The tired debate between those who believe in nation-building and those who scoff at it glosses over a major difference between top-down and bottom-up society-building. The starting point for a bottom-up approach is the communitarian recognition that societies — even modern, so-called “mass” societies — are not composed of just millions upon millions of individual citizens. Instead, most societies are communities of communities. Most people come in social packages. They are greatly influenced by the communities of which they are members and by their natural leaders. These communities are not necessarily residential — the traditional village — but may be ethnic, religious, or based on national origin.
This is not to suggest that individuals do not have degrees of freedom or that their behavior is determined in full by their communities. It merely points out that communities have a profound effect on what seem like individual choices, from voting to purchasing to eating and beyond. Moreover, American national society was formed to a significant extent only after the Civil War. Before that, most Americans’ prime loyalty was to the colony, state, or region in which they lived. When asked overseas, “Where are you from?” Americans used to answer, “I am Virginian” or “I am Bostonian.” Only after the 1870s did more and more Americans respond, “I am American.” Only during the Reconstruction period did the Supreme Court stop referring to the United States as a plurality (“The United States are”) and start referring to the nation as a singular entity (“The United States is”). It took a very bloody war and a generation of society-building afterwards to make the South and the North into one political community — a process that is still ongoing. Many other countries we now know as nations were similarly cobbled together, including the United Kingdom, Germany, Italy, and Switzerland.
I point to these familiar pieces of sociological history because we tend to ignore them when we deal with countries that have not yet made much progress along these society-building, communitarian lines — whether or not they are called nations or have flags, seats at the United Nations, and diplomatic representatives in the capitals of the world. For instance: Iraq, Pakistan, and above all, Afghanistan.
Think tribes, not a nation
Given the power and import of communities (often referred to as “tribes”), the issue here is not whether we can or should avoid engaging in nation building, but how we proceed. Do we make our starting point the notion that there is a central national government, whose troops and police we can train as a national force and whose administration of justice and social services we can improve? Or, do we realize that such a center-to-periphery approach is unworkable, and that we need to build from the periphery to the center? This does not mean that we should go find individual citizens to “empower” and work with them. Instead, we should look at places like Afghanistan as lands in which several tribes lie next to each other. (I use the term “tribes” loosely, referring to ethnic and confessional communities whose members have tribe-like ties to one another, ties they do not have to members of other communities.) In other words, there are many societies in which nation building cannot start from the center — and those who insist otherwise pay a heavy price.
Howard Hart, a member of the cia clandestine service for 25 years and former cia Chief of Station in Islamabad, observes that “Afghan” is purely a geographic distinction and that “there is not, and never has been, anything remotely approaching a shared national identity.” He finds that tribal loyalties in the region are paramount, and “warlords” (or tribal chiefs) are loath to subordinate themselves to a higher authority, especially one fostered by foreign powers. I would temper this argument a bit and suggest that these tribal leaders have some sense of national citizenship. However, and this is the crucial point, when their tribal loyalty conflicts with their national one, their tribal loyalty tends to prevail. J. Alexander Thier, the Director for Afghanistan and Pakistan at the U.S. Institute of Peace, and Jarat Chopra, a former professor at Brown University, write that in Afghanistan, “family and tribal affiliations outweigh all others.” And according to a recently published article in Special Warfare, a quarterly of the United States Army, “[tribal elders] are not willing to place a united Afghanistan over advancement of their particular tribe.”
This tribal primacy is in part due to the fact that many of these nations were created by foreign powers drawing lines on a map that clumped several tribes, who had little in common and much separating them, into a single territory. These tribes were not asking for a more encompassing community — for a nation — but were forcibly joined by colonial powers. Moreover — as in several countries that we now consider well-established nations — the tribes are now vying with one another, in part through civil wars, for control over the nature and leadership of the union-to-be. I am not saying that such strife is to be welcomed or that, to the extent possible, one should not seek to find nonviolent ways for, say, the Shia and Sunni or the Pashtuns, Tajik, Uzbek, and Hazara to work out their differences and form stable communities of communities. However, anyone familiar with the history of Western nations should not act as if these nations did not pass through similar violent phases.
