After seven years of the Bush administration's neglect and mismanagement of Afghanistan, President Barack Obama was prompt in ordering the deployment of 21,000 more U.S. troops. Over 55,000 U.S. soldiers will soon be on the ground there. The replacement of General David McKiernan with General Stanley McChrystal at the head of U.S. operations in Afghanistan is also intended to increase force projection there. The United States' allies are under pressure to follow suit, if not with combat troops, then at least with training and money. All are concerned about the Taliban's recent success at persuading thousands of young Afghan men to sacrifice themselves to fight the foreign occupation. The Taliban's followers have pushed the Afghan government and its allies out of large swaths of the countryside and crept up to the gates of Kabul, bringing an alternative administration and sharia courts to the vacated areas. The Taliban leader Mullah Muhammad Omar recently offered, ironically, to give safe passage to NATO forces that choose to leave the country, just as the mujahideen offered safe passage to Soviet troops two decades ago.

Although sending more troops is necessary to tip the balance of power against the insurgents, the move will have a lasting impact only if it is accompanied by a political "surge," a committed effort to persuade large groups of Taliban fighters to put down their arms and give up the fight. Both the recent interagency white paper on U.S. policy toward Afghanistan and Pakistan and Obama's March 27 speech announcing a new U.S. strategy for Afghanistan acknowledged that integrating reconcilable insurgents will be a key complement to the military buildup. Yet U.S. policymakers have not adequately developed a vision of how to achieve reconciliation. Admitting their lack of knowledge about the precise character of the insurgency, they equate reconciliation with merely cajoling Taliban foot soldiers into crossing over to the U.S. side.

Such a minimalist approach is unlikely to deliver peace. What is required instead is a nimble, sophisticated political campaign that is built on a proper understanding of the nature of the insurgency and that, combined with the reconciliation efforts of the Afghan government, the United States' NATO allies, and Pakistan, enables insurgent commanders and their supporters to realign with the Afghan government. The overriding lesson of the U.S. experience in Iraq—first its failures and more recently its successes—is that no occupying power can hope to quash an insurgency by killing and capturing its way to victory. It must make friends, especially among its enemies. In Afghanistan, a counterinsurgency strategy that includes a credible attempt at reconciliation is more likely to achieve stability than one that relies solely on foreign troops and victories in the battlefield.


The idea that large groups of armed men bent on killing Americans and other Westerners can be persuaded to change sides may seem fanciful at first. But it is not—at least not in Afghanistan. After continuing uninterrupted for more than 30 years, war in Afghanistan has developed its own peculiar rules, style, and logic. One of these rules is side with the winner. Afghan commanders are not cogs in a military machine but the guardians of specific interestss—the interests of the fighters pledged to them and of the tribal, religious, or political groups from which these men are recruited. Few factors have motivated individual Afghan commanders over the years more than the desire to end up on the winning side. They have often switched camps midconflict. In doing so, they have not declared their loyalty to a new cause or a different tribe; they have argued that changing circumstances, such as a shift in the balance of power, demanded a strategic realignment. Their rationale was obvious: in a war that drags on, changing camps means living and holding on to power, as well as saving one's family and one's village. Thus in Afghanistan, battles have often been decided less by fighting than by defections. Changing sides, realigning, flipping—whatever one wants to call it—is the Afghan way of war.

Afghanistan's recent history is replete with examples of commanders choosing to flip rather than fight. In the most recent civil war, which lasted from the collapse of the erstwhile Soviet-backed regime in 1992 to the Taliban's capture of over 80 percent of Afghanistan in the fall of 1998, the heads of mujahideen groups constantly shifted their allegiances. The Uzbek general Abdul Rashid Dostum was the Tajik commander Ahmad Shah Massoud's friend first, and then he was his foe. The Hazara leader Abdul Ali Mazari fought against the Pashtun headman Gulbuddin Hekmatyar before fighting by his side. More than the fighting, it was this flipping that decided major outcomes. Constantly shifting alliances meant no single group could gain the upper hand; it was mass revulsion at the mujahideen warlords' depredations that eventually allowed the Taliban to persuade many factions to side with them. By the time the Taliban reached Kabul, their ranks were teeming with fighters once allied with someone else.

