How we view corruption makes a difference. Take for example corruption in the process of land transactions in Cambodia. Cadastral officials double as “fixers” (neak rot ka). They charge the person trying to sell their land a lump sum to run their papers through the relevant offices in the process of formally registering their transaction. This money goes to paying both formal and informal, or under-the-table, fees. Typically, the fixer underprices the land in order to lower the seller’s tax liability and free up money to cover the informal costs of the operation. In the end, however, the seller usually pays as much or more than they would have had the process been clean (So Reference So2009, 165).

Such rent-seeking is hardly a mysterious thing, according to the prevailing view. We can diagnose it without further information because it happens everywhere. This is a case of actors breaking the rules and violating their trust as public officials. We know why they are doing it: greed or self-interest. Although we may be aware of mitigating circumstances, in this case that the salaries of cadastral officials are abysmally low and irregularly paid, it doesn’t change the fact that rent-seeking is bad behavior. It gets in the way of the formal process working properly and results in distinct costs to society. It diverts money away from public coffers and into private pockets and makes formal registration more expensive, putting it out of reach for most Cambodians, who must resort to informal and less secure transactions. Thankfully, we know how to stop rent-seeking, in theory at least. We have to change the incentives surrounding the practice, for instance by raising the salaries of civil servants, paying them on time, training them better, penalizing further infractions, and perhaps also making the whole process more efficient by lowering land taxes or automating transactions.

But is that all there is to it? What happens if we try to understand the bad behavior of cadastral officials in the larger canvas of Cambodian society and history? We might identify three processes as bearing on the “construction” of cadastral rent-seeking: the development of a land market and regime of private property rights, the development of a political order based on patronage, and the development of procedures for handling land transactions.

Economic restructuring in the 1990s and 2000s bred a new kind of insecurity. People had more money than before but needed it more than ever, and often found themselves in debt as a result of various financial shocks (Ballard et al. Reference Ballard, Sloth, Wharton, Fitzgerald, Murshid, Hansen, Runsinarith and Sovannara2007). Insecurity was not just economic but normative. Land had acquired enormous value as a commodity, but the rules regulating its transaction, being new, lacked legitimacy. The new rules clashed with the old, and most Cambodians found them confusing and difficult to follow (Grimsditch et al. Reference Grimsditch, Leakhana and Sherchan2012, 97). People with office or means, however, were able to exploit these rules. Given a situation of normative ambiguity, corruption becomes less about the violation of the rules than about people’s changing relationship to them, including, crucially, their ability to manipulate or simply follow them.

The development of a land market went hand in hand with the establishment of a political order based on patronage. While patron–client ties have long characterized Cambodian politics, contemporary patronage networks were largely created to capture the resources unleashed by economic restructuring. Within this order, people buy appointments to senior government positions and, once installed, are expected not only to recoup their “investment” but also to make money for their political bosses (Un Reference Un2004, 135). These clients are themselves patrons to clients down the line, including our lowly fixer in the cadastral office. The “rent” money he collects is divided into three portions at the end of every month. One portion is distributed within the office, with a greater share going to higher-ranked officials. Another portion is sent upward to senior officials in the land ministry. The third portion is kept for general expenditures, including contributions to the electoral slush fund of the ruling Cambodia People’s Party (CPP) (So Reference So2009, 199–200). These flows – of money, favors, and obeisance – are binding. They tie together state agents big and small and cannot simply be cut off in one domain without the repercussions being felt in another.

Now let’s take a look inside the cadastral department itself. Since its founding in 1989, it has been hamstrung by several problems. The agency lacks resources and equipment and its agents education and training. The most pressing problem, however, is that the relevant actors lack sufficient “incentive” to do their jobs. (The term incentive here does not appeal to universal self-interest but to the need for money constructed in the process of economic restructuring.) The formal registration process is complex and requires that multiple actors coordinate their activities. The seller has to file a land transfer registration form requiring the approval of the district and provincial governors. Then they have to have the sales agreement authenticated by cadastral officials and pay a fixed service fee. Finally, they have to render the 4 percent sales tax to the tax department. They can only advance to the next step by completing the one before it. Without “lubrication,” the bureaucratic machinery is liable to stall.

Given this situation, we might view rent-seeking as a way to solve, or at least mitigate, these problems. Rent-seeking generates the resources necessary for bureaucratic functioning. It provides the various actors involved in the registration process with incentive to coordinate their activities both “horizontally” (between tax and cadastral officials) and “vertically” (between rank-and-file bureaucrats and their patrons in the land ministry and CPP). Moreover, by navigating a process that most Cambodians find bewildering, the fixer provides a real service.

Embedding corruption in these processes of social organization fundamentally changes how we understand it. We gain a more complex account. It becomes clear that there is more to rent-seeking than a strictly utilitarian logic. We see the practice being worked out within structural situations distinguished by normative ambiguity, neopatrimonialism, and organizational incapacity. It is within these situations that self-interest is constructed. We see actors develop a course of action in response to organizational pressures, constraints, and opportunities. This knowledge changes how we view and evaluate them. The venerable distinction between corrupt and organizationally minded or “Weberian” bureaucrats starts to blur. We begin to wonder whether the fixer is obstructing the formal process or the one responsible for actually making it work.

A more complex account doesn’t make us blind to the costs of corruption but it does complicate our efforts to solve the problem as such. Consider: The World Bank (2000) published a report on corruption in Cambodia recommending that the government pay civil servants a living wage, promote them on the basis of performance, allow for greater public oversight of their duties, and require them to declare their assets. These recommendations may look great on paper, but it is hard to imagine how they would work in practice given what we know. We know that the problem of corruption is larger than the individual; it is organizational and systemic. People are enmeshed in social networks organized around patronage and rent-seeking. Higher pay may make them less desperate, but it will not be enough to extricate them from these networks. Further, the recommendations assume that the Cambodian state has the resources and capacity to implement them and, more importantly, that it will pursue them in good faith, that is, that they will not be disfigured by the logic of patronage. Clearly, reality is more complicated – but that’s the point. We need to be able to appreciate these complications and take them into account if we hope to develop a more realistic approach to the problem.

