Full Text

CORRUPTION, MORE often than not, seems to resemble a plague. Afghanistan, where the CIA and British intelligence (in competition with the Iranians) have quite literally been handing over duffel bags stuffed full with taxpayer money to Prime Minister Hamid Karzai and his associates, is perhaps the most prominent example of its invasiveness and hardiness. Nothing seems to be able to eradicate it. Immunization efforts fail. Mutations occur. The only course seems to be to attempt to adapt to it. For despite the efforts expended by several American presidents on behalf of Karzai's administration, the United States has no surer way of ensuring influence and access to Karzai and his advisers than through direct cash payments into a slush fund designed to purchase the loyalty of important and powerful personages within the Afghan government. The bankruptcy of the Western strategy in Afghanistan could hardly be expressed in more vivid terms. Such failures in Afghanistan, not to mention Iraq, have occurred while the broader (and noncoercive) dimensions of "state building" or more generally "development" have also paid less-than-stellar returns. It is difficult to avoid the conclusion that the project of implanting "good" institutions in non-Western societies, whether through conquest (as in Iraq and Afghanistan) or through consensual, noncoercive means (as in Cambodia), has turned out to be a thankless task.

But is corruption really the source of the problem? Rather than viewing it as a pathology, as most Westerners seem to do, it is better to understand it as a type of currency used to establish and manage power relationships under certain systems of authority. As such, it is neither inherently unstable nor illegitimate. If the international community wants to eradicate corruption in the developing world, it is imperative to understand what it is, how it works, and why it is a potentially stable and legitimate system. Doing this requires stepping back and viewing the evolution of political order through a different set of lenses than most people are accustomed to, but the potential payoff for doing so is a greater sensitivity to how foreign societies actually work-and a deeper understanding of why changing them is so very difficult.

FOR SEVERAL millennia now human societies have created political structures that can be termed "states." What we call the "modern state," however, is a historically unique phenomenon that emerged organically in Western Europe by the nineteenth century and has been characteristic of Western political organization ever since. This modern state is defined by several characteristics, each of which is necessary for an entity to be properly termed a state. These characteristics are a monopoly on legitimate violence over a defined territory and population over which no higher authority exists. Defined as such, it is clear that what is being described is a type of political authority and critically, not a type of administration. Having a monopoly on legitimate violence over a defined territory and population does in fact require organization and administration. It does not, however, require a particular type of "administration" (specifically, it does not require what Max Weber called a "rational-legal" bureaucracy). It is, as is well understood, entirely possible (and logically coherent) to have a modern state operate according to principles other than those that define modern Western societies (that is, an administration or bureaucracy that functions on the basis of meritocracy underpinned by specific liberal notions of fairness and ethical conduct). In other words, it is critical to bear in mind that the problem that the development community is seeking to confront is not primarily a problem of administration (although these too of course exist): it is, instead, primarily a problem of authority. Having a modern state (replete with modern administrative forms) does not imply anything about how power and authority actually function within it. The modern state does not create the emergence of modern legal-rational forms of authority, and so simply creating those administrative structures will not do anything to guarantee that they actually function internally in a salutary way.

The World Bank has pithily described the development project's essence as how to turn the Congo into Denmark. The logical place to start answering this question is by examining how Denmark became "Denmark." When these policy makers turn to the academic literature on Western political development, what they find is that it is principally a story of the emergence of modern forms of governance and above all those of the state. There is an implicit and explicit assumption in the European political-development literature that the emergence of the modern West is tied to the rise of administration. The wonders of modern Western civilization become the positive externalities of the emergence of the formal structures of the modern state. One important implication of this is that much of the academic literature on "state formation" actually focuses on the wrong end of the problem: it is focused on explaining how the administrative structures of European societies evolved and attained their "modern" forms when the real problem is not the existence of bureaucracy but rather its internal logic. And this internal logic is covered by the domain of the concept of political authority. Another way of putting this is that we need to distinguish between the sources of administration (the concern of the state-formation literature) and the sources of the actual behavior within them.

IT HAS been entirely normal in human history for one society to attempt to varying extents and with varying degrees of consequence to alter the institutions of another society. Such attempts can be and have been made across several dimensions of social organization, including religion, property and kinship, among others. The development project can be understood as being a part of a very long history of societies trying to change other ones to resemble themselves.