The Taliban have long marketed themselves as the Afghan faction most successful at maintaining internal cohesion and avoiding defections, but even their alliances were fluid when the U.S.-led war began in 2001. Large infusions of U.S. dollars, overtures from mujahideen and tribal figures alongside whom commanders and fighters had battled earlier, and the promise of honorable positions in the new order prompted many early realignments. Taliban leaders were willing to negotiate. In December 2001, with the movement in full collapse, Mullah Omar publicly offered to surrender the Taliban's stronghold, Kandahar, to Afghan tribal leaders. Soon after the Taliban fell, a brother of a top mujahideen figure aligned with al Qaeda, Jalaluddin Haqqani, participated in consultations with Afghan government patrons in Khost, a subtle indication that the family was seeking entrée into the new Afghanistan. It was only after these overtures led nowhere that the family's network joined the insurgency.

For all their reputed fanaticism, in other words, Taliban commanders will leave the movement and shift allegiances if the conditions are right. In December 2004, the senior Taliban commander Abdul Wahid announced that he had reconciled with the Afghan government. His move was justified, he argued, because he had essentially been released from any obligations to Mullah Omar in December 2001, after Mullah Omar asked him to lead the delegation that would surrender Kandahar to pro-coalition forces and thereby forsook his exalted position as "Commander of the Faithful." This rationale allowed Wahid to keep affirming his commitment to building an Islamic state in Afghanistan even as he announced that the Afghan president, Hamid Karzai, was his new leader. There are plenty of similar examples of Taliban commanders who have turned: the Hotak brothers of Wardak Province, who had held senior positions in the Taliban; Nur Ali Haidery Ishaqzai, the director of Ariana Afghan Airlines under the Taliban; Abdul Salam Rocketi, once the Taliban corps commander in Jalalabad and now a member of parliament; and Arsala Rahmani, a deputy minister under the Taliban turned senator today.

But such cases are still too rare, and more Taliban leaders must be encouraged to defect. One way to achieve this is to make it easier for them to borrow Wahid's argument. This would mean portraying those who align with Kabul and the coalition forces as patriotic Taliban truly devoted to the causes of Islam and an independent Afghanistan and those who persist in opposing progress by the central government as unpatriotic. Reconciliation in Afghanistan requires distinguishing the "good" Taliban from the "bad."


So far, Afghan leaders and their U.S. backers have made only halfhearted, ill-funded, and largely futile efforts to exploit the willingness of Taliban commanders to switch sides. Reconciliation with the Taliban has never been a sufficiently high priority for the Karzai government or Washington for either to put in place a coherent strategy encouraging defections from among the movement. Worse, the United States' misguided approach to the detention of suspected Islamist terrorists in Afghanistan, spurred on by political insiders in Kabul (mostly members of the non-Pashtun Northern Alliance) eager to harass personal rivals, drove people who might otherwise have cooperated into the insurgency. In other words, the people charged with stabilizing Afghanistan forfeited one of the most powerful tools at their disposal. In February 2002, for example, Wakil Ahmad Mutawakil, the Taliban's foreign minister and a man widely regarded as the most reasonable Taliban leader, voluntarily approached the new Afghan authorities expressing a desire to join the new order. He was arrested, detained in the U.S.-run prison at Bagram Air Base for 18 months, and then held under house arrest.

There have been many other such cases. Abdul Haq Wasiq, the Taliban's deputy minister of intelligence, and Rahmatullah Sangaryar, a senior field commander from Uruzgan, were shipped off to the U.S. prison camp at Guantánamo Bay, Cuba, after seeking government protection. So was Sahib Rohullah Wakil, the head of the Salafi movement in eastern Afghanistan, who had been elected to the Loya Jirga, or grand council, that convened in June 2002. Scores of fighters belonging to Wasiq's, Sangaryar's, and Rohullah's networks were once ready to recognize the Afghan government and lay down their arms, but they have not. The message from Washington and its Afghan allies could hardly have been clearer: hold out an olive branch, and you will go straight to jail.

