Upholding public institutions in the midst of conflicts: the threat of political corruption

ABSTRACT Scholars and international organizations engaged in institutional reconstruction converge in recognizing political corruption as a cause or a consequence of conflicts. Anticorruption is thus generally considered a centrepiece of institutional reconstruction programmes. A common approach to anticorruption within this context aims primarily to counter the negative political, social, and economic effects of political corruption, or implement legal anticorruption standards and punitive measures. We offer a normative critical discussion of this approach, particularly when it is initiated and sustained by external entities. We recast the focus from an outward to an inward perspective on institutional action and failure centred on the institutional interactions between officeholders. In so doing, we offer the normative tools to reconceptualize anticorruption in terms of an institutional ethics of ‘office accountability’ that draws on an institution’s internal resources of self-correction as per the officeholders’ interrelated work.


Introduction 1
Since the mid-nineties, scholars and international organizations have recognized the importance of the corruption of public officials and institutions (in short, 'political corruption') as a key security issue and, as such, a threat to institutional action. In particular, they have emphasized the causal link between political corruption and a host of conflict-fuelling activities, such as arms trafficking and the support of terrorism. This awareness has also led to the identification of political corruption as one of the key factors in war to peace transitions. Political corruption may indeed undermine people's confidence in new public institutions, hamper economic development, and make scarce public resources go to waste. Institutional reconstruction programmes, moreover, may offer the terrain for even more political corruption to develop as concerns the distribution and employment of external donor aid (Zaum and Cheng 2012). As a consequence, the need for mainstreaming the fight against political corruption is well recognized in the context of postconflict institutional reconstruction. 2 We believe that, in the context of institutional reconstruction, the relation between political corruption and the upsurge of conflict is not limited to a causal link, but runs even deeper. Indeed, as we expound in the article, this relation concerns the very structure of institutional action and, therefore, the link between political corruption and conflict is also constitutive. This kind of link deserves closer study from a normative point of view. Such a scrutiny is important to enhance the understanding of the nature of the relationship between conflict and political corruption, and of the normative implications of this relationship for upholding public institutions in the midst of conflicts and understanding anticorruption programmes within that framework.
When public institutions are newly established after a conflict, public rules, values, and goals are reconsidered and reset. Moreover, institutional roles for implementing those rules, realizing those values, and pursuing those goals are redesigned and distributed among officeholders. A crucial concern is how to conceive and operationalize those roles and the normative powers -the rights and duties of office -associated with those roles. Such an operation is complex because it should both be capable of enacting the new institution's normative commitments and giving practical guidance for the officeholders' conduct. To address this challenge is critical for making institutional action work, and keeping it as safe as possible from such failures as political corruption consisting in the officeholders' misuse of their normative powers of office.
In this context, it can be difficult and indeed counterproductive to follow the somewhat commonly employed strategy of describing officeholders' rights and duties in terms of set standards governed by coercive rules. In this article, we show how this strategy is particularly troublesome when it is characterized in outward terms. From an outward perspective, such set standards for institutional action come from an external authority, such as an international organization, which is also often in charge of invigilating their implementation. We propose a different approach that valorizes an inward perspective on institutional action. Key to this approach is the mobilization of an institution's internal resources by engaging the officeholders to take on direct responsibility in the construction, exercise, and review of their roles and powers. This approach is centred on an ethics of 'office accountability' aimed to provide officeholders with the normative yardstick for their actions and interactions as they uphold the working of their institutions. Such a commitment on the part of officeholders is generally important for institutional action; we show how it is indeed crucial in the context of newly established public institutions.
By developing our normative framework for understating institutional action and failure, we therefore pursue a twofold normative aim. First, we aim to clarify the 2 For example, the United Nations Office for Drugs and Crime (UNODC) in the Doha Declaration identifies corruptionresistant public institutions as crucial to good governance (https://www.unodc.org/documents/congress/Declaration/ V1504151_English). This preoccupation has led, among other initiatives, to the United Nations Security Council's 'First-Ever Meeting on Corruption' in September 2018, as well as the recognition of the battle against corruption as one of the crucial components for achieving the goals of the 2030 Agenda for Sustainable Development(https://www. un.org/press/en/2018/ga12017.doc.htm).
