An unforeseen result of the U.S. military's stunning success in Afghanistan was the overnight suspension of that country's vicious, 23-year-old civil war. Afghanistan's future—including whether it again degenerates into a terrorist base—now largely depends on what is made of this precious opportunity.

In countries recovering from civil war, the most critical requirement for long-term peace is the demobilization of the formerly warring parties and their integration within a unified military. Angola and the former Yugoslavia provide cautionary tales about the difficulties of military reintegration; Mozambique and South Africa give more hopeful examples of how building a cohesive army can help solidify peace after a national conflict.

In Afghanistan, the process of military integration has barely begun, but it is already close to collapse. Not only are perennial ethnic, factional, and religious disputes hampering progress, but the political elements of postwar transition are moving ahead without the requisite military corollary. Indeed, the interim administration inaugurated in December 2001 never answered basic questions about the size, composition, and tasks of a national army. Meanwhile, the international community remains ambivalent about how it will assist, and what little aid it has promised has been slow in coming.

The dangers of continued delay are growing by the day. The U.S. and allied forces entered Afghanistan to rout the Taliban and al Qaeda; demobilizing the country's many warring factions was not on the agenda. Thus, the operations may have abruptly suspended the civil war, but they have created only a tacit truce without dismantling the full war-fighting capabilities of the armed groups. Many of these groups may now be tempted to either reject the peace process or manipulate it to their advantage. If they do, Afghanistan could plunge straight back into war.


Civil wars can yield three types of disgruntled local parties, or "spoilers," who can derail peace processes. "Limited spoilers" are simply suspicious of promises made by the peace brokers and demand additional guarantees that they will be treated fairly; "greedy spoilers" seek to take all they can get from the postwar reconstruction, even beyond the point of diminishing returns; and "total spoilers," feeling they have no stake in the peace, will try to make it fail at all costs. Unfortunately, Afghanistan today contains archetypes of all three.

In addressing these spoilers, the new national government will need to exert its leadership over a nation in which mistrust of central authority runs deep. Afghanistan as a state was created by late-nineteenth-century British imperialists along borders that, like most colonial divisions, reflected little historical or ethnic logic. The government has usually been controlled by the largest ethnic group, the Pashtuns, who nevertheless make up fewer than half of Afghanistan's roughly 26 million people and are themselves riven by tribal fissures. Hobbled by the Pashtuns' own divisions and opposition from other minority groups—such as Tajiks, Uzbeks, and Hazaras—the government's power outside of the capital, Kabul, has always been limited.

The long history of strained relations between Pashtuns and other ethnic groups makes many Afghans natural limited spoilers. Indeed, they have good reason to view promises of peace with skepticism. Most recently, ethnic relations deteriorated under the predominantly Pashtun Taliban, who came to power pledging to end the post-Soviet chaos and warlordism but instead exhibited equally vicious behavior themselves. The U.S. military intervention last fall then tipped the scales of power in favor of the Taliban's main foe, the Northern Alliance, composed mainly of minority ethnic groups from the north. The interim government established at a UN-brokered peace conference was nominally led by the Pashtun noble Hamid Karzai, but the non-Pashtun leaders of the Northern Alliance wielded the real power in Kabul, with a subset of Tajiks controlling key ministries and the former secret police. Many ordinary Pashtuns thus suspected that the interim government was just a vehicle for minority ambition.

Years of warfare have also created a constellation of regional warlords, quintessential greedy spoilers, who stand to lose a great deal in the transition to a new government. These warlords' power comes from their personal forces of thousands of loyal armed troops, funded by their control of local trading and smuggling routes. Although many of the rival warlords made public statements supporting military integration after the interim government took over, none has made any significant effort to disarm. Several, despite their uneasy truce, have faced off against each other in minor struggles over territory and power. These greedy spoilers remain only nominally linked to the central government and may try to undermine it if they are not given a significant role.

Finally, neither the Taliban nor al Qaeda is a spent force. Few senior leaders of either group have been captured, and both groups still enjoy popular support among many Pashtun nationalists in southern Afghanistan and the tribally administered border regions of Pakistan. With no possibility of inclusion in a future government, these groups can be expected to act as total spoilers, seeking to attack the regime in Kabul whenever and wherever they can. One of the current major security concerns is the possibility of a Tet-like counteroffensive against the government, just when it begins to feel secure.