***

The utilitarian approach identifies agency with the individual and thus espouses a view of action as largely disembedded from society and history. It represents corruption in the abstract, as a generic problem. This leaves us with a weak grasp of it as a social and historical object and a false sense of control over the problem.

In opposition to this view, we argue for understanding corruption as embedded in social space and time. Embedding is a basic move in sociology, perhaps the fundamental one. Study the social fact within its proper milieu, Durkheim (Reference Durkheim2014 [1895]) exhorted in The Rules of Sociological Method. The reason, of course, is because social and historical context is constitutive of the object itself. Stripping it away represents a “scientific disaster” (Abbott Reference Abbot1997, 1171). We end up reducing the phenomenon to its definition; in the case of corruption, typically, some version of Nye’s (Reference Nye1967) classic formulation: the abuse of public office for private gain.Footnote ¹ The definition alone is insufficient to identify the object sociologically. It is only the beginning of investigation. The crucial work is embedding or supplying the context structuring the phenomenon. While there are many viable ways to embed corruption, as we will see, we advocate a comparative historical sociology of corruption in particular. This approach has in mind a view of corruption as “a moving object,” that is, as subject to variation across social space and transformation over time. It focuses on the processes through which a course of action (practice) is worked out in relation to historically specific structural conditions. By tracing these processes and embedding “corrupt” practices in the situations where they were developed and make sense, we gain a deeper understanding of these practices and are in a better position to evaluate them. We are also able to make better comparisons, comparing objects shaped by similar processes rather than objects identified by definition alone.

This approach represents an intervention in the scholarship on corruption today. The questions characterizing the early scholarship about how to understand and evaluate “corrupt” practices appear settled. In the prevailing framework, corrupt action is obvious; it is clear what it is and why people do it, and it is obviously bad. The only real question is how to stop it, hence the single-minded focus on anti-corruption. This framework looms so large in the contemporary scholarship on corruption, particularly on corruption in the developing world, that it gets in the way of understanding the realities on the ground. The aim of this volume is to recover a more sociological way of seeing corruption and, in particular, make the case for a comparative historical sociology of corruption. We proceed, first, by situating our intervention in the context of the rise of a disembedded approach to corruption. Here we trace the development of the scholarship from the “old sociology of corruption” in the 1960s to the asociological view behind contemporary anti-corruptionism. Second, we review a selection of more or less embedded approaches in anthropology and sociology with the intent of showcasing their diversity and the manifold possibilities they index for studying corruption. Third, we describe what a comparative historical sociology of corruption entails. Fourth, we highlight the stakes, arguing that a disembedded view of corruption encourages two kinds of misperceptions, idealism and essentialism, with real-world costs. Finally, we preview the chapters ahead.

1.1 A Brief History of Disembedding: From the Old Sociology of Corruption to Contemporary Anti-corruptionism

A view of corruption as embedded used to be central to the study of corruption. In the old “sociology of corruption” of the 1960s, McMullan (Reference McMullan1961), Leys (Reference Leys1965), Nye (Reference Nye1967), Huntington (Reference Huntington1968), Scott (Reference Scott1969), and others – political scientists and yet deeply sociological – approached corruption as bound up in the process of modernization. On the one hand, they defined corruption in terms of the transgression of the public/private distinction. Corruption, Leys (Reference Leys1965, 221) wrote, “breaks some rule, written or unwritten, about the proper purposes to which a public office should be put.” On the other hand, they understood this rule to be a construction, an ideal everywhere but particularly difficult if not impossible to follow for countries in the throes of modernization. This was because modernization made the public/private distinction salient in the first place. It also created new sources of wealth and power and led to the expansion of public roles and activities. These changes served to define the category of corruption as well as provide both incentive and opportunity for corrupt behavior. In other words, corruption emerged in the process of modernization. Within this context, transgressive behavior was all but inevitable given the novelty and fragility of modern standards. Huntington (Reference Huntington1968) pointed to the weak institutionalization of these standards and McMullan (Reference McMullan1961) to the discrepancy between them and attitudes on the ground.

These scholars went beyond grounding corruption structurally. They saw it as functional. Corruption fulfilled certain requirements associated with building a market economy and modern state. It facilitated political participation. “At a high level [corruption] throws a bridge between those who hold political power and those who control wealth” and at a lower one it serves to incorporate subordinate groups into the political system (McMullan Reference McMullan1961, 196). It stimulated economic development by enabling capital accumulation and the circumvention of bureaucratic hurdles. It helped build political parties by generating the resources used for patronage (Huntington Reference Huntington1968, 59–71).

Despite their functionalism, early corruption scholars were not insensible to its costs. McMullan (Reference McMullan1961) begins his “Theory of Corruption” with a litany of the ills caused by corruption: injustice (for the poor, whom corruption puts at a disadvantage), inefficiency, mistrust of government, waste of public resources, discouragement of enterprise, political instability, and so on. “Understanding is desirable,” he remarked, “but it is wrong to underrate the evil consequences of widespread corruption” (181). This acknowledgment was tempered by a structural analysis, however, one which preceded and informed thinking about interventions; hence corruption was not colored entirely by an anti-corruption attitude, as it tends to be today. In general, these scholars rejected a “moralizing approach” to corruption as being myopic, involving a view of corruption stripped of its structural determinants, and ethnocentric, involving some version of the complaint “Why does the public morality of African states not conform to the British?” (Leys Reference Leys1965, 60).

The impetus to do something about corruption was also, perhaps, allayed by the faith that corruption would sort itself out in the course of modernization. Huntington (Reference Huntington1968) posited that corrupt behaviors would diminish over time but also that some of these behaviors would become accepted and even formalized, and in this way behaviors and rules would align. In other words, he believed that eventually political institutions would consolidate and the public/private distinction become ingrained. Scott (Reference Scott1969, 1156) pointed to political machines in the United States dying “a more or less ‘natural’ death” as a result of industrialization. People became less poor and thus less susceptible to material blandishments. They developed horizontal loyalties to social class and occupational groups, which superseded their vertical ties to political patrons and parties. Scott supposed the same would happen in transitional countries. There was some uncertainty on this point, however, given the recalcitrant realities of development. Scott concludes his paper wondering why political machines in the “new nations” were not developing as quickly or in the same way as in the United States. McMullan (Reference McMullan1961, 181) notes that if corruption is “a passing phase” then it is unlikely to pass quickly.