constitutive (not just causal) link between political corruption and conflict; second, we aim to develop a new normative justification for an inward approach to anticorruption apt to capture the specificities of the challenges that political corruption poses to newly established public institutions in the aftermath of a conflict. 3 To pursue this twofold aim, we start from an understanding of political corruption as a dysfunction of institutional interactions. This understanding is coherent with most current views of what corrupt behaviour is, but it is distinctively capable of explaining why political corruption is constitutively (rather than causally only) a threat to the public institutional system. 4 As we show in the article, this explanation is particularly promising from the inward perspective of anticorruption we want to develop. Indeed, it has the merit of offering an internal insight about the nature of political corruption irreducible to the external consequences of this phenomenon and the attempts to address them from the outside of an institution (for example, through the constraining action of international organizations). In the central sections of the article, we make our point through the illustrative aid of some recent international programmes that address political corruption as a consequence or a cause of conflicts. In the last two sections, we draw on our normative discussion of these programmes to expound the constitutive relation between conflicts and political corruption. The implications of focusing on this constitutive relation will enable us to model the inward approach to anticorruption understood as a component of an institutional ethics of office accountability. In this framing, we show the limits of current outward legalistic approaches to anticorruption centred on positive (through the application of good standards and incentives) or negative (punitive) interventions notably upheld with the aid of international organizations. Our argument for an inward approach to anticorruption offers the normative tools for justifying anticorruption programmes aimed at mobilizing an institution's internal resources of self-correction by sustaining the officeholders' interrelated action and their capacity of self-scrutiny. In this vein, our goal is not to offer a recipe for engineering more or less ideally stable public institutions. Rather, our goal is to provide a framework for the normative justification of those interventions aimed at engaging officeholders in realizing an institutional environment that is supportive of institutional action in nonideal circumstances.

Upholding public institutions and the threat of political corruption
Political corruption is the corruption of public officials and institutions. In what follows, we build on our recent philosophical discussions of political corruption (Ceva 2019;Ceva and Ferretti 2021;Ferretti 2019) to develop the idea that political corruption is best explained as a deficit of 'office accountability' within a public institution. To understand the nature of political corruption, from this perspective, requires clarifying the nature and structure of institutional action.
3 Given the focus on political corruption and the reconstruction of public institutions in the aftermath of conflicts, we shall not discuss the qualifications that may extend our considerations to such private organizations as corporations or private associations. While we recognize the role of these entities in reconstruction programmes, our discussion focuses on the social and political institutions of the state. Any extension of our study requires a separate work, which accounts for both the analogies and the disanalogies across the different domains. 4 The institutional dimension of corruption has been widely discussed from a teleological perspective by Thompson (2018), Lessig (2018), Miller (2017), and, with reference to the context of democracies, Warren (2004).

Institutional action and the structural interrelatedness of institutional roles
One strain of the recent debate in institutional theory revolves around an opposition between reductivist and nonreductivist approaches to the nature and structure of institutional action. An example of the former approach derives from John Searle's constructivist view according to which institutions only exist insofar as they are collectively believed to exist (Searle 2010). This view entails that collectively accepted institutional rules create social institutions. Institutional action may not therefore be reduced to the individual attitudes, beliefs, and actions of the institutional members. Moreover, constitutive rules establish institutional roles, as well as the normative powers (the rights and the duties) attached to those roles; institutional roles are thus conceived as institutional facts (Searle 2010; see also Tuomela 2002). By contrast, reductivist approaches define institutional action as the coordinated action of a number of agents who occupy a role in the institution. In the teleological version defended by Miller (2010), for example, institutional action is directed to the realization of some collective ends (the institutional purpose), and the source of the duties of institutional role occupants lies in the collective goods for which the institution is designed to provide. 5 In this article, we follow an approach to institutional action that bypasses this opposition (Ceva and Ferretti 2021). Like the reductivist view, this approach starts from institutional roles (rather than institutional rules), and conceives of institutional action in continuity with the action of the individual officeholders in the exercise of their institutional mandate. Unlike that view, however, our approach to institutional action rejects the idea that institutional purposes and roles are the exclusive source of the normative powers (the rights and the duties) that accrue to the officeholders. The structural interrelatedness of institutional roles is the source of a special set of normative powers of office that are created by the rule-governed interactions that officeholders are enabled to entertain when they act in their institutional capacity. As we expound later, this approach to institutional action has the advantage of revealing the normative importance of the officeholders' structural interactions, an aspect that risks being sidelined were we to focus only on institutional rules or purposes.