Unfortunately, international help in building a new Afghan army has been limited—and the little offered has thus far been so disorganized that it may only make matters worse. The first effort, led by the UN-mandated International Security Assistance Force (ISAF) in Kabul, recruited and trained an ethnically balanced unit of 600 soldiers drawn from most of Afghanistan's 33 provinces. But the program is only a stopgap measure. By the time the ISAF withdraws at the end of the year, it will at best have trained a mere 4,000 troops (compared to the tens of thousands that will be needed to provide a minimum level of security). Moreover, the exact lines of authority remain unclear.

The United States, in turn, has said repeatedly that it views the building of a national Afghan army as a priority. Many U.S. officials believe such a force could substitute for a large international peacekeeping presence, which they oppose. Accordingly, the Bush administration submitted to Congress in March a $50 million proposal to train and equip 18,000 Afghan soldiers over 18 months, and U.S. forces began some of the basic training pending congressional approval. But it is unrealistic to expect to build an effective, complete force in such a short time. Even 18,000 troops with only basic training, moreover, will still be insufficient to meet the country's security needs.

An added snag is that U.S. special forces have also begun to train and fund separate "anti–al Qaeda" units, often associated with local warlords, to act as American proxies and seek out al Qaeda fugitives in the Pashtun regions of southern Afghanistan. The formation and operation of these units have not been coordinated with Kabul, and, so far, there are no formal plans to integrate them into the future national army. Yet the higher pay in the U.S. units will continue to attract potential recruits away from any force directed from Kabul—a fact that has understandably caused some concern among the Afghan leadership.

As a result, the motley Afghan force currently taking shape is wretchedly small, disorganized, and not clearly linked to an established command structure that represents Afghanistan's ethnic diversity. Meanwhile, no coherent plan has emerged to either integrate or demobilize the assorted warlord forces, nor are the various spoilers being dealt with decisively. If things continue in this way, the present dream of a self-reliant army serving the Afghan people seems almost certainly doomed to failure.


The ultimate success of Afghan military reintegration depends on the political will of the local parties. This will, however, cannot form in a vacuum. The UN, the United States, Europe, and other interested parties urgently need to move the process along. They can do this through four particular types of intervention: first, by pushing the parties to determine quickly, while a tacit truce exists, what type of national military they need; second, by providing critical financial backing; third, by committing themselves to remain engaged and help create an atmosphere in which integration can succeed; and finally, by carefully coordinating the various military aid efforts so they do not end up working at cross-purposes.

Any civil war leaves unresolved grievances and mutual suspicions that linger long after the fighting has ended. During peace negotiations, therefore, both the warring parties and the international community are often tempted to defer the complex issues associated with military reintegration; they hope that the former adversaries will become more reasonable as tensions dissipate. But avoiding thorny questions in this way is generally a strategic mistake. In Bosnia, Angola, Cambodia, and Kosovo, peace negotiations produced only the general outlines of a new, unified military. In all these cases, the lack of detail has since led to frequent misunderstandings and heated arguments that have either stalled or completely stopped the process of military integration.

By contrast, during South Africa's transition from apartheid, military research groups from the forces that had been opposing and defending apartheid, respectively, spent several months working together to outline their views on a joint military. Political leaders on both sides then used these proposals to create a detailed, mutually satisfactory integration program. The process helped all parties adjust to the impending changes and taught the military leadership how to work together constructively, thus creating role models for the ordinary soldiers. Similarly, the peace agreement that ended Mozambique's civil war in 1992 laid out the size and command structure of a new professional army, as well as a detailed time-line for the demobilization of the former combatants. Although it may not be possible to replicate these processes exactly in Afghanistan, efforts must be made to settle the details of military reunification—now.

Regrettably, the international community and the Afghan leadership have not heeded history's lessons. As of this writing, only a single, one-day meeting had taken place among all the major parties to negotiate the future of the Afghan armed forces. Although the gathering of warlords on March 6 brought patriotic speeches proclaiming each faction's good intentions, it did not answer any of the fundamental questions.