Furthermore, what should have been the flagship program in the reconciliation process, the Afghanistan National Independent Peace and Reconciliation Commission (better known by its Dari abbreviation, PTS), has been a lackluster effort. Launched in 2005 and led by a former jihadi leader, Sibghatullah Mojaddedi, the PTS was given an ambitious agenda—to reintegrate former fighters into society—but almost no resources to carry it out. The commission has handed out certificates to former fighters declaring that they have joined the program and so should not be subject to prosecution. But it has never had the wherewithal to protect ex-fighters from retribution by the Taliban or harassment by the government. The PTS office in Kandahar, where the Taliban have their base and which is thus the most crucial part of the country for reconciliation, is a parody of the program. Its monthly budget, barely $600, is supposed to both cover its operating costs and support all the former fighters who choose to defect. The office's efforts have been minimal—it disseminates fliers and provides meager housing subsidies—and its record is dismal. Of the roughly 7,000 people whom the Kandahar office has certified during its four years in operation, fewer than a dozen were bona fide midlevel Taliban officials. The rest were foot soldiers or had no real links to the Taliban. The commission has no program for systematically following up with those it has processed. In the absence of any effective measures by the PTS, former fighters genuinely eager to put down their arms have found themselves caught between Scylla and Charybdis: harassment by either government security officials or insurgents.

Once reconciliation receives the attention it deserves within the overall strategy to rebuild Afghanistan, the Afghan government will need to overhaul its approach to wooing fighters and looking after them. It will need to act strategically, focusing on striking deals with important insurgent networks rather than with average fighters. And it will have to integrate the efforts made by Afghan government agencies and international groups to address the security, political, and economic needs of the networks coming on board.

Of all the shortcomings of the Afghan government and its NATO allies, it is the failure to provide security for ordinary Afghans that has most prevented large-scale reconciliation in the country. The Taliban have worked diligently to make the costs of reconciliation prohibitively high. "It is amazing to see how sensitive and scared everyone in Kandahar is to talk about the Taliban and the government reconciling," an Afghan scholar researching the reconciliation conundrum told us in April. "There is no [government] strategy in place to defy antipeace and antireconciliation attempts." Indeed, so far, the weakness of the Karzai administration and the steady spread of insecurity across the country's Pashtun areas, in the east and the south, have boosted the position of those insurgents who favor continuing the conflict.

In order for reconciliation to work, ordinary Afghans will have to feel secure. The situation on the ground will need to be stabilized, and the Taliban must be reminded that they have no prospect of winning their current military campaign. If the Afghan government offers reconciliation as its carrot, it must also present force as its stick—hence, the importance of sending more U.S. troops to Afghanistan, but also, in the long term, the importance of building up Afghanistan's own security forces. Reconciliation needs to be viewed as part of a larger military-political strategy to defeat the insurgency, like the one Washington has pursued recently in Iraq: win over the insurgents who are willing to reconcile, and kill or capture those who are not.


A focused campaign to win the cooperation of significant elements within the Taliban can succeed. For one thing, there is popular support for reconciliation in Afghanistan. In a nationwide poll sponsored by ABC News, the BBC, and ARD of Germany and conducted in February 2009, 64 percent of the respondents stated that the Afghan government should negotiate a settlement with the Taliban and agree to let the group's members hold office if they agree to stop fighting.

One model of inclusion is the talks between the Taliban and Afghan officials that took place in Mecca under the auspices of the Saudi government last fall. By hosting and endorsing the process, Riyadh generated greater engagement from core Taliban leaders with its initiative than had been generated by previous ones because of the moral authority the Saudi kingdom has within the movement. Informal feedback we received from insurgents suggested that the Saudi process helped promote dialogue and prompted different parts of the insurgency to contemplate what an eventual settlement might involve.

That said, it would be a distraction to focus too much on the prospect of a comprehensive settlement: in the short and medium terms, it seems highly unlikely that Taliban leaders will be willing to strike a broad deal with the Afghan government. They might not even be capable of doing so, because the Taliban is not a unified or monolithic movement. Some leaders and commanders who are influential within the movement are open to rapprochement, but a dialogue conducted through a single authorized channel could be hijacked by Taliban hard-liners: no Taliban leader would be prepared to openly challenge the hard-liners' resistance to dialogue. Reconciliation is an incremental process, and it should start before the pursuit of any comprehensive settlement.

One important step is for the Afghan government to tailor its approach to the needs of the fighters. The Taliban are predominantly Pashtun and conservative, but the movement also contains legions of men who fight for reasons that have nothing to do with Islamic zealotry. For many, insurgency is a way of life. The fighters are affiliated with particular commanders and receive comradeship and protection within their group. Unless they protect a drug-trafficking route, they tend not to be highly paid, but an occasional stipend from their commander is better than unemployment. And even if many fighters are fundamentally nonideological, membership in an insurgent network—in which elders and peers tell them that opposing foreign forces is virtuous—offers a kind of respectability. A well-organized reconciliation program would thus have to offer substitutes for all these benefits: comradeship, security, a livelihood, and respectability.

