Abstract
In this essay I advance two arguments. First, while Afghanistan was once a state built on a social contract that demanded a balance of both modernist and traditional goals, a combination of exogenous shocks shattered that equilibrium, resulting in a territory that is no longer governed by a state, and that continues to generate what economists might call ‘negative externalities.’ It is these negative externalities that have drawn in overlapping interventions (the most salient of which remains Operation Enduring Freedom, or OEF), all of which have failed or will fail. Second, the key feature of Afghanistan’s disintegration is the flight of its public servants – the people most directly necessary for any state to work. I argue that Afghanistan does possess sufficient human capital for effective government, but that its public servants are effectively bounded by tribe, clan and valley. I conclude by showing that state-building demands public servants and public service, and that as long as Afghanistan fails to devote resources towards acquiring and maintaining a minimal critical mass of such people, it can be expected to continue to generate negative externalities that harm its people, its neighbors and the international system more broadly.
Keywords
There was a time when the costs and risks of distant harm – civil, ethnic, and religious war; pollution; organized crime; revolution; famine; terrorism; female infanticide; disease; and so on – could be radically discounted. If one had the good fortune to be born in say, Osaka or Marseille, or San Francisco or Oslo, then the troubles of say, Afghanistan or Zaire (now the ambitiously renamed Democratic Republic of Congo), were distant troubles, and the likelihood that a dictator’s abuses, a war’s ravages or a smuggler’s enterprise could affect (much less kill) you or fellow citizens was very remote.
All that began to change in the late 1950s, when the global communications revolution that began with the container ship (and accelerated through the Internet and the mobile phone) began to make distant damage much more local. There have always been ethical objections to the social and political consequences of the centuries-long process of wealth concentration north of the Equator, but following the collapse of the bipolar order of the Cold War, what we have witnessed is the replacement of an easily-ducked moral argument – for fairness, justice, redress, remedy, and compensation – with an interest-based argument for ‘intervention’ to stabilize, or rebuild ‘failed’ states (Rotberg, 2003; Ghani and Lockhart, 2009).
By nearly any definition one chooses, contemporary Afghanistan is a failed state. Its government, nominally seated in Kabul, has progressively limited capacity to enforce laws or collect taxes beyond its confines; and such capacity as it currently has is likely to disintegrate once the International Security Assistance Forces (ISAF) depart later this year. It is important to emphasize two points in this regard before proceeding with the larger argument. First, the ‘failure’ of the Afghan state was not inevitable, although for long-standing geographical, cultural and historical reasons, the formation of a real state on the territory would demand the highest level of political, social and diplomatic skill and patience. Second, it hardly follows that Afghanistan needs to be a state governed from Kabul, or that both its own people and the citizens of neighboring states (or indeed the international community) could not be better served by something other than the reconstruction of an Afghan state along the lines currently sought by say, the United Nations.
Arguments
As reflected in this special issue, most essays about contemporary Afghanistan begin with a story about the dreadful impact of the lack of supply of some needed factor or element. Afghanistan should be supplied with economic assistance, agricultural assistance, construction assistance, security personnel, lawyers, a judicial system, training, weapons, diplomatic and food aid, roads, bridges, rail links, aqueducts, hospitals, medicines, textbooks, teachers, schools, electricity, and on and on. The basic form of every argument about Afghanistan as a problem reduces to the same thing: supply whatever the key lack is, and the outcome will be peace, justice, or at a minimum, something called ‘stability’. The outcome variables are necessarily unqualified or vague because the bulk of the suppliers called upon to save or aid Afghanistan live and work well beyond its borders, and to the extent that they remain willing and able to supply some key-turning lack in Afghanistan, their motives may be considered interest-based. No one, for example, wants to preside over a government sustaining another 9/11 type attack, or to be accused of ‘standing idly by’ while would-be Afghan schoolgirls are assaulted for simply wishing to learn to read.
Very few analyses begin with what matters most: demand. The first question is: who, in the territory marked on all our maps as ‘Afghanistan’ wants a centralized state governed from Kabul? How widespread, in other words, is the demand for an Afghan state contiguous with its formal borders? Empirically this is a difficult question to answer, but let us consider the logic of a broadly negative response first. If demand drives supply (rather than the reverse), and there remains insufficient demand in Afghanistan for an Afghan state, then logically the supply of government capable of governing such a state will be lacking. It follows that the gap between what is needed and what is demanded would need to be supplied from somewhere else. It also follows that, should domestic demand for government not rise (or worse yet, decline) and the external supply stop, the state as such would fail.
Of course, since we remain in the realm of logic, it is worth considering whether the supply of a thing can increase the demand for it, and the answer, as everyone knows, is under certain circumstances, yes. Here we think in particular of commodities that have an addictive character, such as tea, coffee, tobacco, heroin, cocaine, alcohol, and so on. It is worth applying this logic to the question of the supply of government and the demand for it in Afghanistan. The second question is: has the provision of various sorts of lack (in particular, government) in Afghanistan increased the demand for government by Afghan people, and if so, where?
This is a terrific question that leads to others. Let us assume that the answer is ‘yes’. We can then ask another great question: what do most Afghans think government, or better still, good government, is? What are the bases of authority and legitimacy – two crucial components of government in Western political philosophy – in Afghanistan, and are those different in any important way from what we might expect in, say, Jakarta or Buenos Aires or Prague?