Design versus trends
In discussing the ways in which Afghanistan and other such unstable states can become the kind of nation that we hold is best for them (and us), we tend to start with what political scientists call “design questions.” Implicit is the notion that we are builders and have the needed raw material to build regimes following a blueprint we lay out. Typically, we seek a modern state with a strong national army and police force, a low level of corruption, a proficient civil service, a trustworthy justice system, democratic institutions, and open markets. And we take it for granted that loyalty to the central government will take precedence over loyalty to any member community, to any tribe.
This position is sometimes associated with the policy of nation-building that Governor George W. Bush scoffed at and then promoted once he became the president of the United States, pursuing it relentlessly in Iraq. President Obama’s first strategic review of the situation in Afghanistan, completed in March 2009, set much more limited goals, namely “to disrupt, dismantle, and defeat al Qaeda in Pakistan and Afghanistan and to prevent their return to either country in the future.” However, the advocates of nation building soon interpreted this as necessitating winning the hearts and minds of the population, which in turn requires providing them with security, economic development, and an efficient, functioning government.
One of the most important advisers to General Stanley McChrystal, Anthony H. Cordesman, a senior scholar at the Center for Strategic and International Studies, calls for the U.S. forces “to ‘hold’ and keep the Afghan population secure, and ‘build’ enough secure local governance and economic activity to give Afghans reason to trust their government and allied forces.” Representative Jane Harman, a leading member of the House Intelligence Committee, believes that “it’s too early to abandon a strategy focused on protecting the population and rebuilding the country.” General McChrystal believes that we must “ . . . promote good local governance, root out corruption, reform the justice sector, pursue narcotics traffickers, [and] increase reconstruction activities.”
Instead of starting with such a “design” approach, though, we must first ask what the reality on the ground looks like and in what direction it is changing before we intervene. The main reason we must adopt this focus is that our leverage is much smaller than we tend to imagine. Hence, we must focus the use of those resources that are available to ride trends that are unfolding anyhow and seek to redirect them somewhat, rather than seeking to fashion new ones out of whole cloth. Thus, we should ask at what stage in their sociological history are these failing states? What kind of societal structures do they have? What are their innate dynamics? Then we will be ready to seek ways to build on these existing structures and trends.
One of the greatest insights of the neoconservatives, one which carried them forward for a long time both as public intellectuals and as political counselors, was the recognition that there are great limits to social engineering, that societies tend to be highly resistant to change, and that design drives initiated by governments are prone to failure. Sadly, they ignored the fact that the same grand insight that applied to Los Angeles, Chicago, and New York also applies in full to Helmand and Kandahar, Basra and Swat.
Cases in point
In iraq, it is all too obvious, most citizens’ main loyalties are to major “tribal” groups, especially the Sunni, Shia, and Kurds, and a few other smaller ones (Arabs and Turkmens). Within each of these, there are subcommunities (or clans) that tend to rival each other. These communities command a strong sense of loyalty from their members. They have a clear and strong sense of “do’s” and “don’ts” specific to their community and duties to the common — tribal — good. They have distinct religious beliefs and rituals, tribal leaders, and tribal councils. And they have armies, arms, and ways of raising revenue.
We will probably never find out to what extent the surge in the number of American troops in Iraq in 2007 served as a turning point in the war there, and to what extent a tribal deal made the difference. However, a strong case can be made that a major part of the success attributed to the surge was actually due to the deal the United States made with a group of Sunni sheikhs — a deal known as the “Awakening” — which turned them from a major force of insurgency to one that supported the regime and fought for security. In other parts of Iraq, working with local tribes or clans has also made a considerable difference. For instance, in June 2007, U.S. and Iraqi forces reached agreements with ten Iraqi tribes in the Baghdad area to help in the fight against al Qaeda. In October of that year, the U.S. military reported that attack levels were steadily declining.
Most dramatically, there were next to no American or other coalition forces in northern Iraq, the area in which Kurdish tribes dominate, and yet the area was peaceful. This was the case because the Kurds used their very sizeable armed force, the peshmerga, to maintain security. In a February 2007 interview, Major General Benjamin Mixon, the commanding officer of U.S. forces in northern Iraq and Kurdistan, told 60 Minutes that of the 20,000 troops under his command, a mere 60 or 70 were stationed in Kurdistan. Because Kurdish areas are patrolled by Kurdish troops, “there’s no need” for an American presence in Kurdistan. The U.S. and its allies suffered next to no casualties in this sizable area.
