Gilles Dorronsoro
Dorronsoro’s research focuses on security and political development in Afghanistan. He was a professor of political science at the Sorbonne in Paris and the Institute of Political Studies of Rennes.
More >

The withdrawal of international forces from Afghanistan will leave the country worse than it was before 2001 in some respects. There is no clear plan for the future. Washington will progressively lose its influence over Kabul, and drone operations in Pakistan are not a credible way to fight jihadist groups on the Afghanistan-Pakistan border. The situation will only worsen after 2014, when most U.S. troops are out of the country and aid going to the Afghan government steeply declines.

Key Themes

  • The Afghan political system’s center of gravity—the east and the Kabul region—is gravely threatened by a Taliban advance that will take place in the spring of 2013 following the winter lull in fighting.
  • The Afghan regime will most probably collapse in a few years.
  • Political fragmentation, whether in the form of militias or the establishment of sanctuaries in the north, is laying the groundwork for a long civil war—a dangerous scenario for Western interests.
  • Though negotiations with the Taliban are unlikely before the troop withdrawal, the United States will not be able to pursue its longer-term interests in and around Afghanistan if it is not willing to deal with the Taliban.
  • Poised to take power after the Afghan regime’s likely collapse, only the Taliban can potentially control the Afghan border and expel transnational jihadists from Afghanistan.

Policy Recommendations

The coalition must strengthen security in the east and the Kabul region. Even if it means withdrawing troops more rapidly from the south, bolstering forces in the east will slow the Taliban’s progress and encourage them to take negotiations more seriously.

The United States must end its reintegration policy. The attempt to attract fighters away from the Taliban and “reintegrate” them into society enjoys few successes, fuels corruption, fosters insecurity, and ultimately convinces the population that the Taliban presence is justified.

Washington must not further limit its ability to open negotiations with the Taliban. Listing the Haqqani network, which is part of the Taliban movement, as a terrorist group was counterproductive.

Coalition operations should focus first and foremost on foreign jihadist groups. The Taliban should not be the primary target of drone attacks and night raids.

The United States must develop a new approach to Afghanistan, Pakistan, and India. A long-term regional policy that is not contradictory is needed to stabilize Afghanistan.

The Inevitable Withdrawal

Western disengagement from Afghanistan has entered a crucial period. Afghan president Hamid Karzai recently announced the third phase of the withdrawal of coalition forces. When it is over, 75 percent of Afghan territory will be under the control of the Afghan National Security Forces. Partnership agreements signed by the Afghan and U.S. governments in spring 2012 are incomplete but, while a certain number of unknowns remain with respect to the pace of withdrawal, most combat forces will be out of Afghanistan by the end of 2013. Only a small military force—compared to the boots currently on the ground—will remain in place after 2014.

Withdrawal has become a necessity, rather than a choice, with the alternative being to engage in an increasingly brutal military occupation.

This drawdown is irreversible, whatever the outcome of the U.S. presidential election, because of the U.S. public’s deep disaffection with the war and the irrevocable withdrawal of the European allies.1 The proliferation of damaging incidents—the desecration of corpses, burning of Korans, murder of civilians—point to U.S. troop exhaustion. Furthermore, the Afghans’ rejection of the foreign presence is so great that the coalition no longer has any political capital. The withdrawal has therefore become a necessity, rather than a choice, with the alternative being to engage in an increasingly brutal military occupation.

The largest uncertainty lies with the nature of the U.S. presence after 2014. Indeed, current agreements do not specify key elements: the number of troops, the role they will play, the legal protection they will receive, their collaboration with the Afghan army. But it is likely that the White House has already opted for one of the lower-troop-number options—fewer than 20,000 people involved essentially in training the Afghan army, plus an unspecified number of Special Forces. While private security companies will continue to play a role, the military impact of the U.S. presence will be relatively small. In particular, nighttime raids will likely have to be pared down as the quantity and quality of human intelligence on potential targets will decline.

In the end, the withdrawal is the result of a failed strategy, and the coalition is leaving behind a situation that in some respects is worse than it was before 2001. For over a decade, Western objectives have been undermined by wishful thinking and a misunderstanding of Afghan society. Even the much-heralded surge, as planned in 2009, never had a chance to reverse the momentum. The surge’s effect was limited militarily and disastrous politically, as its unsustainable cost led the coalition to set the unilateral withdrawal date, constraining its ability to negotiate with the insurgency.

Moreover, the insurgency has not been radically weakened by the recent military operations; it remains a vital threat to the Kabul government, particularly because there is no sign of a reduction in Pakistani support for the Taliban. In fact, the withdrawal will automatically translate into a Taliban advance, particularly in eastern and southern areas, such as Helmand, where the insurgents are contained only by the constant efforts of coalition forces.

Meanwhile, the Afghan regime is experiencing a deep and multipronged crisis. Its popular legitimacy is weak, an exit of the elite who have been enriched by the war looms, an economic crisis stretches across the country, and a clear mechanism for President Karzai’s succession remains elusive as it is becoming impossible to organize elections in a large number of rural areas. The current regime also cannot be expected to improve governance, which enhances Afghans’ attraction to the insurgency’s shadow government at a local level, particularly with respect to the resolution of private disputes.