To see the importance of this interactive feature of institutional action, let us start with the claim that public institutions, qua institutions, are systems of interrelated, embodied, rule-governed roles (see Applbaum 1999;Emmet 1966). Each institutional role, or office, is entrusted to an officeholder with some specific normative powers (rights and duties) regulated by a mandate that describes the officeholder's obligations, tasks, and privileges associated with that role. Each power mandate is devised in keeping with the raison d'être of the institution: The set of normative ideals that motivate the establishment of an institution and, consequently, its internal structure and the way it works. Different power mandates regulate different institutional roles with a view to ensuring that the interrelated work of the officeholders in their institutional capacity makes the institution work. In this sense, public institutions are groups of agents, whose actions are coordinated through a system of rules, through which the officeholders exercise the normative powers associated with their office with a mandate.
The central feature of this approach to institutional action is to pinpoint that institutional roles are linked in such a way that makes institutional action structurally proceed from the officeholders' interrelated conduct in their institutional capacity: Every officeholder's exercise of their power of office depends on the power exercise of the other officeholders in their institutional capacity. So, to see what a parliamentary assembly is and how it works we may not just refer to the Robert's Rules of Order. A parliamentary assembly is a group of officeholders (e.g. MPs, assistants, spokespersons, clerks) who exercise their role-related powers (e.g. voting rights, confidentiality duties) in view of institutional rules.
Notice now that, for any officeholder to be capable to perform her role, she must be in the condition of exercising her rights and duties of office as established by her power mandate. Because of the interrelatedness of institutional roles, any officeholder's exercise of such rights and duties is conditional upon the other officeholders' exercise of their rights and duties as per their mandates. When this condition is not met, institutional action fails by the failing action of the individual officeholders. This view of public institutions emphasizes that institutional action and failure are a matter for the officeholders' 'interrelated responsibility' (see Ceva and Ferretti 2021: chapter 4). In the context of institutional action, officeholders are not just individually responsible for their own exercises of their power of office; nor are they only collectively responsible for jointly serving the designated purpose of their institution (see, for example, Miller 2017, 134). A focus on the interrelatedness of institutional roles allows us to see the officeholders as a collective, but without assuming their collective agency. 6 Institutional action proceeds from the officeholders' interrelated exercise of their individual agency. Therefore, to assign responsibility for institutional action and failure requires also making institutional action and failure a matter for all officeholders, by recognizing the various patterns of interrelatedness between their roles and the exercises of power that come with them.
To track the different patterns of interrelatedness is helpful in general to understand institutional action. In the particular context of this article, it can also contribute to account for the various ways in which an entire institution may be corrupt in function of the action of those who occupy a role within it (Ferretti 2019;Ceva and Ferretti 2021: chapter 2). Three patterns of interrelatedness can be distinguished. One basic pattern of interrelatedness is summative. This pattern describes a situation in which most or many of the officeholders in a public institution fail to act according to an agenda coherent with the institution's raison d'être. For example, when several border officials allow the smuggling of weapons at the border, the action of the border agency (which, among other things, should prevent smuggling) is compromised in the sense of being corrupt.
A further pattern is morphological. It describes a situation in which just one or a minority of officeholders acts in a way that -via interrelatedness -spreads and impairs the entire institutional action. For example, in the aftermath of a conflict, one major task is to contract for the reconstruction of damaged public infrastructures. Suppose that one of the officers in charge of the allocation of reconstruction funds offers an assessment favourable to a particular private constructor, who will guarantee that part of the allocated money will flow in the officer's private bank account. Other officeholders, unaware of the deal between the officer and the contractor, could then approve in good faith the assessment report presented by their colleague, with resulting inflated costs for the reconstruction work. While none of the corrupted officer's colleagues can be said to act in a wilful corrupt way, none of them is in the position of fulfiling their duty of assigning the contract to the most efficient bidder. In this way, they fail the uphold the raison d'être of their institution as a group, which, on that ground, can be called corrupt.
One final pattern of institutional action and failure is systemic. This pattern has the most complex structure because it typically includes summative and morphological elements. Such elements concur in defining a net of corrupt actions that involve a multiplicity of agents across various institutions. Concrete illustrations of systemic corruption abound in the context of postconflict institutional action. Consider, for example, the system developed in the context of postwar reconstruction in such postcommunist countries as Bosnia-Herzegovina. In that context, the process of rapid liberalization and the availability of funds made it particularly difficult to tame the influence of wartime elites in handling reconstruction projects. The interrelatedness of different public and private, national and international actors within and outside political institutions entangles institutional action in a corrupted system, which makes institutional action fail (Chandler 2002).

