The Trouble With Allies
America Needs a Playbook for Difficult Friends
The United States has been fighting a war in Afghanistan for over 18 years. More than 2,300 U.S. military personnel have lost their lives there; more than 20,000 others have been wounded. At least half a million Afghans—government forces, Taliban fighters, and civilians—have been killed or wounded. Washington has spent close to $1 trillion on the war. Although the al Qaeda leader Osama bin Laden is dead and no major attack on the U.S. homeland has been carried out by a terrorist group based in Afghanistan since 9/11, the United States has been unable to end the violence or hand off the war to the Afghan authorities, and the Afghan government cannot survive without U.S. military backing.
At the end of 2019, The Washington Post published a series titled “The Afghanistan Papers,” a collection of U.S. government documents that included notes of interviews conducted by the special inspector general for Afghanistan reconstruction. In those interviews, numerous U.S. officials conceded that they had long seen the war as unwinnable. Polls have found that a majority of Americans now view the war as a failure. Every U.S. president since 2001 has sought to reach a point in Afghanistan when the violence would be sufficiently low or the Afghan government strong enough to allow U.S. military forces to withdraw without significantly increasing the risk of a resurgent terrorist threat. That day has not come. In that sense, whatever the future brings, for 18 years the United States has been unable to prevail.
The obstacles to success in Afghanistan were daunting: widespread corruption, intense grievances, Pakistani meddling, and deep-rooted resistance to foreign occupation. Yet there were also fleeting opportunities to find peace, or at least a more sustainable, less costly, and less violent stalemate. American leaders failed to grasp those chances, thanks to unjustified overconfidence following U.S. military victories and thanks to their fear of being held responsible if terrorists based in Afghanistan once again attacked the United States. Above all, officials in Washington clung too long to their preconceived notions of how the war would play out and neglected opportunities and options that did not fit their biases. Winning in Afghanistan was always going to be difficult. Avoidable errors made it impossible.
On October 7, 2001, U.S. President George W. Bush launched an invasion of Afghanistan in retaliation for the 9/11 attacks. In the months that followed, U.S. and allied forces and their partners in the Northern Alliance, an Afghan faction, chased out al Qaeda and upended the Taliban regime. Bin Laden fled to Pakistan; the leader of the Taliban, Mullah Omar, went to the mountains. Taliban commanders and fighters returned to their homes or escaped to safe havens in Pakistan. Skillful diplomatic efforts spearheaded by a U.S. special envoy, Zalmay Khalilzad, established a process that created a new Afghan government led by the conciliatory Hamid Karzai.
For the next four years, Afghanistan was deceptively peaceful. The U.S. military deaths during that time represent just a tenth of the total that have occurred during the war. Bush maintained a light U.S. military footprint in the country (around 8,000 troops in 2002, increasing to about 20,000 by the end of 2005) aimed at completing the defeat of al Qaeda and the Taliban and helping set up a new democracy that could prevent terrorists from coming back. The idea was to withdraw eventually, but there was no clear plan for how to make that happen, other than killing or capturing al Qaeda and Taliban leaders. Still, political progress encouraged optimism. In January 2004, an Afghan loya jirga, or grand council, approved a new constitution. Presidential and then parliamentary elections followed. All the while, Karzai strove to bring the country’s many factions together.
But in Pakistan, the Taliban were rebuilding. In early 2003, Mullah Omar, still in hiding, sent a voice recording to his subordinates calling on them to reorganize the movement and prepare for a major offensive within a few years. Key Taliban figures founded a leadership council known as the Quetta Shura, after the Pakistani city where they assembled. Training and recruitment moved forward. Cadres infiltrated back into Afghanistan. In Washington, however, the narrative of success continued to hold sway, and Pakistan was still seen as a valuable partner.
In February 2006, thousands of Taliban insurgents overran entire districts and surrounded provincial capitals.
Violence increased slowly; then, in February 2006, the Taliban pounced. Thousands of insurgents overran entire districts and surrounded provincial capitals. The Quetta Shura built what amounted to a rival regime. Over the course of the next three years, the Taliban captured most of the country’s south and much of its east. U.S. forces and their NATO allies were sucked into heavy fighting. By the end of 2008, U.S. troop levels had risen to over 30,000 without stemming the tide. Yet the overall strategy did not change. Bush remained determined to defeat the Taliban and win what he deemed “a victory for the forces of liberty.”