Consequently, at best the Afghan government will be able to guarantee the security of cities and a few naturally pro-government regions, but it will lose control of rural Pashtun areas and the border provinces of Pakistan, with no prospects of regaining ground.2 Transnational jihadist groups will enjoy an expanded sanctuary on the Afghanistan-Pakistan border, against which counterterrorism techniques such as the use of drones and raids will be only marginally effective. In the long run, it is impossible to control a large geographical area and millions of inhabitants without some support on the ground, especially when the population is clearly opposed to such operations.

Finally, the influence the United States has over the regional players is decreasing; Washington will have no leverage over Pakistan in the next two years because of the logistical necessities of the withdrawal and the unstable military situation in Afghanistan. This in turn will make Afghanistan a staging ground for fights between regional powers, as it was in the 1990s. Today, Iran, India, and Pakistan sponsor competing Afghan political forces and heightened regional competition on Afghan soil is likely.

Resolving security challenges will require a coherent regional policy that involves harmonizing the U.S. approach to the three main players— Afghanistan, Pakistan, and India.

Ultimately, finding a military solution to the Afghan crisis is a lost cause. The United States is forced to contain a large-scale insurgency with very limited means and a failing Afghan government—a situation that recalls certain aspects of the 1989 Soviet withdrawal that resulted in the progressive isolation of government-held areas. Yet today’s discussion tends to focus on the survival of the Afghan regime, while the key issue is to define a strategy for the longer term. Security challenges in particular cannot be resolved by, or even based mainly on, counterterrorism operations, especially the use of drones, which requires a strong reliance on Pakistan. It will require a coherent regional policy that involves harmonizing the U.S. approach to three main players—Afghanistan, Pakistan, and India.

A Taboo Comparison? Only rarely have systematic comparisons been made between the Soviet occupation and withdrawal and the unfolding Western experience. Ideological and political sensitivities may explain this aversion, yet studying the similarities and differences between the two experiences can be enlightening and useful.

In both cases, foreign powers attempt to impose a social model of modernization that is not acceptable to the local population, apart from the urbanized elites. It is not surprising that they, as former Afghan communists, have embraced the Western project, notably the (cautious) liberalization of mores and economic modernization. Both withdrawals also focus security forces on the “useful Afghanistan,” while guerilla fighters control the territory in the border provinces with Pakistan. Finally, Pakistan’s support for the Taliban directly parallels Pakistan’s support for the mujahideen in the 1980s. In both cases, Islamabad did so for essentially the same reasons as part of its competition with India. The regional power game—a factor that destabilized Afghanistan in the 1990s—is being revived and looks much the same.

There are two major differences. First, the current regime does not possess the ideological and social cohesion of the communist regime, and its ability to survive militarily has not been demonstrated. The officer corps, in particular, is still politically divided and not battle hardened. Second, the Taliban form a united movement with few rifts, compared to the infighting of the mujahideen in the 1990s. It is therefore unlikely that local agreements will be sufficient to halt the Taliban advance, as was the case in 1989‒1990. 

The Insurgency’s Resilience

In early 2010, the surge began with additional U.S. troops sent to Afghanistan in the hope of reversing the balance of power, pressuring insurgents to support the Afghan government, and giving the latter room to build its army and improve governance. Military operations of unparalleled intensity played out in 2010 and 2011, and there was a significant increase in insurgent, coalition, and civilian losses.3 Yet because of the limited number of available troops and the difficulty of standardizing the practices of a very diverse coalition, a coherent counterinsurgency strategy was never implemented at the national level.

Faced with pressure from the insurgency, military solutions were favored to the detriment of a political approach. U.S. forces gave precedence to tactics they believed had worked in Iraq: essentially, the systematic elimination of insurgent leaders through nighttime raids.4 Thousands of Taliban—group leaders and local officials belonging to the shadow government—were killed, despite the temporary slowdown in operations in early 2012 demanded by President Karzai.

The limits of this approach are now becoming clear. For the most part, the goals of the surge were not met, in part because of the Pakistani support for the Taliban; the only progress took place on a very local level and will not survive the departure of American troops. With the population massively opposed to the raids, these tactics coincided with the abandonment of the effort to win hearts and minds.

Wherever the military pressure became too strong, the Taliban avoided head-on combat, preferring to adopt indirect techniques such as planting improvised explosive devices and carrying out targeted assassinations. The Taliban adopted a defensive posture in the south in places like Helmand and Kandahar, but they maintained a political presence there, since, given the lack of noticeable improvement in local governance, the government was not able to gain popular support. The Taliban made some headway in the north a few years ago, but the coalition managed to stabilize the situation. In the east, however, the Taliban strengthened their position. In all, the number of attacks launched by the Taliban was higher in 2011 than in 2010.5 There was a sharp drop in the insurgency’s military activities during the first half of 2012, but the summer of 2012, with a high level of military activities on the insurgents’ part, indicates that the momentum is still on the Taliban’s side.

The targeted eliminations of thousands of Taliban had only a transient, local impact because the insurgency’s leadership lives in Pakistan under the protection of the Pakistani army and has not been directly affected by the strikes.

The targeted eliminations of thousands of Taliban had only a transient, local impact because the insurgency’s leadership lives in Pakistan under the protection of the Pakistani army and has not been directly affected by the strikes. Moreover, mid-level leaders have been quickly replaced, which indicates a healthy level of institutionalization. While the caliber of these leaders has probably declined, their loyalty to the organization can only be stronger, given that their affiliation with the Taliban is increasingly their only source of legitimacy. It is also Taliban policy (as it was in the 1990s) to regularly replace district and provincial leaders to avoid the establishment of strong local players and problems of corruption.

