Impartiality and the Definition of Corruption

Impartiality as a Procedural Norm

There are some realms in which the norm of impartiality is not only unobjectionable but inescapable. A judge must be impartial when deciding a dispute between two parties. After all, the very office of an independent judge arises from the difficulty of judging in one’s own case. Impartiality is an a priori, necessary condition of the office. This is why there is universal disapproval of judicial bribery. This does not mean that it is universally true that judges may not receive gifts: on the contrary, in numerous historical periods gifts to judges were expected as signs of respect for their authority and often functioned as a kind of user fee, but judges were nonetheless expected to remain uninfluenced by those extraneous objects. To judge is to examine the objective merits of a case. To allow friendship, bias, or pecuniary considerations to sway one’s judgment is to cease to perform one’s role. It is not impossible to be able to judge impartially in spite of having a personal penchant towards one party – one can even imagine someone judging correctly in her own case. But this requires a degree of rational self-mastery few think reliable; thus, it is a fundamental principle of natural justice that no one should be judge in her own case. The office of the judge exists for the purpose of attaining impartiality. But ought this norm to be taken as the indispensable criterion of political uprightness more generally? Is the judge the model for all public office? And does impartiality suffice as a definition of corruption’s antonym?

There is a long tradition of attempting to take controversial matters out of the political realm appealing to law or to technocratic bureaucracy. Rothstein comes to his project with disappointment about democracy’s record as a source of improvements in welfare and legitimacy in the poorer parts of the world. He notes that the ‘quality of government’ – the degree to which government can provide fundamental goods like security, education and health – correlates more with human welfare and perceived legitimacy than does the existence of elections. While not anti-democratic, Rothstein thinks the reform of administration is of prior importance to the manner in which governments are chosen and the nature of the policies to be pursued. And when he looks at survey data from around the world, he notes that there is a higher degree of agreement on what good administration entails than is often noted by cultural relativists. So he argues that he can find a universally applicable set of norms on corruption that can be employed both for cross-national studies and for reform movements. Impartiality is attractive to him precisely because it is, he believes, ‘universal, procedural, uni-dimensional and normative’ (Rothstein, 2014: 745).

To ground this ideal, Rothstein appeals to Brian Barry’s Justice as Impartiality (Barry, 1995; Rothstein, 2011b: 16). Given the irreducible plurality of conceptions of the good, Barry – channelling Scanlon – argues that liberalism requires us to ‘reach agreement with others on terms that nobody could reasonably reject’ (Barry, 1995: 168). The principle that permits this is impartiality. Barry is not making the claim of radical impartialists that we ought to be entirely impartial in every aspect of our lives – on the contrary, in much of our lives, we rightly behave partially, favouring our friends and family and preferring certain notions of the good to others. ‘First-order’ impartiality – being impartial in all aspects of one’s ethical life – is, for Barry, patently absurd. Rather, he thinks we must adopt ‘second order’ impartiality that only concerns the basic rules for living together reasonably with other reasonable people who might differ in their views of the good (a position which allows him to sidestep a number of anti-impartialist arguments such as those presented in Young, 1990). Now, on first glimpse, we might think that Rothstein has committed a category error in placing such weight on Brian Barry’s thesis. After all, Barry is concerned with establishing the foundations of justice; Rothstein is much more focused on the effective delivery of government services. He makes explicit this difference with Barry: he argues that Barry, by concerning himself with fundamental matters of justice, was not as thoroughly procedural as he ought to have been:

A large part of the critique from within political philosophy that has been directed against Barry’s theory is, to our mind, due to his conflation of the input and output sides of the democratic system and his idea that impartiality should not only guide implementation but also the content of public policies (Rothstein and Teorell, 2008: 172).

Nonetheless, Barry’s proceduralist argument does lend itself well to the use Rothstein wishes to make of it, for Barry repeatedly speaks as if the division between first-order and second-order impartiality is the division between politics and administration (Barry, 1995: 143). For Barry, impartiality does not determine outcomes of decisions, but merely the decision-making criteria. Thus, just as Rothstein appeals to impartial administration to escape controversial political positions, Barry wishes to bracket politics because he wishes to find a realm of universal agreement under which the various ‘first-order’ visions of the good can coexist politically. Where Rothstein brackets politics entirely, Barry’s intention is rather to constrain politics with procedural strictures that make it impossible for groups with one vision of the good to dominate others. But both writers evince a thoroughgoing celebration of impartial administration.