An inward perspective on political corruption as a deficit of office accountability
To differentiate between and appreciate the specificities of the various patterns of institutional action and failure is both descriptively and normatively instructive. At a descriptive level, the three ideal types bring out the structures of interrelatedness among institutional roles that characterize various kinds of institutional action and failure. From a normative point of view, interrelatedness brings into focus the basic structural features that make officeholders fundamentally and structurally accountable to one another for the uses they make of their powers of office in their institutional capacity. Notably, officeholders are mutually accountable for the rationale of the agenda they pursue in their institutional capacity which, to uphold institutional action, must be coherent with the terms of their mandate.
To be sure, the officeholders in public institutions are not to be seen as a selfenclosed and self-referential elite, shielded from any external control. Officeholders are in fact also accountable to external enforcing authorities, who may intervene and distribute punishments in case of unlawful conduct. Moreover, such democratic institutions as a parliamentary assembly must respond also to the citizenry. 7 However, in the context of public institutional action, accountability has -also and distinctively -an inward-looking dimension. The distinctive interrelatedness of institutional roles makes officeholders, also and fundamentally, structurally accountable to one another for the uses they make in their institutional capacity of the powers associated with those roles. This is the idea of 'office accountability,' the regulative principle for the officeholders' conduct (Ceva 2019;Ceva and Ferretti 2021). The relational core of the idea of 'accountability' is widely recognized in the debate (see, for example, Bovens, Goodin, and Schillemans 2014;Schedler 1999). However, the notion of 'office' accountability is specific to the institutional context because it presupposes the normative order of rightduty relations, regulated by the various power mandates, that an institution establishes.
Through the lenses of office accountability it becomes apparent that in order to understand and assess institutional action, it is crucial to have access to an inward perspective too. In particular, in the nonideal conditions characterizing postconflict contexts, institutional action may not be entirely understood in view of its capacity to fulfil institutional purposes objectively considered as fixed performance indicators orlet alone -as standards imposed from outside an institution (for example, by international organizations). To adopt an inward perspective on institutional action means to focus on the importance of mobilizing officeholders, in the first person and in their interrelated responsibility, in a self-reflexive and critical exercise of interpretation and appraisal of the normative values underpinning their institution and informing their interrelated conduct. As we elaborate later, such an exercise may be carried out through various (soft law) instruments of officeholders' mutual oversight and answerability, such as channels for internal whistleblowing, sustained by an appropriate ethical training for officeholders. While we revisit such instruments later, what we have argued so far can already explain why an inward perspective on institutional action and failure allows us to pinpoint how necessary it is that officeholders be made interrelatedly directly responsible (and mutually accountable) for upholding institutional action.
To adopt an inward perspective on institutional action also reveals how political corruption, as an institutional failure, is primarily an internal enemy of public institutions, which consists in a deficit of office accountability (Ceva and Ferretti 2021: chapter 1). Specifically, we have political corruption when officeholders use the powers entrusted to their institutional roles for the pursuit of an agenda whose rationale they cannot vindicate as coherent with the terms of their mandate. As we illustrate in the previous two sections, to view political corruption this way helps us to focus on the communicative aspects of identifying the officeholders' corrupted conduct as key to anticorruption efforts. Because the corruption of institutional action proceeds from the interrelated corrupted conduct of the officeholders (via the various patterns we discussed earlier), the identification of corruption may not be performed entirely from the outside an institution (for example, by a legal or oversight authority). Put differently, to identify and address political corruption, the reference to external standards is insufficient; the direct engagement of and mutual accountability between officeholders is necessary too. We can thus situate the normative discussion of political corruption within an institutional ethics of office accountability.
In what follows, we show how this understanding of political corruption is particularly apt for understanding and developing anticorruption programmes in cases of institutional reconstruction in the aftermath of a conflict. In those cases, of course, we must consider institutional action, institutional mandates, and duties of office under less-than-ideal conditions. But especially in those cases, looking at political corruption as a problem of institutional ethics is helpful to understand what kind of wrongs and failures anticorruption should be able to target and correct.