More than a military organization, members of the Taliban make up a political party, which explains their attention to management of the population, via the legal system and taxes, for instance. The coalition forces were not capable of dismantling this shadow government, so the core of the organization was preserved. The Afghan government’s lack of notable progress on governance prevented military advances from having any long-term impact. And while the pace of operations succeeded in sowing disarray among the Taliban on the local level, for example in Kandahar or Helmand, their legal system continues to function, with appeals courts and a swift and generally uncorrupt justice system. According to various sources, the number of districts in which the Taliban effectively administer justice is about 150 out of 398.

A good indicator of the insurgency’s resilience is the failure of U.S. reintegration policy. Devised as part of the counterinsurgency strategy, it is based on the idea of rallying small groups or individuals to the government and is therefore the opposite of a political dialogue with the Taliban leadership. According to the Afghan Peace and Reintegration Program established in September 2010, those who have “reintegrated” give up their heavy weapons, sign an oath of allegiance, receive a $120 stipend for three months, and take part in a reintegration program; small loans are provided to the communities that take them in. Officially, those who have “reintegrated” do not serve in the Afghan Local Police, although they will likely be integrated into that force in the long term. According to official U.S. sources, the United States and its allies are committed to provide some $94 million in funding to the Afghan Peace and Reintegration Program, $20 million of which had been spent by October 2011.6

The coalition considers the reintegration program a success, but it is deeply counterproductive, fueling corruption and disorder without weakening the insurgency.

The coalition considers this program a success,7 but it is deeply counterproductive, fueling corruption and disorder without weakening the insurgency. Moreover, the statistics are low. In 2011, there were around 3,000 “reintegrees” in the entire country, with nineteen in Regional Command (RC) Southwest, 131 in RC South, 979 in RC West, 1,684 in RC North, and 157 in RC East. The cost of those reintegrees was $20 million. The insurgency, meanwhile, has mobilized tens of thousands of combatants. Furthermore, those who have gone to the government’s side are not important commanders, and the program has had no measurable military impact on the insurgency.

In addition, a large majority of those who have been reintegrated come from the north and the west, signaling that the program is not operating as planned. Indeed, far from being the result of military pressure, people are rallying to the government more frequently in quiet areas. In Kandahar and Helmand, where the coalition effort is strongest, “reintegrees” are practically nonexistent. This can be explained by the mechanisms of corruption and political patronage that are at work in most cases. The reintegration program is seen by local strongmen as a way to build their clientele and mobilize armed men on their side. Especially in the north, a large part of those who have been reintegrated are not Taliban but small, often bandit-like armed groups with no political affiliations. The local power brokers who are competing for spots on reconciliation committees use this program to obtain resources for their relatives or allies, building local militias and increasing the risks of disorder. 

The Deconstruction of the Regime

While the Taliban gathers momentum, in 2013 and 2014, the Afghan regime will confront three crises essentially simultaneously: an economic crisis sparked by the drop in Western spending, an institutional crisis as the end of President Karzai’s term approaches, and a security crisis as the Taliban are expected to launch an advance beginning in the summer of 2013.

Western spending will probably drop to less than $5 billion a year after 2014, compared to more than $100 billion in 2011 from the United States alone. It is clear that the United States and its allies are not particularly enthusiastic about financing the Afghan government, partly because of widespread corruption. The civilian surge pledged by the State Department has never materialized, and USAID funding has also begun to decline. Of course, the impact of decreased civilian aid to the population will not be as significant as it could be since much of the funding has not reached the populace; a large part of the aid was rather siphoned off by Western companies and the elite connected to the regime. Furthermore, the aid was not evenly dispersed, with the majority spent in Kabul and in the most heavily disputed provinces.

The withdrawal’s economic impact will be most strongly felt in relation to the reduction of spending linked to the troop presence, particularly on infrastructure projects. The end of the wartime economy will likely trigger an increase in drug manufacturing, but the funds from those enterprises will not be enough to compensate for the reduction. All of this will lead to economic crisis.

Afghanistan’s political elite have demonstrated no ability to act collectively to safeguard the current electoral system.

An institutional crisis will also come to a head with the end of President Karzai’s second term in 2014. The political system has lost all democratic legitimacy in the eyes of the Afghan people. With security conditions making it impossible to organize balloting in many rural areas, it will be impossible to hold new presidential elections. Calling a Loya Jirga, a “traditional” assembly of elders appointed by the president, to elect the new head of the country is probably the only practical way to move forward, but the legitimacy of a president elected in such a process will likely be contested. The parliamentary elections a year later will encounter the same obstacles, without the possibility of a Loya Jirga. Moreover, the country’s political elite have demonstrated no ability to act collectively to safeguard the current political system.

Increasingly, people are developing exit strategies. Regardless of what they tell their diplomatic contacts in Kabul, the political elite are preparing to go into exile, scrambling to obtain foreign passports, moving their families to Dubai, and shifting massive amounts of money abroad. The sharp decline in high-end real estate prices in Kabul signals that disengagement is the most common attitude. In addition, the new Afghan middle class, whose existence is directly linked to the Western financing and support, is not politically mobilized and will not be a factor in the transition process. No organized segment of the urban population is ready to support the current government in the coming crisis.