Another important element of a reconciliation strategy will be to recognize the specific needs of each group. The Afghan insurgency combines, on the one hand, the original Taliban idea that the movement is supratribal and that its fighters are pledged to a single leader and, on the other, traditional Afghan affiliations with multiple local and other groups. Networks of commanders play an important role. The group mobilized by the jihad-era veteran Haqqani, for example, runs one of the insurgency's most effective fronts in Kabul and southeastern Afghanistan out of its base in Waziristan. It has vested authority in one of Haqqani's sons and directly cooperates with al Qaeda and Pakistani jihadi groups; it is only nominally subordinated to Mullah Omar's Taliban in Kandahar. These command relationships differ fundamentally from those of a modern army or political party. The bulk of the Taliban's military operations are conducted by fighters operating within their home provinces, where their relationship to the local population is defined by their tribal status and political backgrounds rather than by the authority granted to them by the Taliban leadership in Quetta or Waziristan.

Reconciliation efforts will therefore have to zero in on the particular characteristics of each group: its tribal links, its traditions, the special conditions under which it functions. Any initiative to approach these groups should be spearheaded by interlocutors who have both credibility inside the Afghan establishment in Kabul and ties to insurgent networks. The trick will be to engage a critical mass of local commanders simultaneously. Reconciliation diplomacy must woo enough commanders in any single area to make recalcitrant fighters feel excluded, and it must enable the government to make a credible case that it can back commanders and their followers when they realign.

This vision of reconciliation concerns itself not so much with getting foot soldiers to lay down their arms as with building alliances with commanders and their troops. Fortunately, this also means the undertaking would not be prohibitively expensive. In Iraq, the U.S. government and then the Iraqi government put 100,000 Sunni gunmen on the payroll, many—if not most—of them former insurgents, for about $300 a month each. That amounts to $30 million a month, a reasonable amount given the costs of the war. In Afghanistan, the same amount could be used to give as many as 250,000 insurgents about $120 a month, which is equal to the average monthly salary of a low-ranking member of the Afghan National Army.


The heart of the reconciliation effort must first be a deal among Afghans. But now that the conflict has become internationalized, peace will also require international involvement. The core rationale for the current NATO mission in Afghanistan is to ensure that the Afghan authorities can prevent the Taliban's al Qaeda allies from exploiting Afghanistan as a base for terrorist operations. If they want to extricate themselves from the insurgency and become part of Afghanistan's new deal, Taliban commanders will have to demonstrate that they have broken with al Qaeda. They will have to exclude the global jihadists from their areas of influence. This is a message that Washington and its NATO partners need to disseminate among the insurgents.

As a tradeoff and to create an incentive for the insurgents to cooperate, the U.S. government and its allies should gradually hand over responsibility for the country's security to the Afghan national forces. After the Saudi talks, several leaders of the insurgency, such as Aga Jan Mohtasim, who was once the Taliban's finance minister, various official Taliban spokesmen, and some commanders of Hezb-i-Islami, started taking more political stands. For example, invoking the specter of the Soviet Union's humiliating defeat two decades ago, they have bandied around ideas for a timetable for the departure of foreign troops. Given the current security situation, no such timetable is likely or desirable. Still, it would be helpful to point out to the insurgents that progress on reconciliation, including their cooperation on security, would be the best way to get the foreign troops out. That is, Washington should substitute the model of the British experience in Northern Ireland for the the Soviet one in Afghanistan.

The United States is rightly committed to ensuring that Afghan forces, principally the police and the army, take over responsibility for the country's security; any U.S. military surge is essentially a temporary fix. On this point, ironically, U.S. policy converges with the insurgents' goals. Although a grand bargain in which Mullah Omar and his followers sign on to a nation-building process supported by the United States is unlikely, miniature versions of such a deal are attainable. The "patriotic" Taliban must be allowed to claim some of the success for the Afghanization of the country's security. Commanders and fighters should be formally associated with or absorbed into the police or the army, for example, which would allow the foreign troops to slip into the background. The real test of the commanders' commitment will be whether they can end attacks in their home provinces, cut off logistical support from insurgent rear bases, and cooperate with civilian activities aimed at their and their fighters' reintegration into Afghan society.