Both Barry and Rothstein insist – against certain critics of impartiality – that impartiality need not mean a lack of particularistic attention. And they both, as we noted, think particularistic affection also has an important place in our lives. But the key to their doctrine is that one must separate one’s roles. Everything comes down to how we separate first-order from second-order impartiality. Or, to put it plainly, how we separate what we owe to loved ones from what we owe to all members of society. Certainly, as Barry notes, many societies in which corruption is rife exhibit little generalized public trust and much more particularistic trust – and the latter bleeds over into the former (see Putnam, 1993; Uslaner, 2008). But between a radical universalism that makes us say ‘mine and not mine’ about the same things and a situation in which most everything is a private concern subject to the logic of particularism, there is a wide range of constitutional possibilities. It is to this that we turn.

Corruption Discourse and the Art of Separation: Defining Who Should Be Impartial Where

One popular rule-of-thumb definition of corruption is ‘the abuse of public office for private gain’. Insofar as the discourse of political corruption depends on a distinction between public and non-public goods and duties, it is a discourse that requires a capacity to divide between the ethical dispositions appropriate to various realms. We accept gift giving and reciprocity relationships with friends, but not with judges – different offices or spheres call for different ethical dispositions. This is precisely what is entailed in Barry’s separation of first-order and second-order impartiality and in Rothstein’s distinction between the ‘input and the output’ sides of government. Ultimately, this entails treating politics as the realm of contest between thick visions of the good, and constitutions and administration as the realm of impartiality. Rothstein draws on Michael Walzer’s Spheres of Justice to argue that different ethical norms govern public office and private affection:

those who exercise public power need to know the borderlines between the different moral spheres. Second, they need to know which norms are appropriate (and inappropriate) in the ‘state’ sphere and how these norms differ from what is legitimate in the other spheres (Rothstein and Teorell, 2008: 174).

It is interesting to note that this theory dovetails nicely with Max Weber’s arguments about the distinct ethical demands of ideal-type bureaucracy and other ‘value spheres’ like politics, commerce, or science (Weber, 1972: ch. ix, ab.2). The ideal-type bureaucrat, Weber insisted, must not be a politician – different ethical characters ought to reign in different spheres, and the impartialist champions of quality of government often take the Weberian ideal-type bureaucracy as a moral ideal. The difficulty with Rothstein’s proposal is not that it insists upon ethical separations. Rather, it is that it treats certain separations as uncontroversial and universally valid. Not all impartialists make this error: Alina Mungiu-Pippidi, one of impartiality’s great champions in corruption research, avoids some of this false universalism by openly avowing the particular European origin of the ideal (Mungiu-Pippidi, 2015: 58), though she retains some false universalism in averring that the ideal is now no longer controversial (Mungiu-Pippidi, 2013: 1262). Her account of the development and importance of formal-universalist norms of governance is indispensible reading, but she neglects the degree to which there are varying ways in which impartiality and partiality might manifest themselves in modern states and that this is a matter subject to philosophico-political dispute. Mungiu-Pippidi prefers to describe impartiality in governmental services as ‘ethical universalism’ (which she contrasts to ‘particularism’), but what she is describing is really a form of ethical pluralism, where different mores obtain in private and public realms. The discourse of political corruption is always about policing lines between ethical realms, and the most important political questions concern how such lines should be traced and where partiality and particularism should obtain.

The clear separation between the value spheres of politics and administration was offered by Weber as a fact of modern rationalization that needed to be respected. But Weber did not give an ethico-political account of how and where lines should be drawn. The ‘rationalized’ manner in which Weber carved up the ethical universe, and particularly his distinction between the bureaucrat (who is depersonalized, rule-bound and unfit for leadership) and the political actor (who has charisma and the capacity for electoral and parliamentary struggle) is highly specific to the needs of the modern, centralized, bureaucratic mass democracy as Weber saw them. But the details of how these ethical realms ought to be carved up calls for philosophical investigation. One’s intuitions about the nature and correct separation between given spheres ought not to be treated as universally valid.