What the development community seeks to do is establish self-sustaining institutions that produce similar outcomes to those obtained in the West in societies that do not possess them. In its current garb, the development project seeks to remodel the social institutions of non-Western societies in order to change their human-welfare outcomes. The problem is that this, as is well understood by now, cannot simply be achieved by replicating administrative forms. The outcomes produced by Denmark are not merely the function of possessing good administrative institutions; rather, they are instead a consequence of the emergence of particular configurations of authority that are deeply rooted in the society. The "efficient" and "fair" functioning of Denmark's administrative institutions is a consequence of the fact that these administrative organs are staffed by Danes who take for granted certain patterns of authority in general. Danes did not become prosperous because of the Danish state: they became prosperous because of the way in which they organized their lives in general. The Danish state is a reflection and a consequence of this deeper change in the way in which Danish social organization evolved. Simply taking Danish-style administrative organs and transplanting them to Afghanistan cannot work because the people who would actually staff them would be Afghans, and Afghan institutions are infused with a different category of authority. The problem is at the level of individual human behavior and its aggregate consequences-not the formal institutional settings in which they are displayed.

The problem that the development community faces is that in order for much of its agenda to work it would have to confront the deeper reality that good administrative outcomes (in a modern rational-legal sense) require a particular kind of society: the administrative efficiencies of the modern West are underpinned by a specific configuration of authority that flows from the society into the state. It is the normal expectation of certain behavioral patterns and the punishment of deviance that sustain authority structures in general and specifically in the modern West. Authority structures in all societies are deeply rooted and ingrained in the very fabric of the daily lives of human beings. They are by definition legitimate (usually by tradition or something attempting to masquerade as such) and therefore shape the reasonable expectations that individuals have with one another. In order for the development project to succeed, it needs to overcome both the sheer inert force of tradition as well as the resistance of societies in which authority structures depend on certain patterns of power relations. Persuading bureaucrats that the real pattern of authority that governs their lives-and which they may have varying degrees of stakes in-is illegitimate requires demonstrating to them that it is in their best interest to engage in new types of behavior underpinned by new patterns of legitimate authority. And such change in authority structures is never easy even when there is legitimate demand for different configurations.

In order for the development project to succeed beyond the actual alleviation of human suffering it will have to find ways in which to incentivize individuals-and especially elites (unless we propose to decapitate them, elites are crucial)-to adhere to different standards of public and private conduct. Otherwise, all that will happen is the creation of administrative structures that will be penetrated and permeated by the natural incentive structures organic to the existing society. The key point of the recent history of the development project has been the failure of liberal mechanisms to prompt other societies to alter their institutional profiles, irrespective of the extent to which they possess the formal administrative organs of the modern state. Even in those cases where tremendous force and violence have been used, as in Iraq and Afghanistan, this project has failed to change the nature of authority in the broader society: the identity of the winners and the losers changes constantly but the fundamental dynamics are the same. And so we are back to the problem of corruption and authority.

HISTORICALLY, MOST societies that have possessed a sufficient amount of permanent public authority to be deemed a state have also been governed by other principles of authority based on patronage and clientage systems that were, of course, also closely intertwined with kinship. Societies such as these intertwine the "private" interests of particular individuals and their networks of kin, clients and patrons into the formal administrative organs of the "state." This, of course, had been the case in Western societies until relatively recently. The notion that someone ought not to profit from an office in an individual sense would have been unintelligible to most people who have ever lived in a state. The issue of profit here must be understood carefully. To say that an individual profits from his office by rent seeking is not a "corruption" of an individual or an entire system. The fundamental functioning of the system depends on systems of patrons and clients because it is through these networks that power is expressed and exercised. It is the grease that keeps the gears of the system running.