Note that, without engaging the entire history of the philosophy of government, we may usefully reduce an entire range of responses into two categories: (1) what is minimally needed; and (2) what is desired beyond the minimum needed. Put this way, we can see that a map of responses to such questions would be a first requirement of government. We might also assume that answers to the first question are more apt to be universal than answers to the second. So, for example, Max Weber’s important definition of the state as that political association with a monopoly on the legitimate use of violence implies that a state is that political association capable of guaranteeing (not perfectly), at a minimum, the physical security of its citizens. Citizens must have a reasonable expectation of walking from home to market and back without being assaulted and harmed; and where this expectation is not satisfied, people tend to leave. Beyond this minimum, there is an economic component that has a similar ‘sine qua non’ character: citizens must have sufficient access to food, water and shelter to guarantee their survival. Again, where these expectations are unmet, people leave. Beyond these minima of physical security and economic survival, however, there remains a wide and healthy variation of responses to the second question: what is desired of government beyond this?
We should not be surprised to find that Afghans subscribe very vigorously to the minimal requirements of government, as we all might. However, it does not follow that a centralized state, much less one covering the whole of the territory of Afghanistan, is either necessary or desirable in order to guarantee this minimalist definition of government, much less something more ambitious. This leads to question three: under what circumstances might we expect Afghans to demand a state, and when in recent times have those circumstances been met?
Taken all together, these questions and their answers form the core of my argument about public servants, public service and the alleged ‘failure’ of the Afghan state (or any state, for that matter). I argue that the geography, climate, history, economy and culture of Afghanistan have all mitigated against the formation of a centralized European-style nation state on the territory of Afghanistan. Even so, beginning with the Anglo-Afghan wars of the nineteenth century, and culminating in the patient and skilled efforts of Afghanistan’s ruling tribal elite, a European-style state did form there by the 1920s (and lasted until 1978). However, a series of exogenous shocks, including the independence of India from Britain in 1947, Soviet meddling in Afghanistan’s internal affairs in the 1960s and 1970s (culminating in the murder of Mohammed Da’ud Khan and his family in 1978), followed by a decade of Soviet military intervention and US countermeasures (which among many other things helped establish and supply a network of radical religious actors), resulted in the complete destruction of that state, the dispersion of its ruling elite, and most crucially, human capital flight. Since 1979, in other words, Afghanistan has become adept at importing religious radicals, small arms, well-intentioned aid workers, administrators and armed forces – and suitcases full of cash. It has not created sufficient indigenous or domestic human capital to re-establish a centralized state based in Kabul, and this is due to a widespread lack of demand for a centralized government.
The problem of failed states
As implied so far, Afghanistan may not be a failed state insofar as the demand for a state there is really an externally driven demand. What has driven this demand is not a general concern for the welfare of Afghan people, but an interest in reducing the costs and risks of something nasty being exported from Afghanistan to its neighbors and beyond these to Europe, Africa, the Americas, and East Asia. The formation of a state in Afghanistan gives those at risk some place to deter, coerce, bargain with or bribe.
It is worth focusing on one or two of these nasty things before moving on to discuss the possibility of reducing the demand for their production, and their subsequent export.
Terrorism
As a former soldier I can admit to a certain amount of amusement at the idea of the use of Afghanistan (or any other failed state) as a base for ‘training terrorists’. Since by definition terrorists will be assaulting (most often murdering) those who cannot defend themselves, the amount of ‘training’ they might require is minimal (and thus hardly demands either the physical space or potential privacy of a failed state). A long line of such murderers masquerading as soldiers have had the same ‘training’, from Himmler’s Einsatzgruppen in World War II, to Slobodan Milosevic’s paramilitaries in Kosovo in 1999, or Joseph Kony’s Lord’s Resistance Army in the present day. Whenever and wherever such murderers have run into real soldiers, they have been rapidly wiped out at almost zero risk to the soldiers. Because nowadays terrorists kill most effectively by means of cleverly designed and improvised explosives, it is unclear why this sort of training could not be done in an urban basement in say Calcutta, Detroit or St Petersburg, rather than in a terrorist ‘training camp’ in Afghanistan or Libya.
That said, there is nothing amusing about terrorism or terrorists, and it is this export that made Afghanistan and the condition of its citizens a major international concern in 2001. In spite of the fact that the group that assaulted the US Pentagon and New York’s World Trade Center also trained in Florida, the most lasting argument for the military intervention in Afghanistan called OEF was to prevent its leaders, the Taliban, from hosting another group that might train to accomplish a similar or worse attack on the United States or another state.
Terrorism of this sort gains its attention over other harmful failed state exports such as narcotics and pollution because it is more immediately lethal and obvious. It also appears to be something that can be easily remedied by a lethal physical response (namely a military intervention aimed at ‘finding, fixing, fighting, and finishing’ those responsible for the attacks; Ridgway, 2007, as cited in Halberstam, 2007: p. 493). On both counts then, and in keeping with the discussion of ‘minimal government’ above, a failed state’s association with the export of terrorism becomes a matter of state.