Walzer has rightly called upon us to theorize the ‘art of separation’ – this is an important insight. But Walzer’s method is controversial: his distinctions between ethical spheres are drawn from local conventions, a position Rothstein incorrectly if excusably takes to be relativistic (Rothstein, 2011b: 19). Without canvassing the debates about Walzer’s method, let us merely note that there is a serious difficulty engendered by Rothstein’s desire to treat ethical separations akin to those he finds in Walzer as straightforwardly universal. Rothstein chides Walzer for being arbitrary in his definition of spheres, but he proceeds to present a model of division between self-regarding and other-regarding activity that is much more arbitrary and question-begging:

In the ‘state’ sphere, the norm is that the exercise of power should be in accordance with the enacted laws and policies and that they should apply equally to ‘all’, as stated in the two principles of political equality and equality before the law. In the ‘market’ sphere, the accepted norm is that behavior according to self-interest is justifiable, but the scope dimension ‘all’ implies that everyone should have equal access to the market[.]… the accepted norm in the private sphere is that we should not behave according to self-interest against our family (or clan) members or friends … (Rothstein, 2011b: 20).

These norms are nothing more – and nothing less – than commonsense assumptions prevalent in liberal-democratic, capitalist states. In many respects, they are compelling (though the devil is always in the details), but there is something questionable about simply declaring them universal on the basis of their plausibility. Walzer’s observation that there are different ethical requirements for different spheres was so fruitful because it opened up philosophical discussion on the nature and justification of these divisions. To take Walzer’s observation as grounds for asserting an a priori set of distinctions is to betray the intellectual project itself.

If the discourse of corruption entails policing ethical separations between realms of public and private, we cannot treat these separations as given. The manner in which the spheres can be and have been divided up differs radically in different historical and cultural contexts. Still, it is important not to dismiss Rothstein’s universalistic penchant. There is some degree of universality concerning distinctions between public and non-public goods. Nonetheless, to say this is to say too little. Not only is there a great diversity of ways in which these distinctions have been drawn historically, but the manner in which such divisions are drawn is the heart of the political itself. Plato made the distinctions between the various roles in the city the core of justice; in the same manner, we ought to see as a central philosophical concern the manner in which we separate private from public and the ethical mores of office from extra-political pursuits. To declare without justification that it is permissible in the field of the market to act on self-interested, profit-seeking bases (where partiality towards those who pay is expected), while in government office such motivations are impermissible is to have decided in advance a number of important questions concerning the nature of public office, its delimitations and its ethical exigencies. In Ancien Régime France, office was an investment; it was venal, something that Montesquieu thought thoroughly consonant with a monarchical constitution (Montesquieu, 1989: V:19). In contemporary, so-called ‘neo-liberal’ states, universal public services are often farmed out to private, profit-seeking institutions (a practice akin to Ancien Régime venality without the trappings of honour). When Barry treats the lines between first-order and second-order impartiality as self-evident and when Rothstein acts as if the lines between the mores of the market and the public service are universal and uncontroversial, they prejudge many politically significant constitutional matters. When Socrates in Plato’s Republic proposes that the guardians should say ‘mine’ and ‘not mine’ about the same things (462c), he eliminates any possibility of their being corrupted by activities extraneous to their duty. When Aristotle argued against this excessive unity and proposed a different way for the individual to relate to the city and the household he was offering a contrasting manner of thinking about the ethics of various spheres. The question is not just about the extent to which we should employ the first-person possessive pronoun – it is about which matters belong in the realm of impartiality and which do not. The manner in which we draw this line is one of the classic problems of political philosophy.

Now, one clear example of corruption that Rothstein thinks we can understand as a breach of impartiality is the practice of petty corruption whereby providers of government services accept money to help some people jump the queue. Few would dispute that an official who administered services preferentially in return for a gift was akin to the judge mentioned above. Impartiality appears a clear criterion. But here we have a classic example in which corruption talk entails making decisions about the proper separation of public and private. We can imagine, for instance, a system in which publicly financed health care is taken to be something available equally, without favouritism, to all citizens. This is, for instance, currently the case in most Canadian provinces, and if a doctor gave preferential access to someone for a private payment it would be considered corrupt. But one province, Quebec, is experimenting with a two-tier system in which, in addition to the public health plan (where everyone is treated equally), there is a parallel private system in which people can pay to receive faster access to some services. Since this was a political decision arrived at by a legitimate government, and since the system is transparent, it would strike most people as moralistic to charge the private-practice doctors or patients with corruption. We might argue that the policy corrupts the public health-care system itself, but that would be a deployment of the term that proponents of private health-care delivery would legitimately consider polemical. The merits of the policy are debatable, but this is clearly an instance of private wealth being employed to purchase differential treatment for what is seen as a government service. Does impartiality as an ideal serve to condemn this practice? Surely Rothstein would say no – the manner in which a given service is provided is something he believes subject to political deliberation and decision (Rothstein, 2011b: 11). Similarly, Barry does not think that justice as impartiality makes any particular demands about the manner in which public services are delivered (Barry, 1995: 97). But surely at a certain point shifting public goods to the private realm will start to look corrupt to anyone with a robust conception of the public good. What would be the difference, say, between two-tiered health-care delivery and a policy that allowed wealthy people to purchase faster access to other government services such as services providing identification papers, or building permits, or even legal proceedings? In each case, we can imagine the society becoming increasingly oligarchic, which egalitarians would likely term corrupting, but since services in such a system would be offered in an ‘impartial’ manner (to the highest bidder), they would not be termed corrupt under the impartialist definition of the term.