Corruption as a consequence of conflict
Political corruption and conflicts are frequent bedfellows, and countries scoring dramatically lower in the Corruption Perceptions Index were also among the least peaceful countries in the world according to the Global Peace Index. 8 Making sense of this correlation seems thus fundamental for understanding the role and scope of anticorruption programmes.
A common explanation for the correlation between conflicts and political corruption pinpoints the causal link between the presence of conflicts and weak public institutions, which are often the source of political instability. To illustrate this causal explanation, consider, for example, the United Nations Development Program's (UNDP's) 2010 report 'Fighting Corruption in Post-conflict and Recovery Situations: Learning from the Past.' As an integral part of its efforts to strengthen public institutional systems, the UNDP has been at the forefront of alerting people about the threats of political corruption to institutional action. The UNDP 2010 Report specifically emphasizes the need to respond to these threats by sustaining local initiatives of participatory planning, monitoring, and decision-making that mainstream anticorruption measures throughout their planning and budgeting cycle. 9 Within this general effort, the core message is that failure to secure a durable peace leads to increased and entrenched opportunities for political corruption. This state of affairs results in a much higher probability of failure of anticorruption programmes, such that upholding the public institutional system becomes extremely difficult and relapses into conflict are not infrequent.
Among the reasons for engaging in anticorruption, the 2010 UNDP Report points at the negative consequences of widespread political corruption for the reconstruction of infrastructures. 10 But, certainly, whether the most conspicuous obstacle to infrastructural projects was the widespread political corruption is a matter of empirical controversy. Other significant obstacles could include, say, the administrative inability to cope with the burdens of loan negotiations with the World Bank and protocols for the initiation of the projects (Mustapha and Mangura 2010). Because these arguments are empirical, rather than normative, these controversies concerning the impact of political corruption are relevant and likely to affect the cost/benefit balance of engaging in anticorruption initiatives within institutional reconstruction programmes. This empirical, consequence-driven approach to the study of the relationship between conflict and political corruption may only lead to underdeterminate answers to the question of when and how to prioritize anticorruption in postconflict institutional reconstruction.
To corroborate the commitment to anticorruption, this consequence-driven empirical approach is insufficient. A useful integration comes from our earlier normative discussion of the structural dynamics of corrupt institutional interactions as 8 Data from the Global Peace Index 2020 (https://www.visionofhumanity.org/wp-content/uploads/2020/10/GPI_2020_ web.pdf#:~:text=GLOBAL%20PEACE%20INDEX%202020%20%20%7C%20%20,the%20measure%2C%20in%20the% 20GPI%E2%80%99s%20case%2C%20global%20peace) compared with Corruption Perceptions Index 2020 (Berlin: Transparency International, 2020: https://www.transparency.org/en/cpi/2020/index/nzl). constituting (not just causing) a threat of its own to institutional action. Focusing on political corruption as constituted by institutional patterns of officeholders' interrelatedness brings to the fore a deficit of office accountability -typical of postconflict scenarios. This deficit is an otherwise easily overlooked or underappreciated institutional failure that, as such, presents itself as a threat of its own to the institutional system. This threat is tangible and may be due either to a country's political history of dysfunctional institutional dynamics (for example, in kleptocracies) or to the occurrence of such dramatic events as armed conflicts, which disrupt institutional interactions and weaken the mechanisms of office accountability. These observations highlight the strategic role of anticorruption in institutional reconstruction. As an illustrative case, take the so-called 'transition period' in the DRC. During the years 2003-6, anticorruption initiatives were hindered by a tradeoff between overall political stability and clarity in the assignment of the respective responsibilities to the parties. The Sun City Agreement inaugurated a two-year powersharing arrangement, which constituted a government and a two-chamber parliament, including appointees from the former government, the rebels' movements, the unarmed political opposition, and civil-society organizations. All the signatories of the agreement were allocated a proportionate share of government ministries and other high-ranking institutional positions. It has been noted that political patronage, rather than competence or merit, was the rationale for these appointments as a compromise arrangement to appease social tensions (Kodi 2008). As a consequence, each appointee remained fully answerable only to his patrons, who could thus retain the exclusive power to remove him from office. This state of affairs is arguably responsible for creating incentives for the appointees to make a corrupt use of their powers of office; appointed officeholders were in a position to obtain immunity from prosecution conditional only on maintaining a good clientelist relationship with their patrons. The multiplicity of stakeholders entangled in the clientelist relations was so diffused that, while an anticorruption legal framework was developed during the transition period, the bodies responsible for enforcing it were so dysfunctional that the work of the newly established Ethics and Anticorruption Commission was totally ineffective (Hussman and Bun 2005). In his analysis of this failure, Muzong Kodi (2008) points out that this ineffectiveness was due in part to a general lack of political will on the part of local political elites to carry out anticorruption initiatives. However, Kodi (2008, 89-92) also indicates a failure of international actors and providers of development aid, who feared that a robust anticorruption action could jeopardize the electoral process and exacerbate social conflicts. The conjunction of such a domestic and international attitude favoured a laissez-faire culture among political and administrative elites.