These questions—the military's role, command structure, and force size, as well as the demobilization of the various standing forces—should be addressed immediately in formal negotiations involving all the Afghan parties. These all-important meetings could be mediated by the UN as a continuation of the Bonn agreements that brought the interim government into being. Although the international community cannot ultimately control this process, it should use its lobbying power, including promises of aid, to shape opinions and induce the parties to reach agreement on fundamental issues.

Even after these meetings conclude, however, the job is far from complete. Outsiders can play an important role in generating trust among former enemies and strengthening their overall commitment to the peace process. Too often, outside powers squander these opportunities by promising too much and then failing to deliver, or by allowing their own disagreements to undermine compliance.

Financial assistance is a prime example. At a donors' conference in Tokyo in December, the assembled governments completely ignored the need to fund an integrated Afghan army. A follow-up conference in Geneva in April did yield pledges of $235 million for this cause, but few of the funds have materialized. Yet lack of security is the major source of instability and human rights violations in postconflict situations. Only "fixing" the security sector can generate the confidence necessary to complete the peace process. Adequate, timely funding, with its distribution monitored by the international community to minimize corruption and waste, is required.

As part of the Afghan aid package being debated in Congress, the United States should double its proposed military assistance with an initial, immediate grant of another $50 million (for a total of $100 million, the same amount spent by the United States on a similar program to train and equip the Bosnian army). At the same time, Washington needs to call on other countries to follow through on their pledges and seek to involve U.S. allies in the Islamic world in the process. All the international pledges should be centralized in a fund administered by a joint military commission composed of Afghan parties and members of the donor community, to ensure that the money is properly managed and that neighboring states do not use their gifts of aid to foment dissent.

The required amounts may be small in comparison to the $17 billion spent on Operation Enduring Freedom by mid-May, but they could determine its ultimate legacy. Outside assistance is not only needed to solidify the future of the Afghan army, and with it the Afghan state, but it also offers an excellent insurance policy against the country's ever harboring terrorists again. Furthermore, it provides a concrete example to the Islamic world that the United States does not abandon moderate Muslim regimes that stand with it against terrorism.

Aid money will be ineffective, however, if the peace process does not win the support of the Afghans themselves. It will be much more difficult to sustain a peace agreement, no matter how well thought out or well funded, if no outside enforcement mechanism exists to check spoilers and arbitrate disputes. In Bosnia, for example, the presence of an international military force has been absolutely critical to keeping potential spoilers in line and monitoring compliance with the Dayton accord.

Yet the extended deployment in Bosnia has become political ammunition for those opposed to expanding the ISAF beyond Kabul. Several senior U.S. officials have been adamant on this point, arguing that Afghan security is best handled by a new, national Afghan army. Unfortunately, that arrangement leaves a dangerous gap in time: recruiting, training, and equipping an army capable of providing real security will take several years. In fact, expanding the peacekeeping force could even speed up the creation of an Afghan army. The peacekeeping forces in a number of countries, including Bosnia, have acted as valuable stabilizers and training partners for young local armies. On this evidence, many argue that a wider international force presence is exactly what is needed to convince local warlords to cooperate.

But with the expansion of the ISAF now seemingly ruled out, the United States must consider stopgap measures to bolster Afghanistan's security—beyond just hunting down the Taliban and al Qaeda. The Pentagon should continue posting advisers with local warlord forces and should sanction air strikes against any greedy spoilers who take up arms to disrupt the peace or challenge the government. In early May, for example, the CIA sent an unmanned spy plane to kill Gulbuddin Hekmatyar, leader of a hard-line Islamic group and a former warlord who was reportedly planning attacks against the Karzai administration and U.S. troops in Afghanistan. The mission to kill him may have failed, but the right message was sent.