Rothstein himself would deplore this outcome – he has argued forcefully for universal social-welfare policies as a means of breaking out of ‘social traps’ (Rothstein, 2005). But he deliberately separates his preferred policy outcomes from his definition of corruption in order to keep the term universal. Rothstein maintains that procedural universality precedes substantive universality – when public services are delivered impartially people are more likely to support generous welfare-state measures (Rothstein, 2011b: 127–130). The difficulty is that insisting on the purely proceduralist ideal of impartiality falls into the old relativist trap of not being able to denounce a feverish city. It is also problematic on its own terms, for a constitution that allows certain essential goods to be distributed on the basis of wealth is a constitution that does not treat people impartially. Oskar Kurer – who, unlike Rothstein, does not limit himself to the ‘output’ or administrative side of the equation – suggests that the norm of impartiality can serve to denounce this type of outcome even when laws are framed to make such unequal treatment perfectly legal. But he admits that the ideal of impartiality ultimately depends upon existing public norms in a given context, and is hence liable to some significant relativity (Kurer, 2005: 233). It is unlikely that Rothstein would want to allow as uncorrupt such radical privatization that made justice itself a purchasable good, but the question of which elements of life are to be included in the sphere of universal goods allocated impartially is hard to determine without some heavier normative lifting than that provided by such content-less ideal.

Those who think health care a fundamental public good might term a society in which basic health-care delivery is a private, profit-making concern radically corrupt; those wary of large public sectors would disagree. The difficulty is that if one wishes to employ the term ‘corruption’ in a normatively robust manner one needs to make important arguments about the appropriate purview of the ‘public’. Barry, attempting to determine how far impartiality should extend permits it to extend somewhat into the realm of so-called private affairs, leaving room for just enough first-order partiality as to permit ethical life but limiting it in order to prevent, say, systemic partialities that disadvantage particular groups (Barry, 1995: 203–207). But the rub is that this ‘Goldilocks’ theory (as Barry calls it) – having ‘neither too much first-order impartiality nor too little first-order impartiality’ (Barry, 1995: 206) – is the core of the political itself. For the political is in large part about establishing the exclusions between public and non-public and the ethical attitudes appropriate to each. It is not enough to argue that whatever services are established as public services should be administered impartially, for it is the nature of oligarchy to transform rule into the advantage of the wealthy. The basic question of what is owed to all and what to particulars and what criteria might be employed to distinguish between individuals is the core of the political itself. One might be tempted to say that the degree to which medical services are publically provided has nothing to do with corruption, but extending this logic to other services associated with the state – security, say, or justice – would entail highly exploitative arrangements where some are personally dependent on others. A radically minimalist state in which most goods are subject to the market and the state’s role is merely the protection of property looks very much like the Aristotelian definition of the deviant regime, in which rule is in the interests of the rulers at the expense of the ruled. Rothstein would deplore such a condition, and he argues against it vigorously (Rothstein, 2011b: 225), but there is no basis on his accounts to term it ‘corrupt’.

To place something in the realm of the market is to place it in the realm of partiality. We block exchanges of justice for this reason (though the private nature of law practice renders justice somewhat partial in most liberal-democratic countries). Where exchanges should be blocked is a central question that should concern students of corruption, but it is a question that the ideal of impartiality by itself cannot address. ¹

Corruption Discourse and Membership: Who Merits Impartiality?