President Barack Obama came into office in January 2009 promising to turn around what many of his advisers and supporters saw as “the good war” in Afghanistan (as opposed to “the bad war” in Iraq, which they mostly saw as a lost cause). After a protracted debate, he opted to send reinforcements to Afghanistan: 21,000 troops in March and then, more reluctantly, another 30,000 or so in December, putting the total number of U.S. troops in the country at close to 100,000. Wary of overinvesting, he limited the goals of this “surge”—modeled on the one that had turned around the U.S. war in Iraq a few years earlier—to removing the terrorist threat to the American homeland. Gone was Bush’s intent to defeat the Taliban no matter what, even though the group could not be trusted to stop terrorists from using Afghanistan as a refuge. Instead, the United States would deny al Qaeda a safe haven, reverse the Taliban’s momentum, and strengthen the Afghan government and its security forces. The plan was to begin a drawdown of the surge forces in mid-2011 and eventually hand off full responsibility for the country’s security to the Afghan government.
Over the next three years, the surge stabilized the most important cities and districts, vitalized the Afghan army and police, and rallied support for the government. The threat from al Qaeda fell after the 2011 death of bin Laden at the hands of U.S. special operations forces in Pakistan. Yet the costs of the surge outweighed the gains. Between 2009 and 2012, more than 1,500 U.S. military personnel were killed and over 15,000 were wounded—more American casualties than during the entire rest of the 18-year war. At the height of the surge, the United States was spending approximately $110 billion per year in Afghanistan, roughly 50 percent more than annual U.S. federal spending on education. Obama came to see the war effort as unsustainable. In a series of announcements between 2010 and 2014, he laid out a schedule to draw down U.S. military forces to zero (excluding a small embassy presence) by the end of 2016.
By 2013, more than 350,000 Afghan soldiers and police had been trained, armed, and deployed. Their performance was mixed, marred by corruption and by “insider attacks” carried out on American and allied advisers. Many units depended on U.S. advisers and air support to defeat the Taliban in battle.
By 2015, just 9,800 U.S. troops were left in Afghanistan. As the withdrawal continued, they focused on counterterrorism and on advising and training the Afghans. That fall, the Taliban mounted a series of well-planned offensives that became one of the most decisive events of the war. In the province of Kunduz, 500 Taliban fighters routed some 3,000 Afghan soldiers and police and captured a provincial capital for the first time. In Helmand Province, around 1,800 Taliban fighters defeated some 4,500 Afghan soldiers and police and recaptured almost all the ground the group had lost in the surge. “They ran!” cried an angry Omar Jan, the most talented Afghan frontline commander in Helmand, when I spoke to him in early 2016. “Two thousand men. They had everything they needed—numbers, arms, ammunition—and they gave up!” Only last-minute reinforcements from U.S. and Afghan special operations forces saved the provinces.
In battle after battle, numerically superior and well-supplied soldiers and police in intact defensive positions made a collective decision to throw in the towel rather than go another round against the Taliban. Those who did stay to fight often paid dearly for their courage: some 14,000 Afghan soldiers and police were killed in 2015 and 2016. By 2016, the Afghan government, now headed by Ashraf Ghani, was weaker than ever before. The Taliban held more ground than at any time since 2001. In July of that year, Obama suspended the drawdown.
When President Donald Trump took office in January 2017, the war raged on. He initially approved an increase of U.S. forces in Afghanistan to roughly 14,000. Trump disliked the war, however, and, looking for an exit, started negotiations with the Taliban in 2018. Those negotiations have yet to bear fruit, and the level of violence and Afghan casualties rates in 2019 were on par with those of recent years.
Why did things go wrong? One crucial factor is that the Afghan government and its warlord allies were corrupt and treated Afghans poorly, fomenting grievances and inspiring an insurgency. They stole land, distributed government jobs as patronage, and often tricked U.S. special operations forces into targeting their political rivals. This mistreatment pushed certain tribes into the Taliban’s arms, providing the movement with fighters, a support network, and territory from which to attack. The experience of Raees Baghrani, a respected Alizai tribal leader, is typical. In 2005, after a Karzai-backed warlord disarmed him and stole some of his land and that of his tribesmen, Baghrani surrendered the rest of his territory in Helmand to the Taliban. Many others like him felt forced into similar choices.