To take the ancient Roman example: from the beginning to the end of the empire, the political system-both administrative and military-was based on intertwining networks of kinship and clientage. Indeed, the Roman Empire itself was a consequence of the extension via many different mechanisms of Roman senatorial clientage networks over the entirety of the Mediterranean world and then some. When we speak of the Roman provincial administration, what we are referring to are the great Roman families governing provinces through local elites who were tied into the broader system of clientage. So while the Roman army may have created the empire in one sense, it was only through the drawing of other elites into the orbit of the great Roman senatorial families that Roman rule was actually conducted and sustained. When Romans confronted barbarians, they dealt with them in exactly the same way because there was no other alternative: Roman authority depended, ultimately, on the networks of patronage and clientage (occasioned with violence, of course) with various barbarian groups and therefore through relationships of dominance and control that both sides instinctively understood. In this context, to say that Caesar or Pompey or Crassus or any of the other Roman senators out to make a name and fortune for themselves "profited" from their office is to misunderstand what that really means. What was actually going on was that individuals and their networks used their own resources gained through long careers of office holding to enhance the power and dignity of the Roman people and their republic and later empire. The public interests of the state were completely dependent on the proper functioning of "private" networks based on clientage. Roman patrons had to profit from their offices because if they did not they would have been unable to carry out their duties as Roman patrons and thus to be useful agents of the Roman state. There cannot, therefore, be "corruption" in a strict sense in a society whose public and private authority structures are infused with kinship and clientage networks. It is quite simply the way things are done.

A FURTHER illustration of this point is granted by taking another famously "corrupt" place: India. Any attempt to understand corruption in India must begin with the recognition that the fundamental unit of analysis there is kinship, understood in a broad sense. In India the language of kinship infuses both public and private discourse. It is the measure of ethical conduct and propriety. It is the standard by which an individual's worth is adjudicated. All aspects of Indian social organization are infused by kinship and its precious networks, including the economic sphere (something that has been very much noticed bythe Economist) and of course the public sphere.

Indian social organization is also fundamentally defined by a variety of group identities that have become increasingly corporatist in nature and increasingly embedded into the fundamental way in which power is exercised. India, of course, began independence with a deeper experience of European institutions (outside those of the settler colonies) than any other society in the colonial period. And it therefore emerged with the best administrative inheritance in the postcolonial world. While the generation that led India to independence sought to create a rational-legal bureaucratic state, its leaders did not foresee the extent to which the state that they inherited (including its noncorrupt elite Indian Civil Service administrative structures and personnel) would become permeated by the authority patterns of the society, including the establishment of patrimonialism within the administrative organs of the state itself. And these patrimonial tendencies have only grown stronger since independence. Once the founding generation passed away, much of the value system that had underpinned the clean and relatively efficient administration in the early years gave way to increasingly strong dynastic and clientage networks. The fact that Rahul Gandhi is not the current prime minister of India is simply one of the more amusing puzzles of contemporary Indian politics: it is positively odd that he is not. And so the problem in India is not fundamentally a problem of administration; it is a problem of authority. "Corruption" is what makes the whole system work: without it the Indian state would be unable to placate all of the noisy constituencies demanding a cut of the patronage pie. The Indian state is the principal arena in which a vast network of patron-client and kinship relations mediates how the spoils of the system are distributed. Being powerful in India requires being able to deliver goods to family, friends and dependents: it is the very currency of power. Not using your office in this way means to deliberately alienate the networks of power and dependency upon which normal life depends.

It is also important to emphasize that the West tends to fixate on the inefficiencies and often-tragic outcomes that this system produces but fails to recognize the significance of the phenomenon as a whole. Indian political conflicts are mediated through patronage networks. The system can, therefore, be viewed as a species of conflict resolution. It is by offering rewards to powerful individuals and constituencies within the Indian body politic that the Indian state navigates its treacherous communal, regional, linguistic and, of course, religious cleavages. The fact that India has survived as a democratic state-let alone a "united" state-is a testament to the abilities of a system like this to produce relative stability. One of the few positive consequences of this system is that the Indian state has always been able to deploy its vast reservoir of patronage powers and status to bind to it those who would otherwise be potential opponents of its existence. One way to understand just how important this is for the overall stability of the Indian state is to consider what happens when these networks fail to function properly. The Sikh insurgency in the state of Punjab in the 1980s was in some sense (though not completely) a consequence of a failure to come to an agreement over the relative distribution of spoils in the greater Punjab area. And most of the violent regional conflicts that have simmered and occasionally flared since independence have a dimension of failed patronage distribution. The Indian state has to confront its problems with the tools that it has and pay due attention to established and legitimate mechanisms of conflict resolution. It is a terrible and tragic outcome of this logic that India cannot produce a Western-style administrative state. But this is a consequence of a deeper problem of authority. The bureaucracy is itself a mechanism used to solidify and tie important constituencies to the state; given that India is poor, there are never enough resources to fully operate the system. As a result, many populations are starved of patronage, and these then become potential and active opponents of the Indian state.