By the 1970s, this approach had fallen out of fashion. A new generation of scholars rejected its functionalism and teleological bent. Rightly so, in our view. The functionalism of these accounts left little room for human agency. If corruption exists to fulfill developmental imperatives, then corrupt behavior is mechanical and actors empty vessels simply carrying out their preordained functions. This oversocialized view of human action as dictated by system requirements is untenable theoretically (Wrong Reference Wrong1961). The account’s teleological bent is an artifact of its association with modernization theory. The obvious problem is that it has been sixty years since these accounts predicted that corruption would be “tamed” and yet it continues to plague nations no longer new. The real issue is not how long it is taking for corruption to pass away – adherents of modernization theory would probably argue that it has not been long enough – but rather that the faith that developing countries are moving in the right direction, as opposed to being stuck in place, has been profoundly shaken. Sociologists largely turned away from the study of corruption. Some approached the topic critically, adopting world-systems and dependency perspectives, and idiographically, exploring the diversity of practices and meanings of corruption on the ground (Osrecki Reference Osrecki2017).Footnote ²

By the early 1990s, the study of corruption had been taken over by a disembedded view. The proponents of the new model were primarily economists in academia and the World Bank. They understood corruption within a rational actor/principal-agent framework, basically as individual utility-maximizing behavior. This framework was adopted and elevated by powerful organizations, particularly the World Bank and International Monetary Fund (IMF) (with which the chief proponents of the economic model, Susan Rose-Ackerman and Vito Tanzi, were affiliated), as well as the US government, the Organisation for Economic Co-operation and Development (OECD), and the United Nations (UN). Key events helped put corruption, so conceived, on the global policy map: the founding of Transparency International in 1993, a speech given by World Bank president James Wolfensohn denouncing the “cancer of corruption” in 1996, the signing of the OECD Anti-Bribery Convention in 1997, and the passage of the UN Convention against Corruption in 2003. Further, the new model complemented the broader push for economic liberalization ongoing at the time. A view of corruption as an impediment to market integration and economic growth supported prevailing development policy (Hough Reference Hough2013).

A narrow focus on bribery facilitated the quantification of corruption and its comparison and ranking globally (Johnston Reference Johnston2005). International organizations, governments, and corporate service firms (tasked by companies with assessing the financial risk of their investments abroad) began to generate troves of new cross-national quantitative data on corruption. These data represented a particular way of knowing corruption and constructing the problem. Study after study linked corruption to underdevelopment, nondemocratic governance, and all manner of social harm, including poverty, poor health, low life expectancy, and inequality (Ades and DiTella Reference Ades and DiTella1997; Treisman Reference Treisman2000; Montinola and Jackman Reference Montinola and Jackman2002; Lambsdorff Reference Lambsdorff and Rose-Ackerman2006). Not incidentally, these studies tend to adopt the conventional definition of corruption as a violation of the public/private divide. In so doing, they take for granted an ideal largely associated with countries in the so-called developed world. They assume a context where the public/private distinction is bright and political institutions relatively strong. Under these conditions, corruption represents deviation from a “Weberian” standard of rational-legal bureaucratic procedure. Is it any surprise then that corruption is found to be prevalent in places where these conditions do not hold? Or any wonder that Western countries cluster at the top of the rankings? Treisman’s (Reference Treisman2007) review of the corruption scholarship in economics and political science highlights this circularity. Corruption is conceived in terms of underdevelopment and then discovered to be associated with a lack of development. To describe the mutual shaping of measures and concepts, Picci (Reference Picci2024, 50) resorts to the visual metaphor of M. C. Escher’s Drawing Hands. “One hand, corruption, drew the other, its measures, which in turn partly shaped our understanding of corruption.”

The ranking of countries in terms of corruption is also problematic. The preeminent scale, Transparency International’s Corruption Perception Index (CPI), measures perceptions of corruption and not actual incidence and then mainly the perceptions of experts and Western business leaders. It is not clear whether respondents in different countries are interpreting corruption in the same way. The interval scale gives a false impression of accuracy (Andersson and Heywood Reference Andersson and Heywood2009). Despite these issues, the CPI remains hugely influential. It is used in academic research and to inform policy decisions. In some cases, donors make development aid conditional on a country improving its corruption score. The annual publication of CPI scores is front-page news in many countries and a topic of fervid discussion on the ground. A country’s ranking shapes how people view their governments and themselves.

Ranking and the focus on costs informed an approach to corruption almost exclusively in terms of anti-corruption. The aim of this approach was not to understand corruption qua social phenomenon, as it had been historically, but simply to eradicate it. The rationale behind anti-corruptionism expanded to include political goods in the 2000s. Corruption came to be articulated within a good governance framework emphasizing strong institutions, the rule of law, and provision of key public services – “Denmark” in short, as scholars affiliated with the World Bank referred to it (Pritchett and Woolcock Reference Pritchett and Woolcock2003). In this framework, corruption stands in the way of getting to Denmark, while anti-corruption means promoting the three institutional virtues of transparency, accountability, and integrity (Kaufmann Reference Kaufmann2009).

The anti-corruption field grew enormously over the course of the 2000s. Michael and Bowser (Reference Michael, Bowser, Wolf and Schmidt-Pfister2010) estimate the value of goods and services procured by donor agencies for anti-corruption projects to have ballooned from $100 million in 2003 to almost $5 billion in 2009. Sampson (Reference Sampson2010) describes the formation of an anti-corruption “industry” consisting of a set of actors organized around a common framework, inspired by key texts, making use of the same tools and indicators, and implementing similar projects and programs. It has even spawned an academic critique serving more to legitimize the framework than seriously challenge it. To call the anti-corruption field an industry is, of course, a criticism. It suggests that a degree of standardization has set in, narrowing the thinking on corruption and taking important issues and questions off the table because they fall outside of or are at odds with corruption orthodoxy. As Johnston (Reference Johnston2005, 18) put it, the problem is “too much consensus.”