If the distinction between ethical realms is fundamentally political, so too is the decision about who is to be given standing in the impartial realm. Neither Barry nor Rothstein nor Kurer conceives of impartiality as a norm that applies to all human beings. Rather, they assume the limits of the nation state. Of course, anyone who has ever crossed a border has experienced partiality in its strongest form (and might have experienced corruption as well). Kurer makes clear (though perhaps inadvertently) the degree to which the definition of impartiality depends upon prior decisions about membership and merit. He defines impartiality in the following manner: ‘those who are equally deserving ought to be treated equally’ (Kurer: 230), a principle cited approvingly by Rothstein (2011a: 230; 2011b: 15). Neither author appears to have noticed that this is precisely the definition of justice proposed by Aristotle (Politics 1280a). But of course to note this would be to note how insufficient the definition is, for Aristotle proceeds to the more important and fundamentally political questions of who is worthy of equal treatment and what types of inequality have normative importance. Since such fundamental ethical philosophy is sometimes unwelcome in practically minded social sciences, a great deal of reflection on corruption leaves it aside. But it is at the heart of the matter (a similar point, without the appeal to Aristotle, is made by Kolstad, 2012: 243). If the content of particular policies is left out of one’s definition of corruption, many systematic abuses of political office for personal gain will slip through the cracks.

Consider citizenship status itself. Certainly, the manner in which citizenship law is employed to support domination and, indeed, to encourage the law’s own subversion is not without importance to the subject of corruption. Much of the American agricultural sector is dependent upon exploiting an underclass of undocumented migrants. While the laws are administered ‘impartially’, granting equality to equals, the system is premised on a wider partiality that permits systemic abuse of the public for private gain. Even when such systems are institutionalized without entailing law breaking on the part of employers – in Gastarbeiter-style temporary migrant programmes – there remains a highly dubious form of systemic partiality that one might wish to term corrupt. Citizenship itself is partiality, and citizenship regimes can serve to create exploitation and corruption. ²

One might be tempted to reply that the philosophical problems raised by corruption are difficult enough without introducing the thorny problem of the ethics of immigration control and citizenship: perhaps we ought merely to take the existing nation state as given in our ethical reflection. The difficulty is that the question of membership cannot easily be set aside – numerous unjust and corrupting exclusions have been enabled by the drawing of distinctions between in and out. One need only consider the so-called ‘Bantustans’ that were employed to enforce Apartheid to see this point.

If citizenship itself is no less than official partiality, what about policies whose very content is to discriminate between citizens? Certainly, there are types of ‘positive discrimination’ with laudable aims. Rothstein argues that it is perfectly acceptable to have so-called ‘affirmative action’ policies where a disadvantaged group is given preferential access to positions from which its members have been historically excluded (Rothstein, 2011b: 22). It is merely important, he insists, for such policies themselves to be enacted impartially. Now, while such policies may be entirely justifiable, it is difficult to accept that they are truly ‘impartial’, or to see in what manner Rothstein believes they are to be distinguished from partialist policies with more nefarious goals. Were one to say, ‘I only give government posts to people in my extended family, but I am completely impartial in my selection of them’, one would not be treated as properly impartial. Naturally, there are numerous political grounds to distinguish between policies that discriminate positively and negatively (and those who profess not to see a difference do not merit consideration), but if we are to claim that partiality is permissible on the policy-side of things, then once again we start to see a term with little normative weight. Perhaps, as Ian Shapiro (2015) has recently argued, when impartiality is praised, the real normative work is done by the opposition to domination. After all, affirmative action policies exist because systemic inequalities render the exercise of impartiality itself a means of supporting systemic domination (consider, for instance, the strict, impartialist examination system of the French Grandes Écoles, which functions to recreate a fairly homogenous ruling elite).

Rothstein is aware of the difficulty raised by impartiality’s empty formalism: ‘self-evidently, normatively despicable policies, such as an apartheid system, can in theory be implemented impartially. However QoG [quality of government] as impartiality shares this problem with representative democracy, since there is no guarantee that democracy will equal justice’ (Rothstein, 2011b: 25). But this deflection does not answer the difficulty. For apart from its greater formal systematization, the apartheid system is not fundamentally distinguishable from other systems of clientelistic politics in which certain regions or ethnic groups are systematically favoured over others. Unless we adopt a radically relativistic definition of corruption in which only that is considered corrupt which offends against existing laws, we cannot accept the impartial exercise of existing laws as the hallmark of governmental quality.