Other exports: narcotics
Afghanistan’s Helmand province is among the world’s leading producers of opium poppies. This is true in spite of the fact that Helmand farmers (among others in Afghanistan) have been the target of sequential and overlapping campaigns intended to encourage them to substitute other cash crops for opium poppies, and to severely punish them when they are found cultivating poppies. The United Nations Office on Drugs and Crime (UNODC) reports variation in the likelihood that, year to year, farmers in each Afghan province are likely to grow poppies (UNODC, 2012). The specific estimates are of less interest than the UNODC’s judgment as to the determinants of a farmer’s decision to grow poppies. The biggest driver is insecurity: farmers who are less secure are more likely to grow poppies. The other major factor should be unsurprising as well: economics. Farmers who are paid a real subsidy to grow a different cash crop are more likely to do so. These two metrics should sound familiar, because they reduce to empirical evidence of the extent of state authority. Essentially, where the Afghan state has reach, poppy cultivation is apt to fall off or remain non-existent; and where the reach of the state is compromised or nonexistent, poppy cultivation remains at last year’s levels or is set to increase. Even more importantly, the security and economic assistance provided are reciprocally reinforcing (the better the security, the more aid can be distributed effectively, and the more aid, the more security, as young males substitute into agriculture and away from violence in order to earn enough cash to marry) and almost entirely dependent upon outside aid. Once the outside support declines, the process reverses and opium cultivation multiplies.
This matters because the value of the crop depends upon it being converted to heroin, one of the most addictive narcotics known. Very little of the opium grown in Afghanistan is consumed there, which means that Afghan farmers are exporting the costs of their crop to other countries in exchange for what, in relative terms, are tiny returns (about 3% of the value of the crop stays in Afghanistan). Of course 3% of a US$4b crop adds up to a healthier income for Afghan farmers than any other crop they could grow.
Again, narco trafficking did not (and does not) gain the same traction as terror as an export, even though it is a relatively simple matter to show that the impact on importing countries is far more lethal than terrorism. As one UN study from 2009 put it, ‘the number of people who die of heroin overdoses in NATO countries per year (above 10,000) is five times higher than the total number of NATO troops killed in Afghanistan in the past 8 years’ (UNODC, 2009). It should go without saying that, as devastating emotionally and psychologically as the attacks of 9/11 were, they pale in comparison to the toll that illicit (and licit) addictive drugs take on every importing society.
Yet however one chooses to apply metrics, states and the citizens they are tasked with governing are gravely and chronically harmed by substance abuse, including heroin and cocaine. This means that poppies grown annually in Afghanistan are killing Americans, Japanese, Britons, and especially Russians and Chinese. There are three reasons narco trafficking as such does not hold the same priority as terrorism. First, with relatively few exceptions, the global demand for illicit drugs is satisfied by the current offerings of marijuana, cocaine, opium and heroin. Crack cocaine, crystal methamphetamine and oxycontin remain three innovations whose impact was and remains devastating, but on balance the threat of ‘new’ addictive drugs for current addicts and their loved ones and communities is slight as compared with the threat of radiological, chemical or biological attack by terrorists. In other words, the domino logic that motivates a great deal of the costly and counterproductive reaction to transnational terrorism does not have as much traction in relation to illicit narcotics as it does to terrorist capabilities. Second, terrorism kills deliberately and publicly, whereas heroin overdose kills privately and inadvertently (from a dealer’s perspective an overdose is bad because a dead addict is a lost source of rents). Third, a consumer of an illicit drug (or licit drug, for that matter) exercises some choice, and therefore takes some personal responsibility, in the risk she or he takes in consumption, whereas for victims of terror there is no choice, and therefore no responsibility.
Other externalities
I list terrorism and narco trafficking as the two main negative externalities associated with failed states because together they are sufficiently emblematic to make it un-necessary to catalogue others. However, three others deserve at least some mention here.
First, there is human or sex trafficking. Second, there is communicable disease. Third there is pollution. I mention the first because trafficking in young people (mostly females and children) for sex is dramatically more profitable when addictive drugs are added into the mix. As vulgar and hurtful as it is to speak of, much of what customers want in a sexual service depends upon a veneer of compliance, and sex traffickers often deliberately addict their victims to drugs like heroin in order to obtain the compliance they need both to attract customers and to make their business model work. By this logic we can again say that a subsidiary consequence of an Afghan farmer’s decision to grow poppies is the sexual enslavement of some portion of the world’s most vulnerable population: young women and children.
Communicable disease can also be an artifact of failed states, although in this case it is more difficult to indict Afghanistan specifically (a welcome respite). However, the argument goes like this: once the minimum threshold of public servants within a state with a functional healthcare system is crossed, local and indigenous healthcare is compromised, leaving citizens within the state much more vulnerable to communicable disease, and leaving the state much less capacity to isolate outbreaks than it might otherwise expect. A strong contemporary example of a failed state whose collapsed healthcare infrastructure led not only to domestic injury, but also to the deaths of citizens of neighboring states, is Zimbabwe in 2008 (Dugger, 2008). The accession of Robert Mugabe to the presidency of Zimbabwe in 1987, and his subsequent land redistribution policies, led to a progressive hemorrhaging of public servants or human capital flight from Zimbabwe. This in turn led to a failed state, in which over time investment in infrastructure of all sorts (including healthcare infrastructure) declined markedly. By 2008, cholera – a disease whose causes and treatment are both well known – took root and spread. Again, not only did Zimbabwe’s citizens suffer, but citizens of neighboring states (Botswana, Mozambique, South Africa and Zambia) also died as a result of the Zimbabwean government’s lack of capacity to prevent the spread of the disease. The epidemic was only stopped by permitting Mugabe to publicly blame the epidemic on foreigners while privately allowing international aid workers to supply Zimbabwe’s lack.