The Western withdrawal will also lead to a security crisis as the Afghan National Army and the Afghan National Police struggle to confront the Taliban. Following the example of the communist regime after the Soviet withdrawal in 1989, the Afghan government will have to streamline its security system and focus on withstanding pressure from the insurgency. The land seized by the insurgency is gone for good because the Afghan army’s structure is essentially defensive and thus incapable of recapturing lost territory. Maintaining control of Afghanistan’s major cities and main transport corridors is thus the only realistic goal.

But the Afghan army will struggle to meet even that objective, with force numbers dwindling and limited skills. According to official figures, after peaking at 350,000, the Afghan security forces (police and army) should stabilize around 230,000 because of funding limitations.8 However, the number of soldiers that can actually be mobilized will probably barely exceed 100,000 in 2013, and their ability to carry out operations without coalition assistance will generally be limited. The coalition’s training of the army, which should in principle continue after 2014, is running into numerous problems, notably the proliferation of “green on blue” attacks on international forces launched by Afghan security forces. These reflect a growing rejection of foreign forces as much as Taliban penetration. Furthermore, one in seven soldiers reportedly deserted in 2011, which limits the effectiveness of training and points to a serious motivation problem.9

In addition, the Afghan National Police, whose force numbers will stabilize around 100,000, play a major role in the counterinsurgency effort—often to the detriment of normal operations. They are more mobile than the Afghan National Army, have more contact with the population, and represent the only available official interlocutors at the local level.10 But therein lies the problem—the police force is essentially local and defensive. The same holds for the Afghan Local Police, essentially local militias that should include 30,000 Afghans by the withdrawal. They have been the source of numerous incidents with the population.11 While they may occasionally help contain the insurgency, their destabilizing role will strengthen with the withdrawal and the oversight of their activities will diminish.

The Afghan National Army hardly ever leaves its barracks because of the prevalence of improvised explosive devices and limited air support, which will be even more limited in the future. At most, it will maintain its fixed positions, leaving the rural areas to the Taliban. Furthermore, a significant portion of Afghan National Army and Afghan Local Police forces will be involved in a static defense of the major roads, further limiting their offensive capability. They will focus in particular on the Kabul-Kandahar-Herat road, the Kabul-Jalalabad road, and the Kabul-Kunduz/Mazar-i Sharif road in the north, which are vital to the Afghan economy. The Taliban can, if not totally shut down traffic, which would not make much sense from their perspective, at least attack convoys and set up temporary roadblocks on these thoroughfares. And the nighttime raids, which are the coalition’s last tactically aggressive weapon, will be harder to carry out going forward due to Afghan government opposition. 

The Taliban’s Approach

The Taliban probably number less than 50,000 but are likely to recruit more broadly as the movement increases its territorial control. Their strategy will become more offensive as the withdrawal changes the balance of military power. With this shift, the Taliban could muster hundreds of men, perhaps thousands for a military operation—enough to capture a district capital.

In preparing for the post-withdrawal period, the Afghan National Police and Afghan Local Police have become the Taliban’s main targets. And the insurgency’s targeted assassinations, which are to some extent a response to NATO’s strategy, are directed mainly at figures likely to unite pro-government forces at the political level. Due to the regime’s lack of strong institutions, these assassinations alter local political configurations and generally cause further fragmentation, which facilitates the Taliban’s advance. The assassination of the president’s half brother, Ahmed Wali Karzai, in Kandahar in July 2011, for instance, significantly weakened the Karzai clan, while that of Burhanuddin Rabbani, the former president and former leader of the Islamic political party Jamaat-e-Islami, disrupted the political equilibrium in the province of Badakhshan and politically weakened the north. These assassinations also put pressure on individuals and facilitate agreements that are currently being negotiated between the insurgency and the Afghan National Army in certain outposts where the balance of power is too unfavorable.

Beginning in spring 2013, the Taliban will probably step up attacks throughout the country to test the Afghan National Security Forces’ defenses. Progress is most likely in their two strongholds, the south and the east. In the south in particular, the Taliban can expect to gather strength in northern Helmand Province and near the city of Kandahar—the coalition will progressively withdraw its forces from that area and the Afghan army is incapable of holding land captured by the coalition in 2009‒2010. The insurgency may thus gain a psychological victory starting in summer 2013, and their progress in Helmand will have a direct impact on security in Kandahar and on the Kandahar-Herat road. However, the strategic challenge lies in the east Kabul region along the Afghanistan-Pakistan border. There the Taliban are at a considerable advantage because they are just a few hours’ march from their sanctuary in Pakistan. Coalition forces’ evacuation of the provinces of Kunar and Kapisa and the insurgency’s undivided domination of the border provinces, as well as those of Wardak and Ghazni, is a particular threat to the regime’s survival.12

The coalition is likely underestimating the extent of the Taliban advance in 2013, which will depend on several factors—notably the number of residual coalition troops and the degree of Pakistani support. For instance, there is good reason to believe that in certain districts where the insurgents are firmly in control, negotiations have already taken place to quietly transfer power to the Taliban. And the psychological effect of the withdrawal and advance could cause districts that are in principle defensible to fall as a result of panic. Moreover, the Western withdrawal could allow a more direct Pakistani presence on the ground.

But the Taliban may also make mistakes. Frontal attacks could lead to failure and demoralization, as happened in Jalalabad in 1989 when the Afghan army, without direct Soviet involvement, successfully defeated the mujahideen. Excessively harsh treatment of the population could also undermine the Taliban’s influence in undecided areas. But unlike 1989‒1992, it is extremely doubtful that local agreements will break the insurgency’s momentum and stabilize the military map. The Taliban are more than just a military force—they are a political movement seeking to control territory and to topple the regime.