Defining corruption solely in terms of existing laws (as did Nye in his influential 1967 essay) has long been criticized. In recent times, Lawrence Lessig has sought to define systemic corruption in terms of perfectly legal but highly distorting systems of oligarchic dependencies undermining democracy. In a similar spirit, Daniel Kaufmann (Kaufmann and Vicente 2011) has popularized the term ‘legal corruption’ to describe systemic and legal abuse of the public for private gain. This is extremely important as it catches the most important ways in which goods are diverted away from public use to private interests. This does not solve the difficulties, as Kaufmann’s term itself ultimately rests upon controversial claims about the public good, but it does offer an important corrective, capturing a great deal of rent-seeking and state capture that would otherwise be missed. Rothstein himself applauds Kaufmann (Rothstein, 2011b: 208), but he does not appear to have taken Kaufmann’s argument to heart, for his insistence on framing anti-corruption as impartiality in the ‘output’ side of governing reinforces the widespread tendency to describe corruption only in terms of existing laws. Rothstein’s critics have often noted this point – Marcus Agnafors (2013), for instance, laments the normative emptiness of Rothstein’s procedural ideal and argues that impartiality ought not to be fetishized. It is one norm among many and should periodically cede to other values. Agnafors offers a useful correction, giving a complex account of the quality of government that offends less against widespread moral intuitions than does Rothstein’s unidimensional impartialism. But the difficulty goes beyond the fact that procedural norms have insufficient ethical content – it is that to define corruption adequately will entail answering a fundamental question concerning the principles of justice upon which we distribute offices, honours, goods and membership in a political order.

From False Neutrality to Political Overreach

Another difficulty with treating impartiality as the opposite of corruption is that it (like many proceduralist ideals) slips substantive goods in through the back door. Barry devotes much space in Justice as Impartiality to defending proceduralism against charges of hidden first-order normativity by the likes of Alasdair MacIntyre, but since he never addresses MacIntyre’s philosophical arguments about the nature of reason, it is hard to know what to make of his unreflective appeal to the Scanlonian criterion of reasonableness (West, 2000: 32). And, indeed, the moment one attempts to get impartiality to do any serious ethical work, one finds particular visions of the good slipping in. Take, for instance, a judgment Barry expresses about the ability of a private club to exclude members. If the club has an important role in the business life of a city, Barry thinks there are good grounds to prevent exclusions on the basis of sex, race or creed, but he asserts that second-order impartiality would not demand that clubs be prevented from excluding someone because they just don’t like him (Barry, 1995: 16). As difficult as such a distinction would be to police in practice, Barry’s intuition is correct: there is a categorical difference between the one type of partiality and the other. The first type (systemic racism) is more socially destructive than the second (personal antipathy). Barry, however, tries to make sense of this intuition by suggesting that the personal rejection is actually impartial because everyone has to face the possibility of not being liked by the club members (Barry, 1995: 16). This is as unconvincing as Rothstein’s claim that positive discrimination does not constitute partiality. Perhaps the principle at work here is not that arbitrary personal dislike is actually impartial (it is, after all, the very definition of partiality!), but rather that there is a wider political good in declaring racial, gender and religious discrimination intolerable.

Many other judgments Barry makes evince this type of slippage, as first-order goods find themselves doing work they are not meant to do in his theory (Shapiro, 2015). Still, for the most part, Barry and the other impartialists under consideration here try to keep their first-order moral considerations out of their second-order political arguments. But, this can have drastic decisionist consequences. Consider a recent article by Emanuela Ceva and Maria Paola Ferretti, who adopt Kurer’s definition of corruption as the breach in the impartial exercise of public office (Ceva and Ferretti, 2014). The case they examine concerns the provision of abortion in Italy, a country with publically funded health care. Italy made abortions available under the national health-care plan. However, there is a provision allowing doctors to refuse to perform the service for reasons of conscience. In more religious parts of the country, doctors refuse in large numbers (partially, in Ferretti and Ceva’s opinion, due to informal incentives); thus, though the policy permitting abortion exists, these doctors effectively exercise undue influence and render the service near unattainable in many regions. Ceva and Ferretti argue that this is a form of institutional corruption, a systemic breach of impartiality in the delivery of a service.