A final externality worth introducing is pollution, but here I want to simply make the point that, although it is true that when a failed state takes in toxic waste in exchange for cash, that such waste compromises the health of local and regional citizens, it is not true that the bulk of the damage from pollution worldwide comes from failed states. On the contrary, when the US Secretary of State calls upon Afghanistan’s Hamid Karzai to reduce opium cultivation, it might be fair for Karzai to point out that the United States leads the world in per-capita carbon emissions, and that these same emissions are at least part of the reason it is easier to grow poppies in Afghanistan than many of the other crops the international community hopes to substitute for them.
In sum, failed states generate a number of negative externalities that, because of the now relatively low costs of communications, have a far more direct and lethal impact on so-called advanced-industrial states. All sorts of organized crime are co-located with failed states, including terrorism, piracy, drug trafficking, sex trafficking, Internet fraud, counterfeiting, pollution, and so on. Beyond their immediate lethality, mostly private and chronic, each of these externalities does trillions of dollars’ worth of economic damage each year, to say nothing of the injury, heartbreak and suffering they cause short of death.
The state
When thinking through the issue of the relationship of public service to the state as such, we begin by asking ‘why a state?’ Is the state, as a form of political association, necessary, and if so, under what conditions might the positive utility of living within a state be expected outweigh the costs of association as compared with other options? Finally, we can ask whether those conditions that make living in a state a net benefit exist in contemporary Afghanistan.
The history of opposition to the state as a form of political association is very much weighted toward two nearly-identical critiques: (1) the Anarchists of the late nineteenth century (Bakunin, 1990); and (2) the Marxists of the same period. The core of both groups’ critiques was that states would invariably intervene in the affairs of citizens to preserve the power of the ruling classes, and for the Marxists in particular, their wealth. In addition to abridging civil and economic liberty within, competing states would invariably draft citizens for costly wars. Taken all together, the critiques reduced to the claim that, for all but a tiny minority of people, the benefits of citizenship would invariably be eclipsed by the costs. The primary difference between Anarchists and Marxists proved not to be their critique of the state, but their solution to the state’s tyranny. For Marxists, states would ‘wither away’ after a spontaneous and transnational workers’ revolution swept away the props of state power in a glorious realization of shared production and consumption. All that was needed was (Marx–Engels) the right historical and economic conditions, or (Lenin) a vanguard of the Proletariat capable of enlightening workers sufficiently to set them on the path to revolution and earthly paradise. For Anarchists, successive assassinations of state leaders (the deed) were expected to awaken a spontaneous popular revolt against the state as an institution, leading to the collapse of the states system.
However, there were two challenges to the argument that costs of citizenship necessarily eclipsed benefits. First, there was the problem of war, and, as Charles Tilly so pithily reminded us, the strong association between the origins of the state as a form of political association and war: ‘War made the state, and the state made war’ (Tilly, 1975). The logic is simple: the state survived competition in a kind of natural selection with alternative forms of political association (e.g. clan, dynasty, tribe, and so on) in a context in which assault by another state was – as Thomas Hobbes would have it in Leviathan – ‘a climate wherein the will to contend was always present’. Thus, even if the costs of living in states were as high as those claimed by its critics, given the manifest horrors of war, this single utility could render those costs bearable and, over time, made bearing them seem unquestionable. Second, neither the Marxists nor the Anarchists could conceive of states being ‘hijacked’, as it were, by everyday average citizens (both workers and the middle classes), who could (and did) turn the power of the state toward the rule of law and toward more egalitarian distributions of wealth and resources. Taken together, these two features of citizenship have resulted in a net benefit to citizens sufficient to make the state, since the late seventeenth century, appear to be both necessary and inevitable.
The chief attributes of the state have already been discussed, but here it makes sense to underline the microfoundations of demand: citizens have recourse in the event of injury, and may seek redress from authorities. Widespread awareness of the possibility of sanctions governed by generally applicable and enforceable statutes, and ranging from fines and shaming to execution, then conduce to stability and predictability, which in turn permit investment, transaction and development. The end result is the potential of reciprocal advance, in which stability leads to wealth accumulation, which increases the capacity of the state to protect its citizens from internal and external threats, which in turn leads to more wealth accumulation, and so on.
Topography, central government, social adaptation
There are two important points to note in this connection. First, the distribution of topography and climate favoring centralized control of large territories is not universal, and where human engineering cannot supply communications, the capacity of the state to insure its minimal obligation of physical security and remedy after injury is necessarily abridged. Second, as we are often reminded nowadays, the likelihood of interstate war – or at least our perception of that likelihood – has declined precipitously since World War II. To the extent that both factors are in play, we can derive two important hypotheses, which bracket the conditions under which the state is likely to be ideal. The first is that, where topography and climate limit centralized government, the benefits of citizenship will be diminished, perhaps to the point where the costs of citizenship outweigh the benefits. The second is that, where citizens do not worry about assault and occupation by other states, the benefits of citizenship will likewise be diminished. Stated more formally:
H1: The more costly it is (in terms of communications across physical space) for a government to extend its control from its capital to its borders, the less security it will be capable of providing its citizens.
H2: The less likely a state’s citizens think the threat of conquest is, the less likely these citizens are to value state citizenship.