Finally, the international community must better coordinate its presently disparate military training programs. Foreign military educators can help former adversaries overcome their mutual distrust, teach responsible civil-military relations, and instill a professional ethic in Afghan soldiers, many of whom have received no formal training. The limited steps taken by the ISAF and the United States toward these goals are positive—but they also contain seeds of trouble. The ISAF and the U.S. forces must coordinate their army training programs both with each other and, most important, with the formation of anti–al Qaeda strike forces in the south. To avoid these local units' adding to the rogue armed forces that pervade Afghanistan, the strike forces—like the U.S.- and ISAF-trained battalions—should be brought under the command of the Afghan government as soon as possible.

A novel option worth exploring is to contract the job out. Such a move would solve any concern over detracting from the peacekeeping or antiterrorism effort. Instead, it would bring all the international programs under the management of a private military consulting firm. A number of such consulting companies, staffed by retired military personnel, have experience in advising nascent militaries and useful expertise in running officer academies and training centers. The U.S.-based firm Military Professional Resources, Inc., for example, presently runs such an operation in Bosnia aimed at binding Bosnian Croat and Muslim units together into a capable national force. Local acceptance of the program has also been critical in weeding out the negative influence of military assistance from radical states such as Iran. If the privatized option is chosen for Afghanistan, however, it must be carefully monitored. The contract should include provisions for a gradual turnover of authority to local officers—otherwise, the firm could seek to profit from long-term local dependence on its services.


If a disciplined standing armed force is not built, Kabul will have to continue relying on the Northern Alliance's loose amalgamation of warlord-controlled forces. This outcome is hardly optimal. The Northern Alliance, a short-term coalition at best, does not represent the country as a whole. Rather, it reflects the agendas of its individual warlord components, which do not always intersect with that of the broad-based national government.

The government needs a new national army, instead, to come to terms with the spoiler problem—or otherwise risks scuttling the peace. The army must be able to reassure and incorporate limited and greedy spoilers, giving them incentives to cooperate and thus creating some breathing room for the nascent central government. At the same time, the force must also be strong enough to deal effectively with total spoilers, such as al Qaeda and the Taliban remnants, who may challenge the government in the future.

Given the warlords' deep-rooted hold over local power structures, the government probably could not crush them—indeed, it would be injudicious even to try. They must instead be convinced to play along. The government needs to show the warlords that they will have a role in the new government and military if they cooperate. Yet the country's leaders must also stand tough and enforce the current tacit cease-fire, calling on outside assistance as necessary, to forestall any jockeying for position among the warlords that could disrupt the peace. Running all political and military aid through Kabul will give it important leverage in this process.

In light of this situation, the mission of an integrated Afghan military should be limited. In the short term, it should focus on certain core sovereign tasks: defending Afghanistan's borders against foreign fighters and preventing terrorists from using the country as a base; heading off renewed fighting between political or tribal factions; and protecting the major lines of communication and commerce.

The new army should avoid the numerous other missions that some have proposed for it, from weeding out criminals and smugglers at border crossings to enforcing bans on poppy production. Not only is it unrealistic to expect a new force to take on so many tasks, but these additional duties would blur the traditional distinctions between military and police functions and could even harm the reintegration process. Targeting the major income sources of local power brokers may also create a showdown that the central government is not yet politically strong enough to win. Instead of compromising the national armed force's legitimacy with these explosive domestic issues, Kabul and the international community should focus on establishing effective local law enforcement to take care of such matters.

A better alternative might be to focus the new military's energies on other, more positive operations as a means to help solidify public support, while at the same time building force cohesion and experience. These auxiliary duties would include weapons disposal, demining, and support for disaster relief operations. Similar exercises were used successfully in El Salvador and South Africa.


Ideas for the size and form of the new Afghan armed forces vary significantly. Some, such as Interim Defense Minister Muhammad Fahim Khan, have proposed creating a national military that combines all the warlords' militias and other standing armed forces. This type of centralized force failed during the Soviet occupation in the 1980s and remains unrealistic today. Undeterred, Fahim has publicly stated that he envisions an army of 200,000 troops, almost twice the size of the Soviet-era force. This proposal ignores the incredible financial burden that a large national army places on a nascent state, at the very time it is least able to afford it. In Mozambique, for example, officials decided to integrate all armed units into a large national force, and to demilitarize later. Political leaders overestimated the resources available to fund this force, however, and troops did not receive their promised pay. The resulting dissension within the force led some parts of the new military to revolt, provoking a renewed crisis that the new government barely withstood. Maintaining a centralized force of the magnitude Fahim has proposed, moreover, would eventually require some form of national conscription. In the past, this proved highly unpopular and undercut public support for both the communists and the Taliban, a mistake the new government would do well to avoid.