The other issue is the moralization of corruption. A recitation of the evils of corruption has become de rigueur in the anti-corruption literature. Work in this genre tends to be punctuated by quantitative and qualitative accounts of harm. These accounts do more than illustrate; they serve to justify the anti-corruption agenda. According to one account (Rothstein Reference Rothstein2011), there is no drinking water in Luanda because of corruption, property rights are tenuous in Santa Lucia because of corruption, and because of the corruption in public hospitals in India, poor mothers cannot hold their babies, who are reportedly withheld until a fee is paid. We would not deny or minimize corruption-related harms but would, however, complicate simplistic and exoticized assessments of them.

Today, fighting corruption is seen as God’s work, a moral project, Sampson (Reference Sampson, Haller and Shore2005) argues. It is clear who the good guys – the so-called integrity warriors – and who the bad guys are. The moralistic construction of the problem produces a sense of urgency. It mandates that we “do something.” The problem is that the feel-good quality of the work supplants a more reflexive posture. The sense that “we’re doing something” can keep us from asking whether what we are doing is any good. Pursuing this question has become more important than ever given the poor record of anti-corruption initiatives even by the World Bank’s own reckoning (Kaufmann Reference Kaufmann2009). Brazil’s Lava Jato scandal prompted the economist Lucio Picci (Reference Picci2024) to rethink corruption in a more sociological key. It was not just the naked politicization of anti-corruption efforts so apparent in Lava Jato that gave him pause but the failure of these efforts in most of the world. In recent decades, only a few, mainly small countries have brought corruption under control (Picci Reference Picci2024, 5; see also Mungiu-Pippidi Reference Mungiu-Pippidi, Mungiu-Pippidi and Johnston2017, 12–22). The real question, Picci writes, is not whether anti-corruption efforts are sufficient but whether the paradigm informing them is sound.

1.2 The Disembedded View and Its Discontents: A Review of Embedded Approaches in Anthropology and Sociology

The dominant view is premised on a utilitarian model of corruption. In this model, corruption is conceived as individual utility-maximizing behavior (Rose-Ackerman 1978, Reference Rose-Ackerman1999). It may take on various cultural hues but there is something universal at the core of it: self-interest. Simply put, people abuse public office or public trust because it is in their self-interest to do so. The issue then is how to better manage self-interest. The problem has less to do with bad actors than with bad incentives, and thus reform requires getting the incentives right and channeling self-interest more productively. A good part of the model’s appeal is its theoretical coherence, even elegance, and the fact that it provides a clear solution to corruption. “Getting the incentives right” tends to mean political and economic restructuring in the direction of more competition, freer markers, and smaller and more efficient government. This account is disembedded from social context insofar as it grounds action in individual self-interest. Because self-interest is universal, corruption can be generalized as basically the same thing everywhere – and inherently a bad thing because it gets in the way of proper processes.

Over time dissatisfaction with a purely utilitarian approach deepened. Political scientists in particular took issue with the lumping together of behaviors which looked like corruption in economic terms but meant different things in different contexts. They increasingly rejected a one-size-fits-all approach as glossing over significant variations in the organization and practice of corruption. The particular discovery was that context matters, and by the 2000s, certainly, a context turn had become discernable in the scholarship in repudiation of the prior behavioral one (Hough Reference Hough2013).

Alina Mungiu-Pippidi (Reference Mungiu-Pippidi2006) argued that corruption in developed countries is different from corruption in developing and post-communist ones. The former had regimes characterized by “universalism” and the latter by “particularism” or competitive particularism. Specifically, the public/private divide was sharp in one context and fuzzy in the other, and thus corruption meant different things. It represented an individual infringement on well-established “norms of integrity” in developed countries whereas in developing ones it was a mode of social organization, a collective adaptation to weak institutions. Consequently, anti-corruption strategies had to be tailored to regime type – for instance, it would be less fruitful to appeal to the government in places where particularism reigned. Similarly, Johnston (Reference Johnston2005) distinguished between different syndromes of corruption in contexts with different combinations of political and economic participation and institutions. Hough (Reference Hough2013) linked specific anti-corruption measures to contexts differentiated by quality of governance. There have also been efforts to understand corruption in terms of the larger political economy at stake and as not necessarily incompatible with economic development.Footnote ³ Indeed, as should be evident from the previous section, we have been building upon many of the criticisms of political scientists and economists themselves, particularly Michael Johnston and Lucio Picci. In short, the scholarship on corruption in political science and even economics is not uniformly disembedded.

From our perspective, however, most of these efforts don’t go far enough. Linking corruption to ideal-typical contexts is not the same as showing how the object has been constructed in specific ones. The aim of these efforts is not to embed corruption but to elaborate a framework better able to accommodate significant variation in the forms of corruption. That is, they seek to enhance the concept of corruption. We are interested, however, in understanding the phenomenon concretely. Thus, we turn to the scholarship in anthropology and particularly sociology, where embedding is central to most approaches to corruption. We highlight the diversity of these literatures, organize them, and underscore a comparative historical approach as being of particular importance today.

1.3 A Comparative Historical Sociology of Corruption

The aim of a comparative historical sociology of corruption is not to provide a better definition of corruption or a more refined typology. It is not to elaborate corruption conceptually but, rather, to describe it in its constitutive contexts, including social space and time. Besides, the conventional definition of modern corruption as the abuse of public trust is usually sufficient to start with because it is only that, a starting point. It is an ideal type serving to bracket a piece of reality for investigation – a means to an end (Weber Reference Weber1949 [1904], 90–95). The end is something else entirely; not a better definition, typology, or classification but a better understanding of how the practices in question came to be, why they persist, and how they are understood and utilized on the ground.

The procedure is to embed, embed, and compare. We configure corrupt practices within the structures or sets of social relations we deem relevant to their emergence and institutionalization (embedding socially), we trace the development of these structures over time (embedding historically), and then we compare not the behaviors alone abstracted from context but the social processes producing different corruption milieus. Corruption may not be the same thing everywhere (universal), but it is also not a wholly different thing in every case (radically particular). We can generalize meaningfully across a set of cases not because the practices we are interested in are by definition the same but because the processes underlying those practices are similar (they foster a certain relation to the state, for instance). That is to say, we compare with an eye to social structures and processes and not simply categorical behavior. Such comparisons are socially and historically bounded, and more analytically coherent for it.