It is not necessary here to test each hypothesis, nor is it necessary to assert that there are no people within Afghanistan who wish for a unified state or who are prepared to stand as public servants within such a state. It is sufficient to point out that there are circumstances in which a rational individual might prefer, in a cost–benefit sense, to live within a political association other than a state or, failing that, to live within a micro state. More on this presently.
It should also follow that, in territories with a topography that mitigates against easy communication, social adaptation may supply a lack that technology may not. In other words, have the peoples of Afghanistan (or Switzerland) adapted socially over time to make what appears to be an insurmountable obstacle to state formation – and effective central government – possible? If so, how vulnerable are those adaptations to exogenous shocks or, more importantly, might whatever those adaptations be also make conquest and occupation much more costly?
One striking feature of contemporary Afghanistan is how similar its constellation of tribes, clans, interests and topography is to an analogous region in which a succession of Russian tsars attempted conquest and occupation and only succeeded at prohibitive cost (approximately 1 million casualties over 30 years). I am speaking about the Avars of Dagestan and the Chechens in the Caucasus in the early-to-mid nineteenth century (Arreguín-Toft, 2005: chapter 3). As in Afghanistan, topography proved to be the most crucial factor in the contest, and as in Afghanistan, students of the region noted social adaptations and attributed the resilience of the Murids to their religious unity under a third Imam, Imam Shamyl. Shamyl’s leadership, the prohibitively daunting terrain (sheer mountain cliffs and defiles), and his Murid’s religiously inspired unity were sufficient to dramatically raise the costs of Russia’s conquest, but given the fundamental asymmetry of forces and wealth, the cumulative impact of Russian campaigns took such a toll that in the end Shamyl was captured and the Caucasus ‘conquered’.
Note that few of these questions ever get asked and these comparisons rarely get made because, again, the initial assumptions that most analysts bring to Afghanistan (to cite but one example) are all blindingly wrong: the state as a form of political association is neither natural nor inevitable; nor is it a form of association that can provide the best security and prosperity for its constituents under all circumstances. Only once we move beyond thinking of the state as axiomatic does it become possible to move beyond thinking about supply-side strategies for helping Afghanistan, and to ask whether there is not a middle ground between what would be good for the Afghan people and what would be good for its neighbors and the broader international community (such as it is).
Public servants and public service
Two points must be made here before drawing broader conclusions about the future of Afghanistan and states like it. First, the logic of the relationship of public service to the existence and functioning of the state as such must be reviewed and applied to Afghanistan. Second, it makes sense to introduce at least some empirical support for the notion that Afghanistan has lost too many public servants to recover as a unified state in the near future (even a confederal one), although there is nothing to preclude the formation of such a state should the right conditions hold and should sufficient numbers of Afghans desire it to come to be.
Public servants in Afghanistan
By ‘public servant’, I do not simply mean a bureaucrat capable of effective administration – Kabul has many of those – but one capable of and interested in serving a state able to govern the territory of Afghanistan as a whole. The point cannot be over emphasized that, so long as Kabul’s public servants are beholden to Hamid Karzai and his clan, they are unlikely to count as ‘public’ in the sense I mean. The serve, and they administer – and both at times to great effect – but they do neither with respect to the country as a whole.
The Islamic Republic of Afghanistan itself has framed the issue of public service as an issue of ‘capacity’, and has put forward a plan to increase its capacity (to govern) in a document called Capacity Development Plan, 2011–2014 (UNDP, 2011). The government of Afghanistan’s understanding of the link between public servants and governance is set out as:
Of late, the value of local ownership of global capacity development has been recognized. While financial resources and technical assistance are vital in post-conflict . . . Afghanistan, these alone cannot sustain capacity development. Mostly these are donor-driven and address short-term needs, while distorting national priorities. Strong capacity, locally generated and sustained, is essential to the success of any development process. Developing countries and organizations should own, design, direct, implement and sustain the process themselves. (Central Statistics Organization, Islamic Republic of Afghanistan, 2012)
What is most interesting about this passage is the notion that Afghanistan is ‘developing’, and the natural question is (as it should be in all such cases), ‘developing into what?’ Presumably, Afghanistan is to develop into a centrally governed, democratic state. Yet the main argument of this paper is that each aspect of that formulation is preposterous: Afghanistan does not have the capacity for central government (because it has insufficient domestic demand for central government), democracy of the kind sought after by its well-intentioned (or self-interested, take your pick) donors is an alien concept of legitimacy in Afghanistan, and Afghanistan is effectively an imagined state: a state in name only (Arreguín-Toft, 2013).
If today we find a dearth of public servants either able or willing to reside in Kabul and help administer the imagined state of ‘Afghanistan’ (and I include the nascent Afghan National Army), this should only serve to highlight just how out of place was the broad political objective of the well-intentioned foreign aid and stability program for the country as originally conceived. Although as Thomas Barfield argues, Afghanistan’s diversity does not, as it might elsewhere, itself imply the impossibility of unity sufficient to make a centralized state work, the cumulative impact of the exogenous shocks have had the effect of increasing the demand for Afghanistan’s provinces and major tribal and ethnic groups to have a say in their government (Barfield, 2010: chapter 6).
Afghanistan was in fact a state from about 1928 until the assassination of Mohammed Da’ud Khan and his family by Soviet-supported Afghan Marxists in 1978. Since then three factors have made state formation (or reformation) impossible. First, the attempt by Afghan Marxists to govern Afghanistan forced most public servants whom the Marxists could not murder outright (all those with the skills and experience necessary to actually govern from Kabul to the frontier) to flee. After 1979 the country was effectively governed by Soviet administrators with help from Afghan Marxists.