Even a somewhat smaller central army of roughly 60,000 troops, as force planners in the United States have proposed, is unrealistic and does not address the spoiler problem. Simply recreating a national army strictly controlled from Kabul will do little to convince individual tribal or ethnic groups that the central government is not seeking to dominate them. In fact, if the interim government's composition offers any guide to Afghanistan's eventual leadership, a national army responsible solely to the central government would almost certainly be dominated by the Tajik-controlled Northern Alliance—and thus would pose a direct challenge to local power bases. This situation could provoke both Pashtuns and some warlords to revolt against the central government in the near future, as they would seek to strike while the government was still weak.

A third suggestion is that the new armed force should not be a military at all. Instead, it should fall somewhere between an army and a police force, akin to Costa Rica's small, paramilitary-style national guard. It would focus exclusively on domestic security and have no heavy weapons, so as not to threaten the authority and clout of regional power brokers. This solution, however, underestimates the strength and determination of the new government's potential opponents. Carrying only small arms, the paramilitary force would be outgunned by the full range of potential spoilers—holdout Taliban or al Qaeda bands, warlord armies, and even individual criminal bands—and unable to integrate the warlords' standing forces.


There is a different solution, however. A force that incorporates the best elements of the proposed models while mitigating their failures could resolve the problems presented by all the potential spoilers. Such a force would have two components: first, a small, ethnically integrated professional army answerable to national political authorities; and second, a "national guard" that incorporates tribal and warlord militias (and in the south, the U.S.-trained anti–al Qaeda troops) into formal units responsible to provincial governments. This solution avoids the pitfalls of the other options on the table by balancing Kabul's various needs: a professional army capable of guaranteeing the state's sovereignty and suppressing any total spoilers; integrating the various armed, greedy spoilers; and assuaging the concerns of potential limited spoilers.

In terms of size, the new army should be large enough to fulfill its duties but small enough to facilitate the army's rapid professionalization and help assure individual ethnic and tribal groups that the army will not be used as a means of conquest or control. A force structure of roughly 30,000 troops would fit these requirements: six maneuver brigades, four light and two mechanized, supported by artillery, engineering, and logistics assets. This force mix, essentially a smaller version of Afghanistan's pre-civil war military, would be mobile, able to operate across Afghanistan's varied terrain, and equipped with enough firepower to deal effectively with likely foes. It would not, however, be heavy enough to be used as an occupying force if hijacked by any one ethnic group.

Recruitment should occur through the mechanism that Karzai's interim administration established for the ISAF programs, drawing from each of Afghanistan's 33 provinces. All new recruits should then train together in ethnically integrated units. The officer corps should be built in a similar manner, with each governor delegating a small number of officers to train at a central academy that incorporates all the ethnic groups.

The ethnically integrated brigades should be based near the major urban areas (Charikar, Gardez, Herat, Jalalabad, Kabul, Kandahar, Kunduz, and Mazar-i-Sharif), to help the government establish national authority without directly encroaching on the fiefdoms of rural leaders. The basing of units or individual commanders should be rotated on a regular basis to prevent the development of overly close ties between particular military units and local political authorities—a strategy used successfully in a number of African states.

For this army to be accepted and trusted by all Afghans, its command structure must be broad-based and unified. A senior leadership drawn from or structured along purely ethnic or partisan lines would make excluded groups suspicious of the army and the regime behind it—at the potential expense of future peace. If a political dispute arises while the new force is being put together, disgruntled groups will then be able to walk out of the entire process with their armed units behind them. The example of Angola, where a failed military integration led to another decade of warfare, stands as a clear warning.