There is something highly questionable in their tendency to treat the exercise of individual conscience as partial (because it favours a particular conception of the good ³ ) and a kind of illegitimate benefit (because, beyond career advancement, Ceva and Ferretti think the anti-abortion doctors are benefiting by advancing their conception of the good in an illegitimate manner). Indeed, treating conscientious claims about the good as partisan ‘interests’ (Ceva and Ferreti: 139) might seem like an abuse of language, but it is necessary for them to frame this political position as an interest if they wish to retain proximity to the widespread view of corruption as the abuse of a public office for a private benefit. But calling conscientious objection ‘corrupt’ would likely strike most readers as counterintuitive. At the very least, the authors would need to demonstrate why conscientious objection to policies of which the authors might disapprove (such as the death penalty, or the bombing of civilians during a war) are not also corrupt acts. Certainly, the situation they describe in which an important service is effectively denied to women is troubling, and it is also perfectly possible for the antinomian criterion of individual conscience to be abused, but the way in which the authors frame the issue here overreaches a great deal.

Barry argues that justice as impartiality ought not to determine contentious questions like the legitimacy of abortion (Barry, 1995: 91–93), and Ceva and Ferretti appear to agree – they place weight on the fact of a democratically arrived at law permitting the practice. Impartiality does not decide the rule, but it renders the political ultimately a realm of pure will and the administrative the realm of pure obedience. But if impartiality means the degree of inflexibility that Ceva and Ferretti think it does, it is a principle inconsistent with liberty of conscience. Indeed, to be consistent, Ceva and Ferretti, working for a public educational institution, would have to follow Max Weber’s advice and keep political concerns out of the classroom and official scientific pursuits (Weber, 2009: 129–158). Impartiality here, rather than evading the political, renders corruption discourse hyperpolitical, for it makes political decisions – whatever they may be – uncompromisingly binding at the level of the individual conscience of service providers.

Corrupt and Corrupting Impartiality

So far, we have seen that the definition of corruption as partiality has many failings. In attempting to avoid politics, this ideal is insufficiently attentive to the most prevalent abuses of public power for private gain. Insofar as its proceduralism does not smuggle in thick conceptions of the good by importing into the concept opposition to specific types of domination, impartiality is an ideal that is perfectly amenable to the type of inequalities that are, on Rothstein’s account, the leading sources of corruption. Most importantly, the distinction between first-order and second-order impartiality and the related distinction between politics and administration tend to reinforce a series of separations between ethical realms that are themselves a primary subject of political philosophy and political contest. To take partiality as an uncontroversial definition of corruption is to foreclose the most important politico-philosophical discussions about the appropriate division between ethical spheres. This final section will indicate briefly that there are philosophical traditions that see impartiality not merely as insufficient, but as itself possibly corrupting.

Naturally, there are numerous variants to this theme. We noted that Aristotle’s opposition to excessive unity in Plato’s Kallipolis is based on the importance of ethical particularity for the good life: where we draw the line between acceptable and unacceptable particularism is a central theme of political philosophy. There are numerous philosophers – from agonists to civic republicans – who are even more sympathetic to fervent partiality towards one’s own and propose that its diminution itself is the source of corruption. Consider, for instance, the political theory of Ibn Khaldun (1332–1406), the medieval philosopher and historian whose Muqaddimah offered an account of history as a cycle of purification and corruption. Ibn Khaldun saw civilization as a perpetual cycle between the rough nomadic Bedouin and the settled city dwellers with their walls and luxuries. In brief, history is forever renewed by the small, cohesive and highly martial nomadic peoples who invade corrupt, sedentary cities, reinvigorating them even as they bring destruction. In their turn, they become sedentary and enervated. They are brought back to their origins by fresh waves of nomadic peoples, who are forever a source of destruction and renewal. Now, the central principle that determines political history is the rise and fall of a kind of collective sentiment of belonging – assabiyah – that gives force to tightly knit social groups. Ibn Khaldun thought that the sense of tribal attachment, ‘group feeling’ as the term is translated (Ibn Khaldun, 1989), was the source of strength that allowed nomadic peoples to survive and conquer in harsh conditions. But once groups settle in cities, they become increasingly corrupted and lose their sense of group feeling. Where previously they did not have kings (for they relied on mores), they start to establish such authorities who rule through clients; where previously they did not need many laws because they were reverently attached to their tribe and highly pious, they now have need of external laws. Their larger population makes them less filial and more individualistic. Their greater luxuries make them less capable of war and more dependent on others. Within four generations, rulers become radically corrupt, overtaxing, ruling through clients, and emotionally separated from the ruled. The regime is generally overthrown either by its clients or by a fresh wave of nomadic conquerors.