Second, as if this were not bad enough, Soviet COIN was (like OEFs) under-resourced in almost every respect, which meant that Soviet COIN strategy would need to focus on mobility, and to concentrate Soviet forces on one province or valley in Afghanistan at a time. The effect was to further Balkanize Afghan national identity. Third, the Taliban fighters who eventually succeeded in unifying the country after the Soviets departed were not capable (nor much interested) in re-forming a state. As Barfield persuasively argues, their chief objective was purging Afghanistan of what they believed to be sin and backward superstition (Barfield, 2010: chapter 5). Against the increasingly resistant Afghan peoples, who had been relatively happy to see the backs of the mujahideen, the Taliban were not able to consolidate their power (i.e. train or attract enough indigenous public servants) before their more radical ‘foreign’ Islamic guests brought down the wrath of the international community upon them as a result of 9/11.
Finally, operation OEF brought with it a new set of foreign administrators following the ejection of the Taliban. Beyond a very successful punitive expedition, the forces assigned to OEF were well aware that they would be inadequate for nation or state building. We will never know whether the four-year grace period following the defeat of the Taliban and their return might have proved decisive had a positive reconstruction and development plan been actually implemented. Much aid was promised, but little was delivered. And much of what was delivered was pocketed by Afghanistan’s warlords and their cronies (Rosenberg and Bowley, 2012).
However, the end result of three decades of war and successive foreign occupations has been the devolution of loyalty to the state we call Afghanistan – always fragile and contingent at best – into historically unprecedented demands by the regions to have a greater say in Afghan politics: in other words, centralized government will need to yield to a more confederal structure and a more acceptable leadership than Karzai can provide. The bad news is that this has made NATO’s broad strategy of rebuilding a state existentially impossible because, for the time being, there is no state to rebuild. The good news is that, historically, other states with complex topography and powerful subnational identities have proven able to work together constructively within a larger state, so there is theoretically nothing to prevent Afghanistan from becoming more akin to Switzerland at some point in the future.
The Afghan national army: a ray of hope?
Starting in 2009, US General William Caldwell was tasked with devoting serious resources to creating an Afghan National Army (ANA) capable of supplanting ISAF’s chief role of securing the population and key values from insurgents. By 2012 – two years before the planned draw-down of ISAF combat forces – Caldwell is supposed to have increased the number of ANA forces (now standing at 300,000) to 352,000. ANA troops should be the single greatest empirical counterargument to my claim that Afghanistan lacks sufficient public servants to succeed as a functioning state, because soldiers are the original and quintessential public servants.
There are two problems with these efforts. First, ANA troops have been ‘trained’ and equipped to operate very much as their ISAF mentors do. Under limited circumstances, this means that they are lethal. As already highlighted, however, most Afghans, whether Tajiks, Uzbeks, Turkmen, Pashtun or Hazara, fear many things besides (and in a few cases, more than) death. So in terms of governance, even when these soldiers can kill Taliban, Al Qaeda, Haqqani and Herb i-Islami insurgents effectively, it is hardly clear that this killing will lead to coercion and stability any more than it has for the ISAF and the CIA. More troubling, ISAF forces gain a great deal of their advantage over insurgents because they are adept at logistics and at supplementing their infantry combat skills with heavy fire support (artillery and air). ANA forces remain largely illiterate, and are still a long way from being able to integrate fire support into their operations. In a sense then, their ‘training’ may be making them more vulnerable on several counts.
Second, in an ISAF unit, there are likely to be soldiers whose family members live far outside the battle space, and moreover, many will not live in the same place (an Air Force sergeant’s parents might live in Ohio, for example, while his sister and her family might live in Illinois). For ANA soldiers this is not true. Most of their families live in specific provinces or valleys in Afghanistan, and under the protection of a specific warlord. As a result, it is not clear whether, under duress, many ANA soldiers would desert their units and take their weapons and skills back to protect their families or join the forces of the warlord whose job it is to do the same thing. We can put this a different way – if ‘Afghanistan’ only exists so long as ISAF and other foreigners stay, then will the bulk of the ANA, at a minimum, effectively revert to Karzai as a warlord after they depart?
One possibility, sad but hopeful, might be that we answer that question ‘yes’. Assume for the sake of argument that Hamid Karzai ends up with an ISAF-trained and equipped military loyal to him. Will that military then not be capable, as ISAF was not, of both sweeping aside the insurgency and consolidating control of Afghanistan?
No.
The ANA is only a national army so long as sufficient ISAF personnel remain. It is only an effective killing force so long as ISAF forces can be called upon to assist with fire support and, as noted above, killing is not the way to conquer or control Afghanistan. Once the draw-down begins in earnest (and ideal strategy and US military opposition notwithstanding, budget constraints and lack of popularity in ISAF home countries seem destined to make a draw-down a reality), expect the ANA to begin dividing along provincial and ethnic lines. Already, ANA forces have been implicated in direct and lethal assaults on ISAF forces, and sadly, this is likely to escalate as ISAF combat forces depart (Rosenberg, 2012; ISAF, 2012).