The initial wave of generals appointed to lead a new Afghan army does not bode well in this regard. Of the 100 or so generals named by Defense Minister Fahim during the interim administration, about 90 are Tajiks hailing from the relatively small Panjshir Valley, north of Kabul. Yet this unfortunate beginning does not have to doom the whole project. Like all appointments made by the interim government, these selections should be reassessed under the 18-month transitional authority appointed by the Loya Jirga (the traditional Afghan assembly). A little finesse should be used in shifting slots and considering future appointments, aiming to restructure the senior military leadership so that it better reflects Afghan demographics. One possibility is to move a portion of these officers into other, potentially more lucrative government bureaucracies.

As is the practice in Western militaries, national political authorities should appoint the senior military staff, while the military bureaucracy should nominate individual unit commanders. The new army may initially have to be relatively top-heavy. This may not be the most militarily efficient structure, but the creation and disbursement of a greater number of senior slots will provide Kabul with valuable political chips to bring all parties, including the warlords, to the table. After the critical postwar period has passed—most likely at least three to five years down the road—the government can slim down the command structure to better reflect professional competence.

This smaller national army structure takes a significant step, but does not go all the way, toward dissipating any perceived threat posed by a central government. Alone it will not convince the major power brokers to accept military integration. This goal, however, could be achieved fully if the force were complemented with regional "national guards," regularized units that bring the existing warlord-dominated militias into the fold. Structured at the province level, these guards would be modeled after the various contemporary and historical frontier militia forces in neighboring countries, such as the Rangers and Khyber Rifles in Pakistan and the Assam Rifles in India. The local political authority would recruit and command these guards, who would handle paramilitary duties such as protecting major public installations and lines of communication. Basic law and order should remain the task of civil police forces.

The formation of provincial militia units provides a further means to incorporate recalcitrant warlords into the political structure. If carefully coordinated with other demobilization efforts, the new militias would become one alternative means of peaceful, gainful employment for the masses of fighters currently employed by the warlords. The regional command structure would also formally delineate the lines of control that are presently a major source of contention. Finally, the presence of provincial guards loosely based around an ethnic identity would offer local communities a sense of assurance that their own kin are guaranteeing their security.

With initial financial help from abroad, the central government would ideally provide the largest part—if not all—of the funding for these regional units. This financial leverage over the warlords would allow Kabul to lock them into an implicit bargain: the government keeps the money flowing and the warlords get a seat at the table, as long as their forces keep the peace and do not engage in predatory activities aimed at the local populace. The sides would gain even greater confidence in each other if the deal also included provisions for the cantonment of heavy weapons.

In the beginning, the Pentagon and the ISAF should be prepared to send military advisers (including perhaps the same special forces officers who have already built relationships with local warlords) to aid these regional units. Serving as executive officers (the second-in-command), these advisers would provide structure and professionalism as well as a neutral presence to guarantee that all sides meet their commitments. The anti–al Qaeda training program should also be expanded and amended to teach the regional troops basic professional skills. After the situation has normalized, the executive slots for the foreign military advisers could be replaced by a rotation of officers on temporary transfer from the Afghan national army, as is the practice in Pakistan and India.


With the removal of the Taliban, the pause in internecine fighting, and the pledges of aid and engagement by the international community, the prospects for peace in Afghanistan appear more positive than they have for decades. Peace, however, is by no means certain.

Whether the challenges of military integration are resolved will determine the success or failure of the entire peace process. Getting this task right requires action from international parties in four critical areas: helping to specify what shape the military should take before a final peace agreement is signed, dedicating a portion of the international aid package toward army integration, creating an atmosphere in which integration can succeed, and coordinating the presently disjointed military training programs. The international aid funds must be used to establish a military structure that is integrated and effectively responds to the spoiler problem. If these relatively limited and entirely achievable steps are taken, the prospects for peace in Afghanistan may have a chance. If not, yet another opportunity to restore hope to the Afghan people will have been lost.

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
  • ANJA MANUEL is an Attorney with Wilmer, Cutler & Pickering.
  • P. W. SINGER is Olin Fellow in the Foreign Policy Studies Program at the Brookings Institution and Coordinator of the Brookings Project on U.S. Policy Towards the Islamic World.
  • More By Anja Manuel
  • More By P. W. Singer