The Taliban are more than just a military force—they are a political movement seeking to control territory and to topple the regime.

Keeping all of this in mind, how great a shock should be expected? Afghanistan has 398 districts, and on the basis of inherently approximate assessments, I believe that about a quarter of these could fall totally—that is, including district capitals—under the insurgency’s control in two years.13 These estimates are rather conservative; that is, they are based on situations in which the Taliban already occupy a position of strength. The capture of these district capitals could be achieved under two conditions: if the government is unable to send reinforcements or provide the necessary logistics to isolated outposts; and if the Taliban are in a situation in which they can easily mass troops, with the proximity of the border being key. In addition, between 50 and 100 other districts will fall under the insurgency’s influence, with functioning Taliban administration (with courts and tax collection) and Kabul officials unable to travel outside the district capitals without a substantial military escort. Furthermore, with the exception of Panjshir and Hazarajat, most districts will be exposed to Taliban military operations. Finally, the cities of Kandahar, Ghazni, and Khost are notorious for their Taliban infiltration, enabling the group to develop intelligence networks and exert pressure on officials and businessmen.

This new political-military map has a number of implications. Success will allow the Taliban to free up forces and focus on their next objective—the capture of provincial capitals. In a few cases, notably the provinces of Paktika, Khost, Kunar, and Nuristan, that goal is achievable by 2014. They will also be able to threaten the main roads, which goes hand-in-hand with seizing certain provincial capitals. And the Taliban will be able to act as an alternative state, with a district-level administration. Agreements with nongovernmental organizations wishing to work in rural areas will become more commonplace because the insurgency will be in control of a large portion of the rural areas, bolstering the Taliban’s position as a key political actor administering at least a quarter of the population. 

Could the Afghan Regime Survive?

At first glance, the mechanisms leading to the fall of the regime seem inexorable. But certain developments could stabilize the situation for a few years, such as a change in Pakistani policy, internal divisions within the Taliban, and a new president. In reality, these developments are unlikely and would come about only as a result of unpredictable events—a major political crisis in Pakistan or the death of the Taliban’s spiritual leader Mullah Omar, for instance.

Instead of supporting the insurgency and providing the Taliban sanctuary on Pakistani soil, Islamabad could in theory take an active role combatting insurgents on its territory. Such a reversal of Pakistani policy could deal a severe, even fatal blow to the Taliban and remains the most certain way to stabilize the Afghan regime. But a change in Pakistan’s policy has been announced regularly for the past ten years, and it has never taken place. Indeed, the Pakistani military never ceased its support for the Taliban and Hezb-i-Islami, another insurgent group, and became an increasingly difficult U.S. “ally.” It is highly improbable that change will come in the future for several reasons.

The security cost would be enormous for Pakistan, particularly if the Afghan Taliban and Pakistani jihadist groups were to join forces for offensive operations in Pakistan. Given the highly unstable situation on the border and the Pakistani government’s inability to (re)establish civilian authorities after military operations (in Waziristan for example), the consequences would be grave. The Pakistani army would need to become more heavily involved along its western border with Afghanistan even though its priority is India. The U.S.-Pakistani relationship has deteriorated significantly in recent years as a result of various incidents, such as recent bombardments along the Afghanistan-Pakistan border and the raid on Osama bin Laden’s compound outside of Islamabad, reducing the likelihood that Pakistan would help the United States meet its objectives. Further complicating matters, the Pakistani army is convinced that India is taking an offensive position in Afghanistan, supporting anti-Taliban groups and Baluchi tribes calling for independence from Pakistan.14 Indian policy is interpreted as a strategic threat to Pakistan. Whatever the reality of those perceptions, they fuel Pakistan’s stronger-than-ever support for the insurgency.

The idea that the United States could exploit divisions within the insurgency to weaken it and stabilize the Afghan government seems increasingly more far-fetched.

The idea that the United States could exploit divisions within the insurgency to weaken it and stabilize the Afghan government seems increasingly more far-fetched. The notion that the insurgency is composed of local groups without a national strategy has by now been discredited, but the role of the so-called Haqqani group and Hezb-i-Islami is still being discussed. The Taliban are by far the most important political-military organization and the only one to operate on a national scale. This is a military conflict and the logic is for one organization to acquire a military monopoly. Some divisions do seem to exist. The Taliban have a competitive relationship with Hezb-i-Islami but enjoy a huge military edge. Hezb-i-Islami is also double-dealing, working actively both with Karzai and the insurgency. But that situation is not a threat to the strength of the insurgency. Furthermore, portraying the Haqqani group as independent of the Taliban, on the same level as Hezb-i Islami, is misleading and—more important—has no practical significance. Indeed, the coalition has never been in a situation, nor will it be this year, in which it could exploit any (potential) differences between the central leadership headed by Mullah Omar and the Haqqani family. The only thing that could test the loyalty of the Haqqani group would be an offer to sign a ceasefire or to negotiate on the Taliban’s behalf, which is not about to happen. It is also unlikely the Taliban would collapse following, say, the death of Mullah Omar. After all, the organization has shown extraordinary resilience since 2001, largely due to the robust clerical network at the top of its leadership structure.