In sum, the ANA is not likely to survive as the nucleus of an Afghan state, and while ISAF strategists were right in their understanding of what they would need in order to achieve their assigned missions – including especially the creation of an effective national security service – they were never given sufficient time or resources to do so. There is supreme irony in this, because the ISAF’s current strategic bind is identical to that of Soviet General Mikhail Zaitsev in 1986: not only no more troops, but fewer, and the fear that all gains patiently earned over the past decade will soon be wiped out. More painfully, as with the Soviet experience, the failure of these well-intentioned efforts is likely to make Afghanistan an increased threat to the security and interests of ISAF home countries after the ISAF departs.
Conclusions
OEF is scheduled to end in 2014. Its goals were, like the that of the Soviets beforehand, to provide security and sufficient stability for a local government in Kabul to consolidate power and re-form a functional state. Afghanistan no longer possesses sufficient public servants to administer a central state. However, the ISAF’s political objectives are unlikely to be met because Afghanistan has changed: the provinces have shifted the age-old question of ‘which’ ruling tribe will govern from Kabul to ‘how’ Kabul’s government is to be received in the provinces. The answer is that, whatever emerges must be more confederal than federal. In order for this to work, Afghanistan will need a leader other than Hamid Karzai, who no longer possesses the requisite legitimacy either within or beyond Afghanistan in order to negotiate this new reality.
In sum, Afghanistan was once a real state. It was never a strongly centralized state. It was never a democratic state. As a state it had been built on a careful balance between modernization and tradition which took constant effort and great skill to maintain. As Barfield summarizes:
For almost a century, attempts to bring about social change in Afghanistan have been led by governments in Kabul determined to modernize the country. For an equal period of time they have been resisted by the inhabitants of rural Afghanistan as well as conservative Islamic clerics who distrusted such changes and saw them as a threat to their traditional way of life. (Barfield, 2010: p. 339)
Barfield adds that governments (like Karzai’s today) that pushed modernization too rapidly or beyond the major urban areas, invariably provoked a violent backlash, but that from 1929 to 1978, Afghanistan’s rulers managed to find an acceptable equilibrium (Barfield, 2010: chapter 6).
The power equilibrium that resulted made it possible for the territory of Afghanistan to function as the state of Afghanistan. The advent of the Cold War and, in particular, the independence of India in 1947 set in motion the forces that would shatter that equilibrium. Afghanistan became a kind of place-holder, an imagined or, in more hopeful moments, imminent state.
Soviet intervention supported a Marxist coup that forced most of Afghanistan’s public servants – those with a genuine sense of Afghanistan as such – to flee. They were replaced mainly with Soviet public servants who were faced with attempting to govern through a corrupt, incompetent and illegitimate leadership in Kabul. The outrage provoked by Soviet military intervention in 1979 led to a fierce insurgency, and Soviet COIN attacked that insurgency valley by valley, province by province, only underlining and intensifying subnational identities. Political commands were relocated to Pakistan, where each warlord developed a coterie of private servants and administrators skilled at war, fundraising, and recruitment (survival). By the end of the war in 1989, Afghanistan had become a collection of micro states, each prepared to jockey for control of the entire country, but each vulnerable as foreign military and logistical assistance dried up.
The Taliban entered from Pakistan with a simple unifying mission: to consolidate jihad in Afghanistan by eliminating what they considered to be Afghan sin and superstition. The problem is that, in both the Marxist and Taliban cases, it became possible to see the ideas being carried into Afghanistan as in some sense foreign ideas (the Taliban were almost exclusively Pashtun, but they were seen as being from Pakistan, and their fighters and leaders acted as Pashtun chauvinists, which alienated potential support from Afghan Hazaras and Tajiks). Foreign ideas may attract foreign proponents and administrators, but not Afghan administrators. In neither case would Afghanistan as such benefit from a sufficient mass of skilled public servants capable of forming and supporting a centralized state (much less a ‘democratic’ one).
OEF forced the unpopular Taliban to flee, and in keeping with state-of-the-art COIN strategy, promised to rebuild Afghanistan’s non-criminal economy. ISAF set about importing administrators and attempting to rebuild the country from Kabul. It did not work. Ultimately, ISAF and foreign aid groups shifted from a top-down to a bottom-up COIN and development strategy, which increased short-term effectiveness but at a cost of developing public servants without a territory-wide identity.
Theoretical implications
In terms of theoretical implications it is best to start with a simple explanation of why Afghanistan, and places similar to it, matter, and why it makes sense to devote resources to helping them.
The first major entries in the post-Cold-War discourse on failed states must count Robert Kaplan’s ‘The Coming Anarchy’, an essay he penned for Atlantic Monthly in 1994 (Kaplan, 1994). In it, Kaplan argued that many states in Africa had disintegrated (we would say ‘failed’ today), and that what economists might call ‘negative externalities’ – including plague, refugees, war, genocide, environmental degradation and organized criminal activity – were likely to spread to the advanced-industrial world.
The first argument – that failed states produced negative externalities – was his strongest. His second argument – that these negative externalities would harm citizens in the developed world – was weak at the time, but proved prescient. After 9/11, it became clear that failed states (Afghanistan) could be sites of negative externalities that could not only reach the developed world, but could do so in lethal fashion.
There are other depressing examples. Chechnya, after de facto independence in 1996, became effectively a state-level criminal enterprise, which began to kidnap and murder Russian Federation citizens, among others. Robert Mugabe’s Zimbabwe is another example. Beyond the destruction of its economy, the collapse of Zimbabwe’s healthcare infrastructure made it the site of a cholera epidemic that killed thousands in Zimbabwe, and hundreds in neighboring states.