While this approach is compatible with many of the efforts to embed corruption in anthropology and sociology, it is distinguished by a central focus on process. To understand what we mean by process, we must dispense with the unhelpful distinctions between structure and its others – mainly culture, process, and agency (with structure usually the disfavored term). Structure refers to concrete, ongoing systems of social relations (e.g., Granovetter Reference Granovetter1985). These relations are laden with meaning and form the webs of significance Geertz (Reference Geertz1973) called culture. These meaningful relations are ever in the process of transformation, and thus, rather than structure we might speak of processes of structuration (Giddens Reference Giddens1984). In another sense, structure refers to durable patterns of thought and action (Durkheim Reference Durkheim2014 [1895]). These patterns are not automatically reproduced but have to be enacted, or actively constructed in situ. People have to “translate” the social pressures of a particular situation into appropriate behavior (March and Olsen Reference March and Olsen1989). They possess a degree of autonomy in doing so but must also contend with various contingencies, and thus outcomes may vary even when structural contexts are similar. With these considerations in mind, we use the word process to get at structure and culture taking shape in time as well as the progressive dynamic of structure and agency.Footnote ⁴

The focus of a comparative historical sociology of corruption is on empirically tracing the concrete, or historically specific, processes shaping the practices labeled corruption. To understand rent-seeking in post-conflict Cambodia, for instance, we embed the practice in the historical processes through which it acquired impetus and shape. The question of which processes are relevant is an empirical one. Earlier, we identified three – economic liberalization, state formation, and bureaucratization – but other analysts may highlight different processes. In general, we advocate studying corruption in terms of three generic processes: emergence, institutionalization, and mobilization. These rubrics serve to underscore the temporality of corruption. Corrupt practices have to emerge and may disappear. They become institutionalized (or deinstitutionalized) over time. Allegations of corruption are deployed to mobilize (or demobilize) people and resources. We have to understand this mobilization with respect to the normative projects and material interests of institutions and individuals as they develop over time.

Emergence is the process whereby corruption acquires a particular form and meaning. To illustrate: Verdery (Reference Verdery2003) followed the social fallout of decollectivization in a Transylvanian community in Romania over the course of the 1990s. Privatization meant the massive devaluation of wealth accumulated during the socialist period. Socialist property was recast in terms of market value – that is, what a foreign capitalist would be willing to pay. Firms were sold for the price of an expensive house and factories still capable of producing goods reduced to scrap metal. As people’s relationship to property changed, so did their relationships to one another. The title of Verdery’s book, The Vanishing Hectare, primarily refers not to the loss of land but to the loss of land’s meaning and value. As a result of privatization, it could no longer serve, as it once had, as a viable source of income and symbol of status and place. To say that the value of land is emergent means that it is not intrinsic to the thing itself (land) but underwritten by a complex web of social relations and their associated meanings. It is a property of social structure, and therefore to abstract it from this context is to misconstrue it.

Institutionalization is the process whereby corrupt practices are incorporated into larger systems of meaning and action and become entrenched. We might speak of the normalization of corruption. In many cases, corrupt behaviors represent solutions to organizational problems. These solutions may be suboptimal or dysfunctional, but they become institutionalized over time as behaviors are rationalized and become routine or baked into organizational procedures (Ashforth and Anand Reference Ashforth and Anand2003). Eventually, they form part of the stock of meanings and procedures to which a group resorts for guidance, passing into group or organizational culture (Vaughan Reference Vaughan1996). Newcomers are socialized into them, learning to view them as acceptable. Deviant or unconforming behavior is penalized, even when “deviance” in this context means following the formal rules. Indeed, once corruption is organizationally embedded, it is misleading to speak of deviance or to single out bad actors. At this point, the issue is conformity and the locus of responsibility largely, although perhaps never entirely, organizational (Silver and Geller Reference Silver and Geller1978). We might identify other processes of institutionalization; the articulation of practices in different domains, for instance. As we saw earlier, the rent-seeking activities of cadastral officials are tied to the patronage practices of Cambodian political leaders. Their ties create a kind of accountability between corrupt parties that works to lock in these practices and protect them from efforts at reform.

Mobilization processes concern the taking up and deployment of the label corruption for particular purposes. We should pay attention to how the label is used and situate its mobilization in the relevant “field of power,” that is, with respect to driving interests, agendas, and institutions. The critical scholarship on anti-corruptionism does this work by identifying the economic, political, and moral interests behind the explosion of anti-corruption activity, connecting the global war on corruption to specific agendas, such as neoliberalism and good governance, and interpreting anti-corruption campaigns in terms of local power struggles (e.g., Brown and Cloke Reference Brown and Cloke2004; De Sousa, Hindess, and Larmour Reference De Sousa, Hindess and Larmour2010; Sampson Reference Sampson2010). In a similar vein, Picci (Reference Picci2024) highlights the politicization of corruption. If corruption is a tool of government (lubricating transactions and providing officials with incentives to do their jobs), then anti-corruption is a tool of politics, as illustrated by the Lava Jato scandal. It provides symbolic materiel ripe for political mobilization.

Finally, we conceive a comparative historical sociology of corruption as an intervention. The relevant context is the disembedding of corruption in recent decades and the alarming extent to which, today, even among social scientists, corruption is regarded as a natural, as opposed to social and historical, object: as self-evident, universal, and evil. We regard this conception as blinding. There is, thus, a great need for reflexivity in the scholarship. Embedding enables reflexivity generally by putting corruption in its social and historical place and particularly by situating its mobilization with respect to particular interests, agendas, and institutions. It entails shifting from a primarily normative to a primarily sociological and historical register. We are led to approach corruption not as a problem to be solved but with a set of questions concerning its emergence, institutionalization, and mobilization.