Finally, the current regime could in principle recover after 2014 if following President Karzai’s departure it came under the authority of a new president promoting reformist policies. However, no new leader is currently in sight, and it is especially hard to see how such a person could control the very powerful networks that are siphoning off international aid and organizing corruption and opium trafficking at a time of waning American influence (which was never decisive in the first place). The provincial networks will grow increasingly autonomous, because the redistribution of state resources through patronage will probably be less effective in the future. Networks linked to drugs or contraband will likely become more independent while the system in general grows more fragile. 

Future Scenarios

With the state on a path to greater instability, several scenarios are beginning to take shape for the period following 2014. In the first, the Taliban’s military advance at the district level eventually enables them to isolate the cities, causing the urban elite to defect to the insurgency and bring about the collapse of the regime. Jalalabad’s roads, in particular, are highly vulnerable; the Kabul-Jalalabad-Peshawar road is already threatened by the insurgency. The fall of the city would probably lead to the fall of the regime, because the loss of one major city will most likely be followed by a general panic and the disintegration of the current regime. This would most likely take several years and would largely replicate the Taliban’s first takeover of Kabul in 1996. A second scenario posits collapse due to tensions within the regime. The strong presence of Hezb-i-Islami’s legal branch around President Karzai as well as the politicization of the army evoke the possibility of a coup d’état, followed by a period of anarchy.15 This scenario is not fundamentally incompatible with the first one and is reminiscent of the ouster of former Afghan president Najibullah in 1992.

One last scenario involving the establishment of a sanctuary against the Taliban in the north has often been debated and was a topic of discussion at the end of the Soviet occupation. If the north were politically unified, it could impose itself more easily on Karzai and potentially determine the course of political developments in the capital, including the course of negotiations with the insurgency.

But given the speed of the withdrawal and the end of Karzai’s term in 2014, there remains very little time to unify the north. Militarily, the situation is relatively stable this year, but the government does not have the means to dislodge the insurgency where it has taken root. Politically, the north is far from stable and very fragmented. Historically dominant parties such as Jamaat-e-Islami, Jumbesh, and Hezb-i-Wahdat take different positions or are divided internally. Furthermore, political parties do not have their own structures and are essentially state parasites, with strongmen who monopolize power at the provincial level and (generally) benefit economically from the system. In particular, they have no military structures independent from the Afghan National Army that would enable them to resist a Taliban offensive.

It seems unlikely that northern political factions could unify politically and militarily to become a player on the national scene. One recent attempt to unite the north into a nationally recognized body has made little progress. The Jebhe-ye Melli (National Front) was launched in late 2011, but the structure has no real military presence on the ground, no program, and its activities are very limited. Nor does the group have any political project, as the parties that comprise it represent local interests or advance ethno-political claims that do not constitute a national platform. Indeed, it is hard to see what kind of program could develop, outside of a vague call for a form of decentralization at the provincial level.

At the end of the day, the most likely scenario is the collapse of the Afghan regime in a few years, after a steady period of weakening. The alternative is the containment of the insurgency, with an indefinite civil war and no prospects of strengthening the current central government. Either way, the coalition will be unable to decisively determine the trajectory of the Afghan regime after 2014. 

In Search of an Objective

So, how should the United States proceed? Washington first needs to get its priorities straight. The Afghan regime’s short-term survival has become the coalition’s goal instead of the means to an end.

Since 2001, the coalition has had two final objectives, depending on the moment and sometimes concurrently: neutralizing transnational jihadist groups and maintaining a position of regional influence. In the years following the invasion, the American military presence was essentially structured to fight al-Qaeda—which had a larger presence in Pakistan than in Afghanistan—as well as Taliban groups (or groups assumed to be Taliban, as many mistakes were made). Due to the limited number of troops, achieving the first goal was up to Special Forces, backed by Afghan militias that were essentially motivated by vengeance and financial gain. That policy was so counterproductive that, in 2004‒2005, the U.S. Army added state building to its responsibilities, notably building up Afghan security forces.

As the policy shifted and the military escalation led to massive U.S. investment, as reflected by the huge bases in Bagram, Shindand, and Kandahar, the objectives were redefined. Afghanistan was supposed to become a long-term ally with permanent bases that could exert pressure on Iran, Pakistan, and Central Asia (the competition with China being implicit). The rebirth of the Afghan state would promote both effective counterterrorism and the indefinite presence of U.S. bases, thus altering the regional strategic equation. For that, further state building was required. One of the consequences of state building was that the Taliban steadily became the main enemy, while al-Qaeda largely disappeared from the Afghan scene. The war against the Taliban was justified by their alliance—whether tactical or more fundamental, depending on the interpretation—with transnational jihadist groups.

Now, the Afghan state is too weak to guarantee the country’s security or to serve as a channel for U.S. influence. With the troop withdrawal, the goals of the survival of the regime and the fight against jihadist movements become partially contradictory and, most important, those ends appear to be disconnected from the means that will be available in the future. The level of resources is largely understood, so analyzing what could be accomplished with the resources that will be available can help predict what policies are most likely to be adopted after 2014. The instruments available to the United States in coming years will be limited and poorly adapted to fighting a large-scale insurgency. There will be a radical and irreversible decrease in resources, which means the idea of using Afghanistan as a channel of regional influence no longer makes sense; consequently, maintaining permanent bases there is no longer a credible objective. The United States must focus on realistic objectives, which raises the question of the effectiveness of counterterrorism and the relations between the Taliban and jihadist groups.