Washington could have done more to address the corruption and the grievances that Afghans felt under the new regime and the U.S. occupation, such as pushing Karzai to remove the worst-offending officials from their positions, making all forms of U.S. assistance contingent on reforms, and reducing special operations raids and the mistaken targeting of innocent Afghans. That said, the complexity of addressing corruption and grievances should not be underestimated. No comprehensive solution existed that could have denied the Taliban a support base.
The ease of the initial invasion in 2001 distorted Washington’s perceptions.
Another major factor in the U.S. failure was Pakistan’s influence. Pakistan’s strategy in Afghanistan has always been shaped in large part by the Indian-Pakistani rivalry. In 2001, Pakistani President Pervez Musharraf officially cut off support for the Taliban at the behest of the Bush administration. But he soon feared that India was gaining influence in Afghanistan. In 2004, he reopened assistance to the Taliban, as he later admitted to The Guardian in 2015, because Karzai, he alleged, had “helped India stab Pakistan in the back” by allowing anti-Pakistan Tajiks to play a large role in his government and by fostering good relations with India. The Pakistani military funded the Taliban, granted them a safe haven, ran training camps, and advised them on war planning. The critical mass of recruits for the 2006 offensive came from Afghan refugees in Pakistan. A long succession of U.S. leaders tried to change Pakistani policy, all to no avail: it is unlikely that there was anything Washington could have done to convince Pakistan’s leaders to take steps that would have risked their influence in Afghanistan.
Underneath these factors, something more fundamental was at play. The Taliban exemplified an idea—an idea that runs deep in Afghan culture, that inspired their fighters, that made them powerful in battle, and that, in the eyes of many Afghans, defines an individual’s worth. In simple terms, that idea is resistance to occupation. The very presence of Americans in Afghanistan was an assault on what it meant to be Afghan. It inspired Afghans to defend their honor, their religion, and their homeland. The importance of this cultural factor has been confirmed and reconfirmed by multiple surveys of Taliban fighters since 2007 conducted by a range of researchers.
The Afghan government, tainted by its alignment with foreign occupiers, could not inspire the same devotion. In 2015, a survey of 1,657 police officers in 11 provinces conducted by the Afghan Institute for Strategic Studies found that only 11 percent of respondents had joined the force specifically to fight the Taliban; most of them had joined to serve their country or to earn a salary, motivations that did not necessarily warrant fighting, much less dying. Many interviewees agreed with the claim that police “rank and file are not convinced that they are fighting for a just cause.” There can be little doubt that a far larger percentage of Taliban fighters had joined the group specifically to confront the United States and the Afghans who were cooperating with the Americans.
This asymmetry in commitment explains why, at so many decisive moments, Afghan security forces retreated without putting up much of a fight despite their numerical superiority and their having at least an equal amount of ammunition and supplies. As a Taliban religious scholar from Kandahar told me in January 2019, “The Taliban fight for belief, for jannat [heaven] and ghazi [killing infidels]. . . . The army and police fight for money. . . . The Taliban are willing to lose their heads to fight. . . . How can the army and police compete with the Taliban?” The Taliban had an edge in inspiration. Many Afghans were willing to kill and be killed on behalf of the Taliban. That made all the difference.
These powerful factors have kept the United States and the Afghan government from prevailing. But failure was not inevitable. The best opportunities to succeed appeared early on, between 2001 and 2005. The Taliban were in disarray. Popular support for the new Afghan government was relatively high, as was patience with the foreign presence. Unfortunately, U.S. decisions during that time foreclosed paths that might have avoided the years of war that followed.