In Pakistan, which is more of a cohesive nation than Iraq and Afghanistan, there are still major tribal forces at play. It is well known that there are seven tribes in Waziristan that govern themselves, have considerable armed forces, and do not recognize the Pakistani government. Beyond that, there is also much tension between the Punjabi-dominated government and the Pashtun tribes in the mountainous borderlands, some of which had an autonomous status until 2002 and seek to be integrated into an, at least, semi-independent Pashtun province. In short, what seems to Westerners a fight between the central government and some rebellious citizens is, to a significant extent, a confrontation between Punjabis and Pashtuns. Thus, even here, in a nation that is more unified than Afghanistan, an analysis of where we are and the ways in which we can proceed should start with an examination of the tribal tensions and the means by which their differences may be worked out (e.g., greater autonomy for the Pashtuns), rather than from the assumption that there is or can be in the short order an effective national government that will command the loyalty of most of its citizens.
In Afghanistan, the Taliban were defeated in 2001, with very few American casualties. During the 2001 air and ground campaign that drove the Taliban out of major Afghan cities, only 12 American service members were lost. The war was instead won by an American-supported coalition of several tribes, mainly Tajik, Uzbek, and Hazara, known as the Northern Alliance.
Security: Tribal bests national
The implications of the cases just briefly visited are especially important for the formation of native security forces. Americans tend to take for granted that these ought to be national forces, although even in the United States, much police work is locally and not nationally controlled. The National Guard can be called up only by the governors of the various states and each unit primarily serves its own state. Yet, in Iraq, after American-led coalition forces removed the Saddam regime, the United States and its allies tried to create a national force in Iraq by insisting that Sunni, Shia, and Kurdish units either disarm or integrate. Moreover, the United States positioned Shia forces in Sunni areas and Sunni forces in Shia areas, on the grounds that they should cease to view themselves as tribal forces and start acting like “Iraqis.” The result was often increased bloodshed. (One may argue that Saddam had a national army. However, it was in effect dominated by one tribal grouping, the Sunnis, with most officers drawn from one clan, Albu Ghafour. The Shia majority felt alienated and oppressed. Indeed, both the Shia and the Kurds fought against Saddam’s “national” army.)
A similar development took place in Afghanistan. In the aftermath of the defeat of the Taliban in 2001, the new Afghan government sought to disarm the tribal forces that had ousted the Taliban, what the government referred to as the amf (Afghan Militia Forces), in favor of fashioning a new Afghan national army. As a result of this Disarmament, Demobilization, and Reintegration (ddr), about 63,000 militiamen were disarmed by 2005. However, there are still a great number of unofficial tribal forces. Estimates of their size run between 65,000 and 180,000.
Moreover, according to Selig Harrison, the United States has helped the Tajik minority gain control of many key levels of government, including the armed forces and important security agencies. These forces often operate in the Pashtun areas in which the Taliban are particularly strong, due to their proximity to Pakistan (the Pashtun are concentrated in southeastern Afghanistan) and because the Pashtun are more supportive of the Taliban than other tribes. The un reports that Tajiks comprise 70 percent of the battalion commanders in the army, and many of the army units sent to Pashtun areas are composed mainly of Tajiks who do not speak Pashto. And while the U.S. supported Karzai in order to put a Pashtun face on the government, other Pashtuns consider him a turncoat, a sort of Uncle Tom.
As Seth Jones, a political scientist at the rand Corporation, testified before the House of Representatives Foreign Affairs Committee,
The United States and others in the international community have focused the bulk of their efforts since 2001 in trying to create a strong central government capable of establishing security and delivering services. This goal is ahistorical in Afghanistan and it is not likely to be effective. In addition, the local nature of power in Afghanistan — including in Pashtun areas of the country which are populated by a range of tribes, sub-tribes, clans, and qawms — makes this objective unpopular among many Afghans, who remain skeptical of a strong central government.