The dynamics on the Afghanistan-Pakistan border demonstrate that the terrorist threat today is relatively marginal and directed essentially against India and Pakistan. Thus, from the U.S. perspective, the problems and solutions after the withdrawal are essentially regional questions. Yet the U.S. withdrawal could have wider ramifications. It is freeing up a very large region, comprising several million people, where jihadist groups can take refuge, and it is creating dynamic conditions that strengthen the sanctuary along the Afghanistan-Pakistan border. The Mohamed Merah case in France also shows that this area remains a magnet for jihadists from Western countries. Merah killed seven people in France in March 2012, including three Jewish children as well as French soldiers whose unit had fought in Afghanistan. A member of the international jihadist movement, he had notably spent time in Afghanistan in 2010 and Pakistan in 2011. 

The Limitations of Counterterrorism

In the future, the American strategy might well be limited to counterterrorism. But the results of such a strategy will likely be mixed and at times counterproductive. Far from offering greater autonomy to American policy, counterterrorism is a technical instrument that increases reliance on local allies. Counterterrorism operations rely on the nations where they are carried out. And in this case, the jihadists acting in Afghanistan are based in Pakistan. The operation that led to the death of Osama bin Laden by circumventing the Pakistani government is an exception to which Pakistan reacted quite negatively. Despite the fantasy of a “purely” technological war without political dimensions, drones do not make it possible to bypass national alliances. If it wants to continue using drones and carrying out nighttime raids, the United States will have to rely on the Pakistani and Afghan governments for intelligence and logistics.

Over the next two years, the United States will progressively lose its ability to influence the Afghan regime and will become dependent on Pakistan.

The need to use Pakistani territory to evacuate coalition equipment makes it difficult at this stage to put decisive pressure on Pakistan. This reliance has a real political cost, and in practice, it will be Pakistan that defines potential targets in the future. Over the next two years, the United States will progressively lose its ability to influence the Afghan regime and will become dependent on Pakistan. Logistically, withdrawing forces will have to go through Karachi, and drone operations depend heavily on Pakistani intelligence.

Furthermore, counterterrorism operations are a source—probably the most important source—of anti-American sentiment in the region. Whatever the real level of civilian losses incurred during the operations, the general perception is clearly one of indiscriminate strikes against the population. This is important, because this sentiment facilitates recruitment for jihadist movements and to a certain extent paralyzes the Pakistani government.

In addition, these are not only counterterrorism operations but also counterinsurgency operations targeting both transnational jihadist groups and the Taliban—which is possibly the biggest problem. It is very difficult for the Afghan government to conduct these operations without active support from the coalition, but the use of counterterrorist techniques like drones and raids against insurgents facilitates the insurgents’ alliance with jihadist groups on the ground. The danger is therefore that the Taliban, which until now had been an essentially Afghan movement, will strengthen its ties with jihadist groups that have a more global agenda.

The only coherent response now is for the coalition to reach an agreement with the Taliban in order to detach them from transnational jihadist groups.

The coalition can no longer defeat the Taliban, which will remain a political and military power in Afghanistan for the foreseeable future. With the reestablishment of a sanctuary for jihadist groups, the only coherent response now—at least theoretically—is for the coalition to reach an agreement with the Taliban in order to detach them from transnational jihadist groups. And if the alliance between the Taliban and jihadist groups is effectively non-negotiable, the United States does not have the appropriate military instruments or obvious solutions to the problem of a reconstituted sanctuary on the Afghanistan-Pakistan border. Then, the argument that the Taliban would not be ready to negotiate is not particularly pertinent, because the United States would have to negotiate. The alternative would be for the United States to pursue a policy of pressure through drone attacks, which suppose a high level of dependence on Pakistan and probably mixed results on the ground since the population is more and more opposed to the strikes, with all of the attendant negative effects.

After 2014, however, negotiations will largely be out of the coalition’s hands. From now on, U.S. leverage over Pakistan and the Afghan parties, which is already limited, will swiftly decrease. Regional actors that sponsor Afghan groups will play a key—and probably negative—role in the next phase of the civil war. It is hard to see how Pakistan and India in particular could find common ground on Afghanistan. The last chance to exert influence in a way that is favorable to U.S. interests is therefore in the fall of 2013, but the regime must first prove itself sufficiently capable of resisting the insurgency’s pressure. The Afghan National Security Forces’ ability to hold on in the east will be critical.

Given the current circumstances, the United States must begin to consider the situation that will be created by the fall of the Afghan regime and the Taliban’s rise to power. Negotiations, if they ever take place, could be between the Taliban in Kabul and the Western countries. Alienating key insurgent groups, by, for instance, placing the Haqqani group on the U.S. list of terrorist movements, is thus counterproductive. The move will have few immediate practical consequences while possibly creating future deadlocks with the entire Taliban movement. The most dangerous situation would be the Taliban’s rise to power in association with jihadist groups with a more global agenda and the (re)establishment of a pariah state on the international scene. Beginning in 2014, the use of drones and nighttime raids must be limited to transnational jihadist groups to send a clear message that the United States targets the Taliban only when they collaborate with jihadists that pose a direct threat to Western countries. 

Toward a Compartmentalized Strategy

After 2014, the level of U.S. support for the Afghan regime will be limited and, after a new phase in the civil war, a Taliban victory will likely follow. The United States does not have a long-term interest in the survival of the current powers in Kabul because they cannot deliver on key demands—securing the border with Pakistan, fighting jihadist groups, and limiting drug trafficking. Even a (relatively) hostile new Taliban force in Kabul will be easier to deal with because, since they will have established their control on the Afghan side of the border, they will be directly responsible for key security issues.