In each case, the international community asked the wrong question: what mix of resources can we supply in order to support rebuilding the state? It did not ask ‘are there sufficient people in this territory who are both interested in, and capable of, running a state?’ In an increasing number of countries where the state as a form of political association has been imported, that answer is ‘no’. Political actors – tribes, clans, ethnic and religious groups – want other things, and do not see the value in undertaking the burden of rebuilding and maintaining a state. This may be due to a moral hazard problem: decades of success in preventing interstate war (Goldstein, 2011) has meant that the relative benefits of life within states have declined as compared with the relative costs. For states in the advanced-industrial world, however, the dilemma is clear: either accept the negative externalities that flow from these ‘failed’ states, which appear to be escalating in intensity and frequency over time, or mount a costly intervention and import public servants (administrators) capable of running a functional state (neocolonialism) (Arreguín-Toft, 2005; Pape, 2005).
Afghanistan is a perfect example of the dilemma. The Soviets and ISAF both imported public servants whose mission was to provide security for a centralized state governed from from Kabul. That makes them neocolonial enterprises, a liability that served insurgent propaganda well.
Logically, the solution to the dilemma is to mount an intervention and transitional occupation analogous in scale and duration to that undertaken by the United States in Germany and Japan following World War II. However, this is politically impossible, both because the international community does not have the resources to do so (the investment would be dramatically more costly in the short term than the short-term costs of crime, terrorism, plague and environmental degradation), and because it would be seen as a colonial, illegitimate and dangerous precedent.
This is to say nothing of the fact that, however hostile to interstate peace, both Germany and Japan were real states prior to the war, and after the war, each possessed a critical mass of public servants sufficient to govern. This is not true in Afghanistan, and it is not true in Chechnya, Somalia or the Democratic Republic of Congo (to cite but three).
I note in closing one other important case which supports my argument: Iraq after 2003. Again, the details of coalition efforts to transition out of Iraq following the defeat of Saddam Hussein’s government are best read elsewhere, but on one feature of the Coalition Provisional Authority’s (CPA’s) early actions most historians of the period agree: its first two orders proved disastrous. CPA Order Number One was to summarily dismiss Iraq’s public servants from their posts, on the grounds that they were Ba’ath Party loyalists. CPA Order Number Two was to summarily dismiss Iraq’s army. In short, the first two acts of the CPA destroyed any possibility of Iraq governing itself as a prelude to the formation of an Iraqi government. To be sure, the CPA imported its own public servants to help manage the transition, but like its chief, L. Paul Bremmer III, these imported foreign administrators had no knowledge of Iraq, its history or culture. No government meant no security, and within a few months coalition forces began to reap the whirlwind in the form of an organized insurgency that continues to this day.
In terms of theory, therefore, we have been asking the wrong question: not how to restore the state, but how to restore demand for states without recourse to interstate war, which today, as ever, contains powerful possibilities for expansion and escalation. Alternately, might other forms of political association besides states not only work to satisfy their citizens’ basic needs, but be able to do so without creating the negative externalities that we associate today with the often mis-named ‘failed’ states?
Policy implications
The policy implications of this analysis are simple. I have argued that the chief reason state building – the overarching goal of all efforts in Afghanistan since 2001 – cannot succeed in Afghanistan is because the territory itself has been Balkanized by successive exogenous shocks into a loose confederation of micro states, whose citizens each depend on and value their own valley or province more than what has for them become an abstract concept: Afghanistan.
If I am right, then in order for Afghanistan to become a state – and here the aim is modest, not a democratic or free-trading state, but a functional state – then it must develop and maintain a corpus of public servants loyal to that larger entity. How can this happen?
The prospects are not good. First, businesses and educated Afghans both appear likely to exit once the ISAF departs (Carberry, 2012). As predicted by the logic of my argument, the chief reasons cited are lack of security and, in the case of public servants, lack of a demand for skills other than war fighting and narco trafficking.
Second, one logical way to get Afghans to unify under a central government is to threaten them with conquest and foreign occupation. Given the repeated failures of a host of foreign interventions over the past 30 years, it is difficult to imagine such a threat being credible.
Oddly, one ray of hope in this category might be the Taliban themselves. Their desire to complete the work of the Taliban of the 1990s might be sufficient to swing support to a skilled and respected Afghan leader capable of forging a new relationship between the provinces and the urban centers. This is in fact the best hope for the re-emergence of a functional Afghan state.
Third and finally, Pakistan continues to exert sufficient ‘spoiler’ costs to complicate the emergence of a new equilibrium between social change and tradition in Afghanistan.
Ultimately the best policy option for the well-intentioned or self-interested (or both) foreign assistance donors is to leave. The short-term costs of following such a policy will not be negligible, either for the Afghan people (especially Afghan women), who can expect years of privation and violence before things improve, nor for the current recipients of Afghanistan’s negative externalities (heroin addiction, drug-related crime and terrorism in particular). However, leaving Afghanistan to its own devices is the only way that Afghanistan will have a chance to begin to build the basis of a unified confederal state with a non-criminal economy. Once Afghans demand it, and once that demand translates into a sufficient quantity and quality of public servants capable of governing a state, then and only then might the international community supply what is lacking in order to facilitate an Afghan vision of the territory’s and its people’s future.
Footnotes
Funding
This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.