One way to comment on corrupt forms of institutional interaction of the kind described consists in pointing out how they were to an extent generally tolerated by fear that targeting them directly through anticorruption could upset a seemingly peaceful status quo and exacerbate social acrimony. But in addition to this factual observation, a normative lesson may be learned too. A normative assessment of these experiences from the perspective we are offering shows that there are reasons not to condone political corruption on prudential grounds because that would hinder the project of upholding the public institutional system. The events in the DRC illustrate the circumstances where one can reasonably hold the normative claim that political equilibria ought not to be based on corrupt forms of institutional interaction, which are inherently inimical to the working of public institutions. In fact, condoning political corruption sanctions forms of institutional interaction that constitutively violate the requirements of office accountability, which, we have seen, is critical to institutional action.
Interestingly, the idea that fostering institutional relations of mutual accountability between officeholders should be a priority of institutional reconstructions emerges underneath such programmes as, for example, the Anti-Corruption, and Integrity in the Arab Countries programme (UNDP-ACIAC). The UNDP-ACIAC is currently piloting a number of initiatives centred on a Conceptual Framework for Corruption Risk Assessment at Sectoral Level in six Middle East North Africa (MENA) countries and territories (see Hunter 2020). One of the elements of this innovative framework is to identify crucial decision points where corrupt activities might take place. Decision points are identified at key junctures -mapped in relation to specific institutional functions and objectives -between officeholders' concrete decisions. Targeted anticorruption actions may thus be developed based on the identification of the institutional roles that appear to be particularly vulnerable to political corruption (and may compromise the entire institutional action). Our approach makes sense of the specificity of this framework as it explains and justifies the centrality of the officeholders' interrelatedness in upholding institutional action.
The UNDP -ACIAC Conceptual Framework straightforwardly exemplifies anticorruption measures addressed to individual decision-making and breaches of formal duties for private gain. But read from our perspective of an institutional ethics of office accountability, the document may also usefully illustrate the importance of identifying relevant junctures between the officeholders' individual decisions through the lenses of the interrelatedness of their actions. What is more, the adoption of an inward perspective that insists on the officeholders' capacity for self-scrutiny and self-correction can provide a justification for grounding anticorruption in a general effort of institutional capacity building by strengthening in-house knowledge and skills. Our discussion so far helps to see why this knowledge and skills are fundamental to the assessment and management of the risk of political corruption. Their importance is explained in terms of their capacity to mobilize a system of office accountability, which serves the twofold goal of identifying deviant decisions as well as the factors that may influence such decisions. Recast through the lenses of an institutional ethics of office accountability, both the task of understanding the structures of institutional decisions and the possible factors that may hinder them belong primarily to a critical and self-reflective process internal to the institution itself.
We have already acknowledged that this inward perspective should not result in a normative view of office as a fully self-referential, if not elitist, practice. The officeholders' institutional action must also respond to external authorities (e.g. law enforcing bodies or various stakeholders). But, we submit, the question of whether, for example, the lower salaries of certain officeholders, or their family's proximity to potential bribers, constitute a realistic risk of political corruption in a certain institutional context can be most proficiently spelled out from the inside an institution, by those who have a more direct access to and knowledge of institutional structures and their predicaments. To wit, this analysis should occur, in the first instance, through the sustained dialogue and mutual interrogation among officeholders with the double function of making their interpretations of institutional action explicit and clarifying their mutual expectations. This is no doubt a considerable effort, especially within relatively weak or developing institutional systems. As we shall see more specifically in what follows, this effort implies the mobilization of manifold channels for the officeholders' mutual oversight and answerability (including, e.g. procedures of internal whistleblowing), accompanied by the development of ethical training programmes. This complexity speaks of the importance of sustaining anticorruption through appropriate tools and procedures as a matter of political priority.