During what should be seen as an intermediate period—between the withdrawal and the Taliban takeover—U.S. options are limited and the administration should focus on preparing for the next phase. There are a number of steps Washington should take:

First, limit drone strikes to jihadist groups and primarily al-Qaeda. Using drones against the Taliban is counterproductive.

Second, avoid anything that could limit the ability of the next administration to open negotiations with the Taliban when they will be in Kabul. Putting the Haqqani group on the U.S. terrorist list was counterproductive in this regard.

Third, define a long-term regional policy since the current situation—three contradictory approaches toward Afghanistan, Pakistan, and India—is not sustainable. The new regional framework should take the form of a “compartmentalized strategy” that emphasizes the fact that the three countries can deliver very different things—from cooperation in counterterrorism to a strategic alliance—and there is no possibility of an integrated approach, especially because the United States lacks leverage over the regional players.

The desirable endgame should be a stabilization of Afghanistan, probably with the Taliban in Kabul. There would have to be a measure of political or economic support from the United States because a difficult relationship between Afghanistan and Pakistan is very likely whatever the regime in Kabul. That is essentially the best situation from a U.S. point of view. A difficult relationship between Afghanistan and Pakistan will give the United States more leverage on both and offer some guarantees that the border will not be totally out of control. The United States should also limit its cooperation with Pakistan on terrorism to the exchange of information and put an end to the (especially military) aid it gives the country.

The United States should limit its cooperation with Pakistan on terrorism to the exchange of information and put an end to the (especially military) aid it gives the country.

Washington should also avoid over-the-top declarations about the necessary convergence of Indian and U.S. interests and define cooperation on more pragmatic grounds. In the broader view, the U.S.-Indian relationship is more important to counterbalance China, but counterterrorism policy requires a certain measure of collaboration with Pakistan. A careful approach focused on a medium-term deal with the insurgency is quite different from the current Indian policy. India is intent on supporting the Afghan regime until the end, hoping that an ongoing civil war in Afghanistan will distract the Pakistani military from the eastern front. The U.S. focus on its enemy defined in a very narrow sense means that India will be in a somewhat uncomfortable position vis-à-vis the United States in the next few years.


1 Twenty-seven percent support the war compared with 66 percent who strongly oppose it, according to the AP-GfK poll. Forty percent are strongly opposed to the war compared with 8 percent who strongly support it. See Mackenzie Weinger, “Poll: Afghan War Backing at New Low,” Politico, May 9, 2012,

2Which validates a posteriori the proposals made in a previous report, “Focus and Exit,” which suggested controlling the “useful” Afghanistan at a lower cost and opening negotiations. See Gilles Dorronsoro, “Focus and Exit: An Alternative Strategy for the Afghan War,” Carnegie Endowment for International Peace, Policy Brief, January 2009,

3Total coalition losses in 2010 amounted to 711 and have declined sharply since then, with 181 casualties in the first five months of 2012 ( About 3,021 civilians were killed in the violence in 2011; see UNAMA, The end of the surge saw a spectacular decrease in losses: 21 percent fewer civilians killed during the first four months of 2012. See

4About a dozen per night in 2012. For targeted eliminations and the drafting of target lists see “‘Capture or Kill’: Germany Gave Names to Secret Taliban Hit List,” Der Spiegel, August 2, 2010,,1518,709625,00.html.

5ANSO, Q1 2012 report,

6See SIGAR (Special Inspector General for Afghanistan Reconstruction), Ten Years of Reconstruction 2001–2011, Quarterly Report to the United States Congress, January 30, 2012, 104.

7Michael Hirsh, “Meet Afghanistan’s Greatest Hope: ‘The Re-Integree,’” National Journal, May 8, 2012,

8Kim Sengupta, “The Taliban Are Divided and in Retreat, Insist Top U.S. Officials,” Independent, May 9, 2012,

9Human Rights Watch, World Report 2012: Afghanistan,

10Joshua Partlow, “Afghan Police Casualties Soar,” Washington Post, August 31, 2011,

11 See “Attacks on U.S and NATO Soldiers by Afghan Security Forces” (map), New America Foundation,

12See map attached to the International Crisis Group’s report, The Insurgency in Afghanistan’s Heartland, Asia Report no. 207, June 27, 2011.

13In seventeen (out of 34) provinces, the Taliban will be in a position of strength after the withdrawal: Farah, Nimroz, Helmand, Kandahar, Uruzgan, Zabul, Ghazni, Paktya, Paktika, Wardak, Logar, Nangarhar, Laghman, Kapisa, Nuristan, Kunar, and Khost. Concretely, it will be difficult for regime officials to travel outside of the provincial capital. In twelve provinces—Badghris, Baghlan, Takhar, Sar-i Pul, Samangan, Kunduz, Kabul, Balkh, Faryab, Jawzjan, Parwan, and Herat—the government can contain the Taliban, but they control at least one district and will be able to operate militarily in most of the province. The province of Ghor is an open space, without government control; the province of Badakhshan is an area with influential fundamentalist movements but for now, at least, they have not taken up arms. In Panjshir and the Hazara-populated regions (spread out over several provinces), the Taliban presence is marginal or nonexistent.

14The cooperation agreement between India and Afghanistan, signed on October 4, 2011, covers different areas including security.

15International Crisis Group, A Force in Fragments: Reconstituting the Afghan National Army, Asia Report no. 190, May 12, 2010.