Beyond the very serious issue of loyalty, the center-to-periphery nation-building approach runs into difficulties because of especially high levels of corruption and attempts to hold the new national forces to Western standards. When local Afghans deal with members of the national police or army, they report that they are exploited, subjected to endless demands for cash and goods, and given little protection. This is due to a culture in which corruption is endemic and to the fact that the soldiers and police are not locals. They take their cut and are soon assigned elsewhere.
In contrast to the national forces, members of tribal forces live with their fellow tribe members, who are often related to them as kin, and while they are far from free of corruption, they tend to show more self-restraint. They have also shown a willingness to fight for their own kind, while members of the national forces often do not show up and avoid fighting. Furthermore, national forces demonstrate a remarkable level of disloyalty to the nation. There are reports of members of the national police selling to the Taliban arms provided by the West, of members of the national army delaying American military operations to give the Taliban time to set up ambushes, and of other betrayals of their government and their Western allies.
The vision of building a central national police and army also suffers because of sociological ignorance. Western trainers and advisers are trying to convert the Afghan army and police forces into units that adhere to Western ethical and legal standards of professionalism. Thus, a Western trainer cannot stand by if an Afghan trainee does not show the proper attitude toward women, suspects, or minors. No wonder, given the deep and strong differences between Afghan and Western culture, that after eight years, the number of successfully trained police officers is still rather small. The training of the army has progressed more successfully, but for the same reasons it is still far below the level needed if national forces are to be relied upon.
All this suggests that the U.S. and its allies in Afghanistan should work with the local tribal forces and their natural leaders, and shift major resources currently dedicated to the national army and police to help shore up the local forces rather than seeking to dismantle them. Each local force can be responsible for security in the areas of its tribe.
Such suggestions are challenged by the observation that the Afghans do not want a decentralized government. James Dobbins of the rand Corporation has noted that Afghans want a central government, but a very weak one. Barnett Rubin, professor at New York University, who has studied Afghanistan for 26 years, writes that he does not know any Afghans who want decentralization. Moreover, the constitution that Afghanistan adopted in 2004 does not accommodate a federation.
These are all well-taken points — if one approaches the subject from the viewpoint of a legal scholar or a political scientist. However, speaking as a sociologist, one notes that Afghans already have a de facto societal federation (or even a confederation) in terms of the ways they govern themselves. In effect, Afghans have a high level of regional, cultural, social, and even political self-determination. They reject the government in Kabul, especially because it was fashioned by foreigners and has lost much of whatever legitimacy it had due to association with drug lords and the heavy-handed manipulation of the 2009 elections. In the long run, one may work with the tribal leaders to integrate their armies into a national force and to slowly move towards implementing Western standards with regards to individual and minority rights and professional, modern modes of fighting. In the short run, working with sizable tribal armies seems much more likely to succeed: This is what I mean by moving from the periphery to the center rather than from the center to the periphery. Instead of trying to break the tribal authorities and undermine tribal loyalties and structures, we must build upon them and move towards coalition building, which may gradually lead to society-wide commitments and forces.
Recently the U.S. and its allies, reluctant to deal with tribal leaders, faced an unfortunate development: The tribes united against the U.S. In the summer of 2009, Admiral Michael Mullen reported that the situation in Afghanistan was “serious” and “deteriorating.” In September 2009, General McChrystal echoed these sentiments, writing, “Although considerable effort and sacrifice have resulted in some progress, many indicators suggest the overall effort is deteriorating.” One of the main reasons: Many Tajiks (led by, among others, warlord Ghulam Yahya), who previously were relatively supportive of the U.S. and the Karzai government have now joined the Pashtun in opposing both.
Natural leaders and elected leaders
An essential feature of a stable and adaptable political system is the availability of institutions that can be used to settle differences without resorting to violence. We tend to assume that these political institutions will be democratic and that various particularistic interests will be represented by elected officials. In this way, as the saying goes, ballots will replace bullets. In Iraq and Afghanistan, the United States and its allies invested considerable effort into introducing free and fair elections, in part to serve the purpose of absorbing intertribal conflicts into political institutions.