The first mistake was the Bush administration’s decision to exclude the Taliban from the postinvasion political settlement. Senior Taliban leaders tried to negotiate a peace deal with Karzai in December 2001. They were willing to lay down their arms and recognize Karzai as the country’s legitimate leader. But U.S. Secretary of Defense Donald Rumsfeld shot down the deal—in a press conference, no less. After that, between 2002 and 2004, Taliban leaders continued to reach out to Karzai to ask to be allowed to participate in the political process. Karzai brought up these overtures to U.S. officials only to have the Bush administration respond by banning negotiations with any top Taliban figures. In the end, the new government was established without the Taliban getting a seat at the table. Whether or not the entire group would have compromised, enough senior leaders were interested that future violence could have been lessened.
Obama’s decision to use airstrikes only in extremis virtually ensured defeat.
After pushing the Taliban back to war, Bush and his team then moved far too slowly in building up the Afghan security forces. After the initial invasion, a year passed before Washington committed to building and funding a small national army of 70,000. Recruitment and training then proceeded haltingly. By 2006, only 26,000 Afghan army soldiers had been trained. So when the Taliban struck back that year, there was little to stop them. In his memoir, Bush concedes the error. “In an attempt to keep the Afghan government from taking on an unsustainable expense,” he writes, “we had kept the army too small.”
The Bush administration thus missed the two best opportunities to find peace. An inclusive settlement could have won over key Taliban leaders, and capable armed forces could have held off the holdouts. Overconfidence prevented the Bush team from seeing this. The administration presumed that the Taliban had been defeated. Barely two years after the Taliban regime fell, U.S. Central Command labeled the group a “spent force.” Rumsfeld announced at a news conference in early 2003: “We clearly have moved from major combat activity to a period of stability and stabilization and reconstruction activities. . . . The bulk of the country today is permissive; it’s secure.” In other words, “Mission accomplished.”
The ease of the initial invasion in 2001 distorted Washington’s perceptions. The administration disregarded arguments by Karzai, Khalilzad, U.S. Lieutenant General Karl Eikenberry (then the senior U.S. general in Afghanistan), Ronald Neumann (at the time the U.S. ambassador to Afghanistan), and others that the insurgents were staging a comeback. Believing they had already won the war in Afghanistan, Bush and his team turned their attention to Iraq. And although the fiasco in Iraq was not a cause of the failure in Afghanistan, it compounded the errors in U.S. strategy by diverting the scarce time and attention of key decision-makers.
After 2006, the odds of a better outcome narrowed. The reemergence of the Taliban catalyzed further resistance to the occupation. U.S. airstrikes and night raids heightened a sense of oppression among Afghans and triggered in many an obligation to resist. After the Taliban offensive that year, it is hard to see how any strategy could have resulted in victory for the United States and the Afghan government. Nevertheless, a few points stand out when Washington might have cleared a way to a less bad outcome.
The surge was one of them. In retrospect, the United States would have been better off if it had never surged at all. If his campaign promises obligated some number of reinforcements, Obama still might have deployed fewer troops than he did—perhaps just the initial tranche of 21,000. But General Stanley McChrystal, the top U.S. commander in Afghanistan, and General David Petraeus, the commander of U.S. Central Command, did not present the president with that kind of option: all their proposals involved further increases in the number of U.S. military personnel deployed to Afghanistan. Both generals believed that escalation was warranted owing to the threat posed by the possible reestablishment of Afghanistan as a safe haven for terrorists. Both had witnessed how a counterinsurgency strategy and unswerving resolve had turned things around in Iraq, and both thought the same could be done in Afghanistan. Their case that something had to be done and their overconfidence in counterinsurgency crowded out the practical alternative of forgoing further reinforcements. Had Obama done less, U.S. casualties and expenses would likely have been far lower and still the conditions would have changed little.
It is worth noting that the much-criticized 18-month deadline that Obama attached to the surge, although unnecessary, was not itself a major missed opportunity. There is scant evidence to support the charge that if Obama had given no timeline, the Taliban would have been more exhausted by the surge and would have given up or negotiated a settlement.