In this respect, a comparative historical sociology of corruption is different from a problem-solving sociology, which, in practice, tends to mobilize a narrow definition of the problem. The latter approach emphasizes the costs of corruption as an impetus for action but seems less interested in looking beyond costs and complicating the picture, for instance, by recognizing the ideological accounting of these “costs.” Our approach would bracket the problem-solving effort in order to create space for understanding corruption in the Weberian sense, that is, grasping its significance in the concrete terms of the social reality in which it takes shape (Weber Reference Weber1949 [1904]). Some might discount our goal as “only” understanding, but we see this kind of understanding as a form of action. As Hannah Arendt (Reference Arendt and Kohn1994 [1954]) observed, understanding is harder to do than simply doing and yet, insofar as it serves to reconcile us to reality, imperative to feeling at home in the world. Understanding would serve to correct a reductive and moralistic conception of corruption. This conception is costly in its own right, as we will discuss in the following section.

1.4 The Costs of Anti-corruption

A disembedded view of corruption promotes two kinds of misperceptions, idealism and essentialism, with real-world costs.

First, idealism. Treating corruption as a generic problem can lead to policy prescriptions that are detached from the reality on the ground and hence inappropriate, ineffective, or perverse in their effects. To illustrate, let us return to the example of corruption in Cambodia’s land market. The country’s World Bank-sponsored land registration program was supposed to improve tenure security and prevent land grabbing. Nearly a decade on, however, the development worker turned academic Robin Biddulph (Reference Biddulph2010) concluded that it did not make much of a difference. It did not make people more secure, increase their productivity, or improve their access to credit. Rather, it led to more conflict, making people keen to title their land and claim communal land and pitting formal title holders against those with only informal titles. He argued that the intervention served primarily as cover for powerful interests. Private titling distracted from the selling off of public land on a massive scale. In addition to 3 million titles of questionable value, the program produced political capital; it served to appease international donors and could be touted domestically as a pro-poor initiative.

The gap between the ideal and reality of the titling effort was the result of systematic misrepresentation, Biddulph (Reference Biddulph2010, 215–219) writes. The land registration program was based on a “caricature” of rural life and an insensibility to the schemes of powerful interests. Part of this had to do with the intervention being conceived with the donor community in mind – an audience that was more interested in solving a problem framed in generic terms (tenure insecurity) than in coming to terms with the complexities of the situation at hand. The need to be effective led to policy simplifications discounting the local context and bracketing power relations. The result was a kind of structural blindness.

This blindness was shared even by development workers on the ground. Biddulph tells the story of Oxfam field workers tasked with protecting Cambodia’s forests. They had to confront the fact that ordinary villagers routinely plundered the forests for their livelihoods, collecting timber and non-timber forest products. When asked about it, they responded by saying that they would have to teach the villagers to value their environment. “If the ideas and behavior of the people did not fit the project document,” Biddulph (Reference Biddulph2010, 218) observed wryly, “then the ideas and behavior should be changed.” This brings to mind Weber’s (Reference Weber1949, 94) caution against forcing reality into the “procrustean bed” of theory.

The persistent gap between the aims and outcomes of development interventions led to disillusion for Biddulph, who fled to graduate school in order to contemplate his “disappointment” as a development worker. He begins his dissertation with a remarkable confession:

Coming to terms with the reality on the ground forced Biddulph to develop a more pragmatic approach to the problem of landlessness. In the dissertation, he recommends working directly with powerful figures, for instance by asking them to accommodate smallholders on the land they control and by crafting interventions that the political elite would feel comfortable implementing.

Some may argue that embedding runs the opposite risk of idealism, that of excusing corruption. We disagree. An embedded approach does not preclude recognizing the problems created by corrupt practices. It simply makes understanding these practices as structured social action – that is, historically and sociologically – the priority. To be sure, this understanding will inevitably complicate a social problems framework, not least by challenging reductive accounts. But this is a good thing insofar as it brings us closer to reality. The hope, of course, is that an embedded understanding will inform more realistic approaches to the problem as such.

Second, essentialism. We might think that articulating corruption in terms of self-interest would pull us away from a cultural understanding of the problem, and yet, ironically, a reductive and moralistic account serves to entrench a view of corruption as a cultural pathology. The relevant conceit is that if corruption is the same thing everywhere, then so is anti-corruption; in other words, what worked in, say, Singapore should work in the Philippines despite the two countries being completely different sociologically. This particular alchemy of commensuration informs Quah’s (Reference Quah2011) approach to corruption in Asia. “Why is corruption a serious problem in India, Indonesia, Japan, Mongolia, the Philippines, South Korea, Taiwan, and Thailand,” he asks, whereas “Singapore and Hong Kong SAR [have] been able to minimize corruption?” (32), holding up these two exceptions – which are, of course, exceptional for good reasons – as models for emulation. (It is a stunning formulation if you really think about it and almost reads as if it were meant facetiously.) In the case of the Philippines, he points to the usual suspects as promoting corruption: the low salaries of civil servants, red tape, the low risk of detection and punishment for corrupt actors, a lack of political will, and culture. With respect to the latter, Quah writes that corruption thrives in the Philippines because of the importance of family ties and a culture of reciprocal obligation, and approvingly cites a description of the Filipino family as a “corruption syndicate” (132). He ultimately concludes that curbing corruption in the Philippines remains “an impossible dream” so long as these factors persist.

Telling people that they need to change and then saying that they cannot because of who they are – their politics and culture – is not an idle gesture. It succeeds in making people feel bad about themselves. This negativity is easily pathologized, or attributed to cultural or national identity, as Smith (Reference Smith2007) illustrates. His informants tend to view corruption as a specifically Nigerian problem, and their sense that “something is wrong with us” fosters much handwringing and self-flagellation. As Chinua Achebe (Reference Achebe1983, 2) lamented in The Trouble with Nigeria, “Whenever two Nigerians meet, their conversation will sooner or later slide into a litany of our national deficiencies.” Similarly, in the Philippines, the nationalist politician Benigno Aquino (Reference Aquino1968), in an article titled “What’s Wrong with the Philippines?,” bemoaned his countrymates’ lack of discipline and propensity for corruption. This sensibility is debilitating, and of course it is. What if anything can be done about corruption if the problem is cultural? Hence, as Olivier de Sardan (Reference Olivier de Sardan1999, 48) observed, “the general feeling of helplessness in the face of an infernal mechanism.” The real problem is that a thin and morally charged account of corruption leads people to view corruption as both endemic and problematic without access to a deeper understanding of why the practices labeled corrupt are endemic and also, just as importantly, why the problem framework is so compelling. Consequently, they end up blaming themselves as the problem. They point to their “bad values” while judging themselves by an impossible standard. Understanding corruption in the context of long-term processes, as practices that have grown out of real structural conditions but are also changing in ways that escape anti-corruption metrics, provides people with a broader and more salutary perspective, one less likely to lead them to denounce their “cultures of corruption” and leave them feeling hopeless.