While India has indeed grown richer and more powerful over the past two decades, none of that progress is the product of a change in authority structures. Quite the contrary. The dynastic tendency in India and the system of patronage have simply grown larger and flashier, but the system remains unchanged and many hundreds of millions of people are the victims of this fact. But it must be understood that the patrimonial system that is India cannot simply be seen in moral terms (although it can, of course, also be viewed through a normative lens). This system does not exist because of "bad" people. Its existence is the result of the numerous compromises that the Indian state had to reach with the many potential opponents of the Indian experiment. And, more importantly, the elite have had no incentive to change it until recently. What is potentially revolutionary in India is not the public denunciation of corruption but rather the fact that the society is simultaneously shifting away from older patterns of kinship and association. In other words, if India is to become like the West in this respect it will be because the authority structures that sustain these practices have lost their legitimacy in a constituency that can do something about it. But it is very important to understand that what sustains this system is that it is in fact a legitimate and accepted way in which to conduct social relationships. The point is that it is a stable system, which is not to say that it is a "good" system.

What we call "corruption" has been the normal and legitimate practice of most human societies and can actually produce certain categories of good outcomes. In order to change a system like this (in the absence of endogenous demand as in, just perhaps, contemporary India) it is necessary to change the incentive structures of the society at a very micro and therefore basic level. We are talking about what the definition of ethical conduct is and the extent to which it is internally policed. Building formal institutions can in no way substitute for the creation of incentive structures that govern actual lives. And whatever else is true about other systems of social transformation that have existed historically, it appears that in our modern age there may be no way to use liberal means to attain liberal ends in nonliberal societies. Changing authority structures is a very big deal and historically has always been accompanied by violence and social dislocation. We cannot anticipate that individuals will alter their daily expectations of normal human interactions without causing an overall shift in the nature of the system of power relations.

ALL THIS brings us back to Hamid Karzai and those duffel bags stuffed with American tax dollars. The United States had a choice about intervention in Afghanistan. One option was a limited intervention along the lines of an imperial raid to settle some frontier and then leave, but done while working through the local authority structures to achieve those goals and accepting the limitation on influence and power that this implied. Or it could have attempted to restructure Afghan society on the grounds that the kind of threat that was emanating from it could only be fundamentally resolved by a change in the basic social organization of that society. Either way, the cooperation of elites would have had to have been secured on the basis of their legitimate interests. And in Afghanistan that meant that the regime through which the United States sought to achieve its goals would have had to establish its authority on the basis of the system that governed the assumptions of most of the people it was called to rule. Karzai had to have the resources with which to create clientage networks because there could be no other way for him to ensure that the administration would actually heed his orders. He had to place kin and trusted clients in key positions and they had to use their positions to further the overall network of influence because that was essential to Karzai's power and perhaps his very life.

And where could these resources possibly have come from? Well, the only significant revenue streams in Afghanistan have been drugs and aid, and both of these have fueled the overall system. The United States should not have been surprised. Indeed, it seems that the CIA, at the very least, understood that there were no other alternatives but to deliver sacks full of cash: Karzai could not rule Afghanistan otherwise, and without funding his patronage networks the United States would have no leverage over him. What is remarkable is not that the CIA chose to bribe Karzai; what is surprising is that it shocks us (and, of course, there is an added level of hypocrisy given the tremendous public and private pressure put on the Afghans to clean up their collective act). For we are now a liberal society, and in a liberal society it is very difficult to make the case that bribery and corruption may be the only tools we have at our disposal because we do not have the power to coerce them to become like us. It reeks and taps into the residual historical anger that caused Western administrations to become modern rational-legal ones in the first place. And yet, it is difficult not to conclude that, in some instances, corruption must be accepted as an undesirable but nonetheless potentially legitimate mechanism for engaging with societies organized along different lines. Perhaps it is time to give corruption a chance.

Vivek S. Sharma is a visiting associate professor of political science at the University of Copenhagen.