But Obama did err when it came to placing restrictions on U.S. forces. Prior to 2014, U.S. airstrikes had been used when necessary to strike enemy targets, and commanders took steps to avoid civilian casualties. That year, however, as part of the drawdown process, it was decided that U.S. airstrikes in support of the Afghan army and police would be employed only “in extremis”—when a strategic location or major Afghan formation was in danger of imminent annihilation. The idea was to disentangle U.S. forces from combat and, to a lesser extent, to reduce civilian casualties. As a result of the change, there was a pronounced reduction in the number of U.S. strikes, even as the Taliban gained strength. Into 2016, U.S. forces carried out an average of 80 airstrikes per month, less than a quarter of the monthly average for 2012. Meanwhile, over 500 airstrikes per month were being conducted in Iraq and Syria against a comparable adversary. “If America just helps with airstrikes and . . . supplies, we can win,” pleaded Omar Jan, the frontline commander in Helmand, in 2016. “My weapons are worn from shooting. My ammunition stocks are low. I do not need advisers. I just need someone to call when things are really bad.” The decision to use airstrikes only in extremis virtually ensured defeat. Obama had purchased too little insurance on his withdrawal policy. When the unexpected happened, he was unprepared.
Bush had enjoyed the freedom to maneuver in Afghanistan for half his presidency and had still passed up significant opportunities. Facing far greater constraints, Obama had to play the cards he had been dealt. The Afghan government had been formed, violence had returned, and a spirit of resistance had arisen in the Afghan people. Obama’s errors derived less from a willful refusal to take advantage of clear opportunities than from oversights and miscalculations made under pressure. They nevertheless had major consequences.
Given the high costs and slim benefits of the war, why hasn’t the United States simply left Afghanistan? The answer is the combination of terrorism and U.S. electoral politics. In the post-9/11 world, U.S. presidents have had to choose between spending resources in places of very low geostrategic value and accepting some unknown risk of a terrorist attack, worried that voters will never forgive them or their party if they underestimate the threat. Nowhere has that dynamic been more evident than in Afghanistan.
In the early years after the 9/11 attacks, the political atmosphere in the United States was charged with fears of another assault. Throughout 2002, various Gallup polls showed that a majority of Americans believed that another attack on the United States was likely. That is one reason why Bush, after having overseen the initial defeat of al Qaeda and the Taliban, never considered simply declaring victory and bringing the troops home. He has said that an option of “attack, destroy the Taliban, destroy al Qaeda as best we could, and leave” was never appealing because “that would have created a vacuum [in] which . . . radicalism could become even stronger.”
Only after the surge and the death of bin Laden did withdrawal from Afghanistan become conceivable.
The terrorist threat receded during the first half of Obama’s presidency, yet he, too, could not ignore it, and its persistence took the prospect of a full withdrawal from Afghanistan off the table in the run-up to the surge. According to the available evidence, at no point during the debate over the surge did any high-level Obama administration official advocate such a move. One concern was that withdrawing completely would have opened up the administration to intense criticism, possibly disrupting Obama’s domestic agenda, which was focused on reviving the U.S. economy after the financial crisis of 2008 and the subsequent recession.
Only after the surge and the death of bin Laden did a “zero option” become conceivable. Days after bin Laden was captured and killed, in May 2011, a Gallup poll showed that 59 percent of Americans believed the U.S. mission in Afghanistan had been accomplished. “It is time to focus on nation building here at home,” Obama announced in his June 2011 address on the drawdown. Even so, concerns about the ability of the Afghan government to contain the residual terrorist threat defeated proposals, backed by some members of the administration, to fully withdraw more quickly. Then, in 2014, the rise of the Islamic State (or ISIS) in Iraq and Syria and a subsequent string of high-profile terrorist attacks in Europe and the United States made even the original, modest drawdown schedule less strategically and politically feasible. After the setbacks of 2015, the U.S. intelligence community assessed that if the drawdown went forward on schedule, security could deteriorate to the point where terrorist groups could once again establish safe havens in Afghanistan. Confronted with that finding, Obama essentially accepted the advice of his top generals to keep U.S. forces there, provide greater air support to the Afghan army and police, and continue counterterrorism operations in the country. The intention to get out had met reality and blinked.
So far, a similar fate has befallen Trump, the U.S. president with the least patience for the mission in Afghanistan. With Trump agitating for an exit, substantive talks between the Taliban and the United States commenced in 2018. An earlier effort between 2010 and 2013 had failed because the conditions were not ripe: the White House was occupied with other issues, negotiating teams were not in place, and Mullah Omar, the Taliban’s leader, was in seclusion—and then died in 2013. By 2019, those obstacles no longer stood in the way, and Trump was uniquely determined to leave. The result was the closest the United States has come to ending the war.