1.5 The Volume

All the chapters in the volume build on the older, more interesting, and humanist tradition of embedding corruption, although not all of them are historical. The volume is inherently comparative by format. Simply reading across the chapters, we are able to see, for instance, that corruption in Cambodia and China are different things, whereas rent-seeking in Cambodia resembles rent-seeking in India. Meanwhile, the focus on structure and process gives us insight into why. The volume is organized around the three generic processes we outlined in this chapter: emergence, institutionalization, and mobilization.

The chapters in Part I of the volume feature corruption as an emergent process. Marco Garrido (Chapter 2) takes three types of corruption commonly associated with the formal process of land registration in Cambodia – the violation of regulations and procedures (a corruption of the rules), patronage practices (a corruption of politics), and rent-seeking (a corruption of bureaucracy) – and embeds them in the processes and situations in which they took shape. The development of a land market and regime of private property rights in the 1990s and 2000s produced a situation of normative ambiguity; the process of state formation resulted in a neopatrimonial political order; and bureaucratization, or the development by tax and cadastral officials of procedures for handling land transactions, underscored a situation of organizational incapacity. Embedding changes how we understand corruption. We come to see it as an emergent practice (as opposed to being a universal one), as systemic (as opposed to being isolated to certain individuals or agencies), and as a way of building bureaucratic capacity (as opposed to being purely self-interested and anti-organizational). We get a better account of corruption as well as a better understanding of the situation on the ground.

Juan Wang (Chapter 3) tracks the changing meaning of corruption and anti-corruption in postwar China. She embeds these changes in the transformation of the country’s social and economic conditions and the evolving priorities of the state, particularly with regard to controlling grassroots cadres. Given the dire economic situation of the Maoist era, corruption was primarily understood as a drain on state resources, and anti-corruption involved punishing wasteful cadres. During a period of famine in the 1960s, corruption meant indifference toward the people, and anti-corruption agents targeted local officials for engaging in practices, for example the misappropriation of public funds, taken to reflect such indifference. After the Cultural Revolution, the priority was rebuilding the party, and corruption came to indicate a lack of party discipline. In the post-Deng era and particularly with the Xi regime, anti-corruption became a means of consolidating political power and checking the economic power of non-state, private sector actors. The chapter underscores the point that the meaning of corruption, and thus anti-corruption, is contingent and cannot be adequately understood outside of history.

Garrido and Wang trace the emergence or identification of corrupt practices in periods of upheaval and restructuring. The structures being assembled or consolidated are very different, however; on the one hand, an authoritarian state and command economy and, on the other, a nominally democratic state and market economy. These differences bear significantly on the nature of corruption in each case. In China, the state gets to dictate what corruption is by deciding where to focus its anti-corruption efforts, whereas in Cambodia, state agents are the primary vectors of corruption. They are the ones seen as getting in the way of “proper” bureaucratic and market processes.

The next two chapters offer another interesting comparison. They embed corruption in a colonial and thus transnational structure of power: the so-called rule of colonial difference (Chatterjee Reference Chatterjee1993) – in this case, a view of the colonized as constitutionally corrupt and untrustworthy. Jack Jin Gary Lee and Kwai Hang Ng (Chapter 4) show that British colonial authorities in Hong Kong conceived corruption as a particularly Chinese or native problem. They discuss an episode where this view was challenged, when senior British policeman Peter Godber fled Hong Kong while under investigation for receiving bribes, and they track the efforts of colonial authorities to address the incident. These efforts culminated in the creation of an anti-corruption agency headed by expatriate British officials and targeting mainly Chinese police officers. The Godber incident not only failed to unsettle the predominant understanding of corruption but served to entrench a culturalist view.

Anurag Sinha (Chapter 5) argues that corruption in British India came to be rearticulated over the course of the late eighteenth and early nineteenth centuries as the British public’s understanding of the colonial enterprise shifted from the effective management of colonial revenue to the good government of subject peoples. The shift crystallized in a power struggle between Richard Wellesley, the governor-general of India, and the East India Company’s Court of Directors. The Court represented Wellesley’s efforts to found a training college in India as a kind of corruption: as flouting established procedure, a case of institutional overreach, and an affront to British national identity. Corruption in the last instance had to do with the acculturation of colonial agents into Indian society. Here again we see an understanding of corruption emerge around questions of imperial rule and cultural difference. Once more, we see the rule of colonial difference being invoked. As Sinha observes, the process of defining corruption was at the same time a process of constructing universal moral standards, that is, of defining the relationship between colonizer and colonized in terms of norm and deviance.

The chapters in Part II of the volume speak primarily to the various ways corrupt practices are connected to and sustained by the larger social and political structures in which they are embedded. Sneha Annavarapu (Chapter 6) considers the institutionalization of the broker or fixer, who facilitates government transactions for a fee. The use of brokers to acquire driver’s licenses in India has become so commonplace, she writes, that some people do not even realize that brokers are not government workers and the service they provide is illegal. She embeds people’s reliance on brokers in an experience of the Indian state as problematic. Brokers thus represent a kind of solution. Given an experience of government services as uncertain, arbitrary, and inordinately complex, they provide a measure of control as well as simplify a potentially complicated transaction. They also allow middle-class clients in particular to stay “clean” by keeping corrupt transactions at an arm’s length. Resorting to brokers has the effect of entrenching the problem, however. It renders the driver’s license meaningless as a qualification and feeds into perceptions of the state as ineffective and dirty.