Recent developments in Pakistan, namely the expansion of the militants' influence and their success in paralyzing law enforcement efforts, are both a threat and an opportunity for reconciliation in Afghanistan. The Pashtun and Punjabi militants of Pakistan are distinct from Mullah Omar's Taliban fighters. But all these groups share a comradeship originally forged under the Taliban's rule in Afghanistan, between 1996 and 2001, during which some 5,000 - 7,000 Pakistani fighters served with the Taliban military at any given time. Now, the Afghan Taliban think it is unlikely that the Pakistani government will pressure them to give up their fight. And they see in the Pakistani Taliban's growing confidence and military strength a chance to get more support for their own armed struggle. The Afghan logic of siding with the winner means that the expansion of the insurgency in Pakistan makes even those Afghan Taliban most amenable to reconciliation with the Afghan government less likely to give in.

This is an obstacle, of course, but also an opportunity. Like it or not, the leaders, commanders, recruiters, financiers, and families of those running the Afghan insurgency are predominantly based in Pakistan, away from the war zone. Thus, a process of reconciliation based on accommodation with key networks of commanders within the insurgency—rather than with the whole of the Taliban leadership or the movement's foot soldiers—would have to start in Pakistan, where these networks are most accessible. And so Washington should encourage and help coordinate a low-profile but intensive dialogue between internationally backed mediators and the networks of commanders in Pakistan. It should include talks about how to exclude al Qaeda from Afghanistan and ultimately also Pakistan, find common ground on security arrangements for Afghanistan, and open up the way for insurgents to extricate themselves from the armed struggle and rejoin mainstream Afghan life. Insurgent networks cohere at least as much on the basis of social and economic interests as on political and ideological grounds; thus, the success of engagement will depend on building confidence. Much tea will have to be drunk to persuade commanders to realign.

To date, the supporters of reconciliation in Afghanistan have assumed that any serious effort to engage commander networks in Pakistan could be disrupted by the insurgency's backers. Now that U.S.-Afghan-Pakistani relations are being recalibrated, Washington is on the lookout for ideas about how Pakistan could do more to support progress in Afghanistan. Pakistan's backing of a low-profile political dialogue, even one that yielded only an agreement not to disrupt the process, would open up a whole new field of possibilities. What is needed is the creation of a sort of reconciliation safe haven in Pakistan. The mediation team should be allowed to operate without fear of government harassment in the areas currently exploited by commander networks: the Northwest Frontier Province, Baluchistan, the so-called tribal areas, and Pakistan's big cities. Such freedom of operation could help build confidence and create a foundation for a future settlement.


The Obama administration and its Afghan counterparts will have to get many things right in order to salvage the reconstruction effort in Afghanistan. Reconciliation with Afghan insurgents is one important component of any new successful strategy. But another three issues will also determine the fate of such a project: the outcome of the upcoming Afghan presidential election, the success of efforts to improve governance, and the effectiveness of the new campaign to boost security.

President Karzai has promoted many of his appeals for reconciliation with the insurgents by pledging his personal support. But given his diminished personal standing and the mounting uncertainty about the future of his leadership, significant progress is unlikely unless he changes his policies or a new president with the authority to cut credible deals is elected. And given that corruption, arbitrary detentions, and factionalized local governance have helped drive so many people to join the armed resistance, progress is only likely where such predatory behavior can be reined in. Finally, it is only if the United States' military surge can demonstrably stem the insurgents' influence in Pashtun areas that militants there will start to believe that they might be able to stay alive if they realign with the government.

If these conditions are met, a comprehensive strategy for reconciliation—launched with tea-drinking diplomacy that involves both Afghan and international parties and creates a safe haven for negotiations with Taliban commanders in Pakistan—could help bring stability to Afghanistan. Such an effort would not end the insurgency, but it would allow the residual military action to focus on rooting out its irreconcilable elements. Time is short, and effective engagement with the Taliban could mean the difference between a protracted, unwinnable conflict and a pragmatic solution acceptable to both Washington and its Afghan allies.

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
  • FOTINI CHRISTIA is Assistant Professor of Political Science at MIT.

    MICHAEL SEMPLE is a regional specialist focusing on Afghanistan and

    Pakistan, with extensive experience dialoguing with the Taliban.
  • More By Michael Semple
  • More By Fotini Christia