Given that the format of the introduced institutions was greatly influenced by the United States, however, they often did not reflect the preferences of the Iraqi or Afghani people. For instance, when it came time to draft an interim constitution for Iraq, L. Paul Bremer, head of the Coalition Provisional Authority, stated that he would veto any document that made Islam the sole source of legislation, despite the fact that the two main Shia parties called for the institution of sharia. The differences between American requirements and local preferences resulted in odd compromises in both the interim and final constitutions of Iraq. For example, Article II of the constitution stipulates that “No law may be enacted that contradicts the established provisions of Islam” and that “No law may be enacted that contradicts the rights and basic freedoms stipulated in this Constitution.” Additionally, the United States pushed to give more authority to the national government and opposed greater autonomy for regional bodies that largely parallel the tribes, a federalism preferred by many Iraqis. In Afghanistan, the U.S. insisted that the constitution be drafted and approved by consensus before the election of the National Assembly and national officials, and it promoted Karzai as the national leader. None of these moves helped lend legitimacy to political institutions that were imported and alien to begin with.
Often, native people have their own institutions and ways of selecting leaders and resolving conflicts — tribal councils, for instance, or community elders. Religious authorities also serve to guide, influence policies, and resolve differences. That is, the people often rely on natural leaders — those who rose to power due to their charisma, persuasive powers, lineage, or religious status, but who were not elected in the Western way. Initially, one had best try to work with them, rather than expect that they could be replaced by elected officials in short order. The same holds for tribal councils and intertribal bodies.
To understand the influence of natural leaders, one could consider the case of Ismail Khan. After the defeat of the Taliban, Ismail Khan, a warlord in Herat, Afghanistan, became governor of the area. Despite his ability to maintain security, Khan’s support of Iran and his refusal to send the tax revenues he collected to the central government, coupled with a wish to strengthen Karzai, led the United States to urge for his removal. Khan was removed from his local post in 2004, a move that resulted in violent protests, sectarian violence, increased crime, and the Taliban making inroads into Herat. Similarly, Governor Gul Agha Shirzai of the Nangarhar province was removed from a previous gubernatorial position for his autocratic, warlord style, but is now seen as necessary to stabilize the province.
While the United States, at least until very recently, tended to avoid working with tribal chiefs, Karzai has been courting them assiduously. Indeed, they played a key role in mobilizing various tribes to support his reelection. (Working with tribal leaders raises ethical questions that require separate treatment. One should note, though, that even if the U.S. and its allies avoid them, the national government the West is propping up does not. That is, they are prominently in the picture, one way or the other.)
Seth Jones reports that a strategy that seeks to build a strong central government and to hold territory with foreign forces is unlikely to work in Afghanistan. He reports that the national presidential elections in 2004 and parliamentary elections in 2005 did little to diminish the power of regional warlords and tribal militias. Even efforts that were made to relocate such leaders and wrest them from their regional power bases were unsuccessful. Instead of attempting (and failing) to break these solid ties, strategists should draw upon them to promote security in Afghanistan. Jones points out that a successful bottom-up strategy must strengthen the local tribal and religious leaders who understand their communities best, so that they may provide security and services. Indeed, he writes, “the most effective bottom-up strategy in Afghanistan is likely to be one that already taps into existing local institutions . . . Local tribal and religious leaders best understand their community needs.”
Similarly, Clare Lockhart, an expert on Afghanistan, told the Senate Foreign Relations Committee, that what was needed was a “‘light touch’ form of governance . . . where formal structures . . . can ‘mesh’ with local and traditional networks and social organizations . . . Networks of traditional birth attendants, hawala dealers, traders, ulema, and teachers can all be mobilized or partnered with for different tasks.” In a 2008 survey, the Asia Foundation found that local representative bodies (both traditional ones such as the shura and jirga and newer ones such as the Community Development Councils and Provincial Councils) enjoy the support of about two-thirds of the population. In addition, almost 70 percent said that religious leaders should be involved in local government decision-making.