Ibn Khaldun was not without ambivalence – he appreciated certain goods of the city, but he nonetheless depicted the city as a font of corruption. The source of nomadic peoples’ fierce purity was their assabiyah. While it promotes the public good and thus might appear close to impartiality, it is in fact the very opposite of impartiality, for it is based on fierce affection for one’s own. Now, this quality is troublesome from a universalist, Islamic perspective – there are Hadith specifically warning against assabiyah. And, indeed, Ibn Khaldun, as a pious religious scholar as well as historian, thought there was something particularly important about the conjunction of assabiyah with the universal mission of a prophet. But the cool impartiality that defines relations in large, law-bound and monarchical cities was, on his view, the soul of corruption and the attitude that led towards tyranny.

This is not to say that Ibn Khaldun had no place for impartiality. It was a necessary part of the judge’s activity. He advised judges, ‘Consider all the people equal before you in your court and in your attention, so that the noble will not expect you to be partial and the humble will not despair of justice from you’ (Ibn Khaldun, 1989: 173). But though this was central to the judge’s role, such inattention to social distinctions and such absence of partial zeal could not be rendered into generalized political virtues without entailing the death of collective virtue. Ibn Khaldun was admiring in his description of the first Caliphs who ‘always entrusted the office only to people who shared in their group feeling either through (common) descent or their status as clients. They did not entrust it to men who were not close to them in this sense’ (Ibn Khaldun, 1989: 174). It would take us out of our way to examine the details of Ibn Khaldun’s famous argument about assabiyah, but it is sufficient for our purposes merely to highlight this as one of a number of important normative positions that conceive of impartiality as part of a corrupted condition. Few thinkers – not even Ibn Khaldun – deprecate impartiality as such or in all areas, but there are radically diverse manners of circumscribing it and diverse ways of celebrating partiality itself as a basis for civic virtue.

Conclusion

This article offered several objections to the attempt to give as a parsimonious and uncontroversial definition of corruption the breach of impartiality. This definition cannot bear the load its champions want it to. On the one hand, it says too little: its proceduralist flight from substantive goods makes it insufficient to speak to a number of powerful intuitions about what a corrupt constitution looks like. On the other hand, it says too much, taking as given a number of contestable divisions between the ethical realms of politics and administration, between inside and outside, and between first-order and second-order impartiality. We concluded with an all-too-brief encounter with a philosophical position that sees impartiality not as the essential characteristic of integrity, but rather as an element of impersonality that corresponds to a corrupted regime. It is important to repeat that the ideal of impartiality is not to be rejected. The point is that where and when impartiality is a good and how to draw the lines between the realms of experience are much larger questions than is realized by impartiality’s champions.

Ultimately, the concept of corruption is not susceptible to an uncontroversial definition, for what is termed corrupt is a fundamental question of political philosophy. Defining corruption entails answering extremely difficult and perennial questions about the nature of office and the appropriate divisions between ethical spheres. This is not to say that the search for a rationally compelling definition is futile – quixotic as they seem, political philosophy’s universalist ambitions are worth retaining even in the face of inescapable pluralism. But what might a rationally compelling definition of corruption entail? To define corruption will require making very large claims about the nature of the good regime; it will entail asking many complicated questions that are at the heart of political philosophy itself. A fitting place to begin would be with the Aristotelian view that a corrupted constitution is one in which rule is in the interests of the rulers at the expense of the ruled. Whether following this polestar will lead us to calling for the world to ‘become Denmark’ (as Mungiu-Pippidi, 2013, puts it) is an open question. But one should not pretend that any proposed definition can be beyond the contestability that is the heart of political terms. If we wish to offer a definition as a contender for universal adoption, we must realize that it will rest upon larger claims about the right manner of dividing private from public, about the nature of terms like ‘integrity’, and, ultimately, about the correct definition of justice itself. And, indeed, any definition of corruption will not merely have important presuppositions – if proposed as a candidate for universal adoption, it will be a fighting creed. Rothstein has sought a definition that is sufficiently clean, simple and uncontroversial to be used as an international measure in order to test propositions about the effects of corruption on well-being. Important as his project is, it is worth remembering that if this definition were to be accepted by the wider anti-corruption movement, it would have the effect of slipping in some contestable norms through the back door. Anytime fundamentally political concepts are treated in an anti-political manner, alarm bells should ring.