is unlikely to be more conciliatory than when it had a weaker hand. Groups like the ISKP, which are more radical, oppose any negotiation and are likely to put political pressure on the Taliban to keep them from making any concessions. In addition, the Taliban and their Pakistani backers recognize the American public’s growing weariness with the war in Afghanistan. If they believe, perhaps correctly, that time is on their side, they will only be less willing to make a deal. 

Pakistan plays a tremendously important and negative role in the Taliban’s rise to power. It helped the Taliban expand in the 1990s, providing massive financial and military assistance. Thereafter, Pakistani intelligence acted as the Taliban’s “primary patron,” according to the 9/11 Commission. Despite promises to the contrary, Islamabad’s support continued. Since 9/11, Pakistan succored the Taliban, offering a haven from U.S. counterterrorism operations and assistance with fundraising and logistics. This prevented the United States from engaging in the leadership targeting that proved successful in devastating the ranks of al Qaeda in Iraq at the end of the last decade. Instead, the Taliban’s cadre have a place to train, regroup, and plan—whether to lay low and lick their wounds or step up operations when there is an opportunity. 

The Trump administration is trying to shake things up by conditioning some U.S. aid to Pakistan on ending support for the Taliban, but despite Trump’s desire to force Islamabad to shape up, Pakistan is unlikely to change. Islamabad has a strong strategic interest in maintaining a friendly government in Kabul as Afghanistan has long proved a troublesome neighbor. In addition, Pakistan’s military and security services contain anti-Western and Islamist elements that favor ignoring Trump’s demands. On a more local level, the areas near the Pakistani and Afghan border, which are ruled less by the central government than by strong cross-border tribal ties, allow terrorist organizations and their allies to operate outside of the Pakistani government’s influence, regardless of Islamabad’s intent. Ending the Pakistani sanctuary completely would require a massive change in policy in Islamabad, as well as U.S. military operations against groups holed up there. 

In addition to its problems with Pakistan, Washington also has a strained relationship with several of Afghanistan’s other neighbors, notably Iran and Russia. For both of these countries, however, disagreements with the United States over Afghanistan are secondary to more pressing concerns, such as propping up the regime of Syrian leader Bashar al-Assad. Additionally, Moscow is preoccupied with controlling eastern Europe and influencing U.S. and European elections. With these issues looming, Iran and Russia will likely use anti-U.S. policies in Afghanistan as a way to gain leverage in these areas.

A final consideration, when weighing the pros and cons of continued intervention, is that even if the Taliban wins control of Afghanistan it may not remain a direct threat to the United States. The 9/11 attacks proved devastating for the Taliban, and if they succeed in ridding the country of a U.S. presence, they may have little appetite in giving the superpower a reason to return. The Afghan Taliban have not supported a mass-casualty terrorist attack against the West in the post-9/11 era. They have not done so even in the face of the ferocious warfare waged against them, suggesting that their theater of operations is local and regional, not international. In addition, the Taliban hesitated to embrace al Qaeda’s global terrorism agenda even in the pre-9/11 era, although it stumbled into a tight embrace with the group in the end. But at the moment, the primary terrorism threat stems from Pakistan, not Afghanistan, because that is where the al Qaeda core has holed up. Al Qaeda, however, is weaker than before. As a result, the U.S. government has needed to carry out far fewer drone strikes on Pakistan in recent years. In 2010, the United States carried out more than 100 drone strikes, but only a handful of them in the last few years. Of course, we can’t rely on al Qaeda to remain down forever. The Taliban may again host terrorist groups if it gains power, and al Qaeda in Pakistan may return if U.S. pressure relents. 


To have a chance at securing the victory that Trump has promised, there would need to be a massive military escalation in Afghanistan, lasting many years and resulting in high numbers of American casualties. If the Trump administration is serious about taking on Pakistan as well, it would need to consider expanding the war there via drones, covert operations, and commando raids—a substantial surge. But Pakistan can easily respond by putting more fuel on the fire and has already made it clear that it is playing a long game. So expanding the war to Pakistan would demand a tremendous amount of effort and time.

From a political perspective, having gone big in the past makes it harder to go small: both Bush and Obama expended enormous amounts of resources to win in Afghanistan. That expectation of victory lingers even as U.S. resources have diminished and the war drags on. Trump’s promise of “we will win” sounds great as rhetoric, but it’s not a realistic policy goal given that he is committing substantially less in blood and treasure. If the United States couldn’t win against a weaker Taliban with 100,000 U.S. troops, then promising the same goal with a tenth of that manpower against a stronger Taliban is unlikely to succeed.

A better option might be to use a hedging strategy. This would involve a limited number of troops relative to current force levels. They would focus on counterterrorism and training Afghan forces. In addition, the spigot of U.S. money to Afghanistan would remain open, enabling the government in Kabul to buy allies and keep its current supporters happy. Washington would also try to increase pressure on Pakistan—admittedly, easier as a recommendation than as an actual policy—while urging the Afghan government to recognize the border between Afghanistan and Pakistan (the Durand Line), whose disputed status is a constant irritant to Islamabad. This approach will not solve all of Washington’s problems in Afghanistan, but it’s a sustainable way of preserving the United States’ interests while limiting its exposure.

This more reserved approach is also more sustainable. The human and financial cost of the U.S. presence in Afghanistan, for example, has declined as the total number of troops has fallen. Even if we include Trump’s troop increase, the roughly 12,000 U.S. troops in Afghanistan still constitute a fraction of the more than 100,000 stationed in Afghanistan at the height of U.S. presence there under Obama. Casualties have plummeted as the U.S. military footprint in Afghanistan has shrunk. In 2016, the United States lost 14 soldiers, down from more than 400 each year in 2010 and 2011. Because of far fewer deployed troops, the cost to maintain the war decreased to perhaps $20 billion or so a year, although this number varies dramatically depending on how the figure is computed. It is hardly small, but neither is it astronomical. 

Yet the inevitability of the latest surge and its meager size are due to two related factors. First, we are hard-wired to fall for the fallacy of sunk cost, and therefore tend to press on despite the obvious futility of additional investment. So we almost always throw good money after bad, especially when the “good money” is denominated in prestige, blood, treasure, and politicians’ reputations. Second, the lurking sense, among both policy makers and the public, that the sunk cost cannot be recovered through additional investment—that we cannot win—guarantees that new deployments will be bite-sized. And the fact that the latest surge will be small in scale makes it all that much easier to undertake. We will have the deep psychological satisfaction of bowing to the sunk cost fallacy, while our rational selves will devise ways to limit the self-defeating results. Striking a balance between the imperatives of cognitive bias and of strategic and political realities requires tight discipline and sound judgment, which, in this White House, are currently in short supply. But the possibility that our leaders are edging toward hedging suggest that such a balance might yet emerge.

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