The reluctance of the United States to build on the tribal power structure is one major reason its attempts to break away the moderate Taliban from the insurgents in Afghanistan have failed. According to Fotini Christia, assistant professor of political science at mit, and Michael Semple, a regional specialist focusing on Afghanistan and Pakistan, individual fighters are generally “operating within their home provinces, where their relationship to the local population is defined by their tribal status and political background.” In Iraq, it was sheikhs who played the major role in the Sunni Awakening movement (and not the Sunni’s elected representatives in Baghdad), and they were the leaders the U.S. commanders turned to in the Anbar region (which includes Fallujah). These sheikhs were the leaders who decided to cooperate with the United States in taking on al Qaeda in Iraq, routing them from the region.
The necessity of working with tribal leaders in Afghanistan is a lesson the British learned in the 1800s, when efforts at colonization were failing. Instead of continuing to engage in a military undertaking, the British paid subsidies to the tribal leaders on the borders but otherwise let them control their own areas, thus turning them away from those who threatened British interests. Working with tribal leaders is still a strategy, albeit in a modified form, which current British officials support.
Working with tribal and religious leaders will not only help us establish security but also lead to a much more efficient use of aid money. To date, usaid projects have been poorly planned and badly administered, reflecting American unfamiliarity with Afghan culture and terrain. In the summer of 2009, Richard Holbrooke called the campaigns to eradicate poppy crops and usaid’s alternative employment programs the “most wasteful, most ineffective program I have ever seen.” To provide one example, in late 2005, usaid funded a project which put people to work building cheap, but labor-intensive cobblestone roads. When the project was completed, the contractor learned that the local leaders objected: Cobblestone hurt camels’ hooves, and they wanted gravel and asphalt roads.
Again, it is not necessary to ally oneself with tribal chiefs no matter how authoritarian they are, although the question stands to what extent a foreign power should intervene in the internal political development of these countries, on both principled and pragmatic grounds. However, it is essential to make the leaders in place the cornerstone of starting an alliance.
Religious authorities can play this role. The United States must overcome its reluctance to work with religious leaders and instead embrace and even favor them — but only those who reject violence, of course. A prime example is Grand Ayatollah Sayyid Ali al-Husayni al-Sistani, the most important Shia cleric in Iraq. He is very influential among the largest Iraqi confessional group (some 60 percent of Iraqis are Shia) and a strong advocate of nonviolence. Initially, the United States sought to marginalize him. The reasons are telling: He is not elected by voters, and thus does not fit the democratic model. The U.S. believes that separation of religion and state should be promoted in other nations despite the fact that many democracies do not draw such a line, and the U.S. is increasingly finding ways to bridge the two worlds at home. American agencies acting overseas, especially the State Department, are very uneasy when they have to deal with religious figures, both because many American civil servants are not devout, and because they believe that it would violate the law to use U.S. taxpayer money to support religious authorities. However, if one returns to what I suggested is the essential prerequisite for sound policies — that one start from where people are, not from where we believe they ought to be — one cannot ignore that many of the most influential people in the countries in which terrorists thrive are religious authorities.
Seeking to marginalize or dismiss these leaders merely undermines much that the U.S. seeks to accomplish — regardless of whether or not it should be in the business of promoting secular regimes in the first place. In short, building up from the periphery means initially accepting the tribal groups and their natural leaders, traditional and religious leaders included.
Ultimately, the question of the extent to which one should rely on tribal forces, and not on a national police or army, to maintain order depends on the goals foreign powers set for themselves. If the main goal is to build a stable democracy and a modern economy, promoting national forces is called for, though still in the gradual, transitional way already indicated. However, pursuing these goals, given the high level of corruption and the lack of commitment of Afghanistan’s leaders to a democratic state makes it likely that the U.S. and its allies will have to stay in that country for decades and commit a large amount of resources there. Thomas Friedman, who is a master at putting things succinctly, wrote in the New York Times, “This is State-Building 101: our partners, the current Afghan police and government, are so corrupt that more than a few Afghans prefer the Taliban. With infinite time, money, soldiers and aid workers, we can probably reverse that. But we have none of these.”
In contrast, if the purpose of U.S. policy is, as President Obama concluded in March 2009, “to disrupt, dismantle, and defeat al Qaeda in Pakistan and Afghanistan and to prevent their return to either country in the future,” much can be done by relying on tribal forces, their natural leaders, and intertribal coalitions. At least, communitarians would argue, this is the level at which all restructuring of failing states like Afghanistan must be started.