Khalilzad, once again serving as a special envoy, made quick progress by offering a timeline for the complete withdrawal of U.S. forces in return for the Taliban engaging in negotiations with the Afghan government, reducing violence as the two sides worked toward a comprehensive cease-fire, and not aiding al Qaeda or other terrorist groups. Over the course of nine rounds of talks, the two sides developed a draft agreement. The Taliban representatives in the talks and the group’s senior leaders refused to meet all of Khalilzad’s conditions. But the initial agreement was a real opportunity for Trump to get the United States out of Afghanistan and still have a chance at peace.
It fell apart. Although Trump toyed with the idea of holding a dramatic summit to announce a deal at Camp David in September 2019, he was torn between his campaign promise to end “endless wars” and the possibility of a resurgent terrorist threat, which could harm him politically. During an interview with Fox News in August, he was distinctly noncommittal about fully withdrawing. “We’re going down to 8,600 [troops], and then we’ll make a determination from there,” he said, adding that a “high intelligence presence” would stay in the country. So when the Taliban drastically escalated their attacks in the run-up to a possible announcement, killing one American soldier and wounding many more, Trump concluded that he was getting a bad deal and called off the negotiations, blasting the Taliban as untrustworthy. Trump, like Obama before him, would not risk a withdrawal that might someday make him vulnerable to the charge of willingly unlocking the terrorist threat. And so yet another chance to end the war slipped away.
The notion that the United States should have just left Afghanistan presumes that a U.S. president was free to pull the plug as he pleased. In reality, getting out was nearly as difficult as prevailing. It was one thing to boldly promise that the United States would leave in the near future. It was quite another to peer over the edge when the moment arrived, see the uncertainties, weigh the political fallout of a terrorist attack, and still take the leap.
The United States failed in Afghanistan largely because of intractable grievances, Pakistan’s meddling, and an intense Afghan commitment to resisting occupiers, and it stayed largely because of unrelenting terrorist threats and their effect on U.S. electoral politics. There were few chances to prevail and few chances to get out.
In this situation, a better outcome demanded an especially well-managed strategy. Perhaps the most important lesson is the value of forethought: considering a variety of outcomes rather than focusing on the preferred one. U.S. presidents and generals repeatedly saw their plans fall short when what they expected to happen did not: for Bush, when the Taliban turned out not to be defeated; for McChrystal and Petraeus, when the surge proved unsustainable; for Obama, when the terrorist threat returned; for Trump, when the political costs of leaving proved steeper than he had assumed. If U.S. leaders had thought more about the different ways that things could play out, the United States and Afghanistan might have experienced a less costly, less violent war, or even found peace.
This lack of forethought is not disconnected from the revelation in The Washington Post’s “Afghanistan Papers” that U.S. leaders misled the American people. A single-minded focus on preferred outcomes had the unhealthy side effect of sidelining inconvenient evidence. In most cases, determined U.S. leaders did this inadvertently, or because they truly believed things were going well. At times, however, evidence of failure was purposefully swept under the rug.
Afghanistan’s past may not be its future. Just because the war has been difficult to end does not mean it will go on indefinitely. Last November, Trump reopened talks with the Taliban. A chance exists that Khalilzad will conjure a political settlement. If not, Trump may decide to get out anyway. Trump has committed to reducing force levels to roughly the same number that Obama had in place at the end of his term. Further reductions could be pending. Great-power competition is the rising concern in Washington. With the death last year of ISIS’s leader, Abu Bakr al-Baghdadi, the shadow of 9/11 might at last recede, and the specter of terrorism might lose some of its influence on U.S. politics. At the same time, the roiling U.S. confrontation with Iran is a wild card that could alter the nature of the Afghan war, including by re-entrenching the American presence.
But none of that can change the past 18 years. Afghanistan will still be the United States’ longest war. Americans can best learn its lessons by studying the missed opportunities that kept the United States from making progress. Ultimately, the war should be understood neither as an avoidable folly nor as an inevitable tragedy but rather as an unresolved dilemma.