In Annavarapu’s account, corrupt practices become institutionalized through rationalization. People find brokers useful and justify their use of them (even as they protest corruption in other forms). For Leslie MacColman (Chapter 7), institutionalization has to do, rather, with the articulation of organizations into larger systems. MacColman emphasizes the connections between the organization of interest and “outside” actors and organizations. Her chapter examines the Metropolitan Police force created by the Buenos Aires government in 2010. Interestingly, mid-ranking officers in the Metro Police had a reputation for being corruption-free while high-ranking officers were seen as being notoriously corrupt. MacColman explains this pattern by tracing the relations the different groups of officers had with other actors and organizations. Mid-ranking officers transferred into the Metro Police precisely in order to escape the corruption-ridden national police force. Further, the virtual monopoly of the national police over illicit protection rackets effectively insulated them from pressures to participate in such rackets. High-ranking police officers, meanwhile, were tied up with municipal politicians and powerful economic actors, including organized crime. The mayor of Buenos Aires, for instance, enlisted the top brass in his effort to consolidate political power. So we might say that one group was structurally protected from corruption while the other was structurally exposed to it. MacColman’s chapter makes the point that organizations do not exist in a vacuum; they are articulated to other organizations and actors, which affect how they function and how corrupt or clean they can be.

Marina Zaloznaya and William M. Reisinger (Chapter 8) find bribe-givers in Russia to be more civically active than their non-bribe-giving counterparts. They explain this interesting result by showing bribery to involve social networks operating just under the radar of an authoritarian state. Bribe-givers have more extensive social networks, they argue, and these networks connect them to a wider range of social actors and organizations. Corruption networks provide a social infrastructure that can be utilized for civic purposes, including protest. In a context where freedoms of expression and association are strictly regulated, these networks represent a “third space” between state and society where people can collaborate and pursue common projects relatively free of state oversight. In this respect, we might understand public sector corruption as actually promoting the kind of civic culture envisioned by Almond and Verba (Reference Almond and Verba1963). Here the institutionalization of corruption is not only a matter of utility but structural opportunity. Limits on civic association make sub rosa corruption networks particularly valuable not just economically but politically.

Studying mobilization involves examining how claims of corruption and anti-corruption are used and situating their use with respect to particular interests, agendas, and institutions. The chapters in Part III of the volume feature new aspects of mobilization. Alex Diamond and Tomás Gold (Chapter 9) examine the corruption narratives of groups occupying very different social positions. Colombian farmers blame corruption for the deficiencies of the state and invoke corruption to justify their cultivation of clientelist ties. Free-market think tank elites across Latin America view themselves as engaged in a hegemonic struggle against leftist populism, which they identify as corrupt. These narratives represent each group’s interpretation of their historical situation based on their respective social locations. By juxtaposing them, we see that the two groups are essentially denouncing each other as the corrupt one. For the farmers, collaboration between political and economic elites (such as the people running think tanks) results in the plunder of state coffers. For elites, a populist mentality begets corruption in the form of patronage politics (in which the poor are enmeshed). The authors conclude that corruption has become the main idiom for discussing the problem of governance in Latin America. People at the top and bottom of the social hierarchy mobilize corruption in order to advance a political vision in line with their particular interests and worldview.

Nicholas Hoover Wilson (Chapter 10) understands corruption as a form of moral accusation involving the mobilization of certain audiences around particular moral frameworks. The nature of accusation depends crucially on which frameworks and audiences are mobilized and how – that is, on the process of mobilization. This process is inherently “promiscuous.” Mobilization efforts are unstable or slippery because they index disparate moral frameworks and invoke different forms of accusation. Corruption allegations, meanwhile, have a tendency of getting out of hand or expanding beyond their original scope and attaching to new targets. Wilson illustrates this promiscuity with reference to the “Negro Plot” of 1741 in New York City, which at first pertained to a simple robbery and then came to include within its scope a series of mysterious fires across the city and the prospect of a slave insurrection. In this process, allegations of corruption against two slaves expanded to involve “most Negroes” as well as poor, Catholic whites.

Fernando Miramontes Forattini (Chapter 11) locates the success of Operation Lava Jato – the most sweeping anti-corruption investigation in Brazil’s history – at the intersection of two distinct efforts to mobilize corruption. First, the United States pursued particular economic and geopolitical objectives (securing access to the Brazilian market and enlisting Brazil’s cooperation in the war on terror) in the name of fighting corruption. These efforts largely floundered until the creation of Project Pontes in 2008. The program brought members of the Brazilian judiciary and federal police together with US officials from the Department of Justice and FBI. Sergio Moro emerged out of this network a self-styled anti-corruption crusader. He launched the second mobilization effort, Lava Jato, driven by considerations particular to the domestic political scene. Forattini’s chapter highlights the contingency of mobilization. Lava Jato was neither the product of a US plot nor the result of Moro’s actions. The first reading is “noirish” and the second “heroic,” he writes; both accounts are deterministic. Lava Jato was the outcome of a perfect storm: the contingent alignment of the interests and objectives of actors embedded in different social and political fields.

The focus of Byron Villacis’ chapter (Chapter 12) is not on mobilization per se but on the agents of mobilizations, anti-corruption experts specifically. Who are they, sociologically speaking? Utilizing a Bourdieusian field approach, he identifies three clusters of power: tenured professors from American universities, economists with ties to international financial institutions, and practitioners with a high level of media exposure. He finds that quantitative skills matter more than bureaucratic experience, and that experts in the Global South with experience in the public sector are relatively less influential. Villacis locates the main proponents of the dominant view of corruption in social space – essentially embedding the disembedded view. By telling us who gets to frame the problem of corruption, he leads us to ask how the social location of these experts shapes the parameters of the problem and thus our approach to it. This reflexivity is the very point of embedding.

As the chapters to follow will make clear, embedding inevitably complicates the conventional understanding of corruption. The distinctions holding it together start to blur. Transgression is shown to be constructed and the public/private divide contingent. The corrupt bureaucrat shades into the “Weberian,” organizationally minded one. We arrive at a conception of corruption that is messier, contingent on the course of certain processes, and normatively slippery, making corruption hard to “fix” in both senses of the word: to pin down and to solve. This messiness is a resource analytically, however. It yields a more complex account, to be sure, but also a more clear-eyed one.