This brings us to a further observation of why upholding the commitment to anticorruption from within public institutions is particularly important in a postconflict situation. It is common that foreign donors to intervene with economic support to sustain a variety of reconstruction projects. This support involves a flow of material resources, which often ends up fuelling political corruption and derailing the transition to political and economic stability. This unwelcome result is due, among other factors, to the large-scale injection of material resources in an environment where the legal and institutional frameworks necessary to manage those resources are fragile. The large amounts of money involved in reconstruction projects and the highly compressed time frame in which such projects are carried out can be expected to create the substrate for political corruption to thrive and doom anticorruption to failure. Without attending to issues of good governance and effective capacity building, the risk is the perpetuation of the status quo, including the conditions under which political corruption flourishes.
These hurdles to institutional reconstruction suggest that transferring institutional models and good governance standards from one country to another has little chance of success if the rationale for upholding the institutional system is not acknowledged, internalized, and acted upon by the officeholders. As several observers have remarked, failing this precondition for the development of local anticorruption commitments, the flow of both material resources and technical aid is doomed (see Holmberg and Rothstein 2017;Zaum and Cheng 2012). In fact, shadow practices are likely to develop, thus exacerbating previous corrupt mechanisms and relations that trade on deficits of office accountability. Also in this case, our approach is distinctively helpful in explaining the centrality of the dynamics of corrupt institutional interaction and how such dynamics are critical for understanding how political corruption is an internal threat to public institutions, and what anticorruption initiatives should consequently aim to achieve.
In an attempt to promote good governance standards, international donors have often followed strategies of 'aid conditionality,' where financial support was made dependent on the introduction of reforms intended to foster good governance, including anticorruption measures. This practice, however, has often been criticized for its managerial character and for inducing the unreflective accommodation of external demands (Marquette 2011). In several cases, aid conditionality has led recipient governments to accept impositions without a genuine commitment to the demanded reforms, introduced on purely instrumental grounds, but dismantled soon after the promised aid was received. This predicament has motivated donors to pay a greater attention to the aid recipient's claim to the 'ownership' of policy reforms. It remains uncertain how such an ownership can be achieved. Of course such an achievement depends on empirical factors that escape any ethical normative theory of anticorruption. To be able to understand the extent to which anticorruption is a matter of institutional ethics of office accountability is nevertheless helpful to clarify the kind of responsibility officeholders are expected to take for institutional action and failure. As seen earlier, this is an interrelated form of responsibility resting on the internal dialogue and constant exchange between officeholders.
The interrelated mobilization of officeholders occurs via communicative processes of self-scrutiny through which the officeholders themselves can identify the risks of political corruption by mobilizing their internal knowledge of institutional mechanisms and critically engaging with their mutual expectations as concerns the working of their institution and their interrelated conduct. The officeholders' direct engagement in such processes is telling of their long-term commitment to sustaining their common effort to curb political corruption. Such a commitment thus rests on much stronger grounds than any externally initiated action aimed at the transfer of anticorruption standards promoted by means of aid conditionality.

Political corruption as a cause of conflict
Failures of anticorruption have often contributed to outbreaks of new violence and relapses into conflict. This has led some observers to see a double causal link between conflict and political corruption. Not only, as discussed in the previous section, does conflict generate political corruption, but also widespread political corruption is often a cause of conflicts as it contributes to protests that may lead to social unrest and civil war. A telling illustration of this correlation comes, for instance, from the 2017 Transparency International (TI) Report 'The Fifth Column: Understanding the Relationship of Corruption and Conflict.' The report explores political corruption as a cause of such political problems as economic inequalities, rent-seeking competition, poverty, terrorism, and ultimately conflict. Consider, for example, how political corruption can exacerbate social divisions and fuel people's resentments of the political class. The TI report regards political corruption as a privileged way for kleptocratic elites to extract maximum resources from the state. These privileges often occur to the detriment of the provision of public services and of civic equality. In this way, political corruption aggravates the many causes of poverty and underdevelopment, which are often at the origin of social and armed conflicts.
Looking at the events of the Arab Spring, starting from December 2010, and in particular the evolution of the events in Tunisia, can further elucidate the role of political corruption in fuelling conflict. The outbreak of the protests is arguably an indicator of the importance of perceived conspicuous political corruption as a cause of the diversion of wealth and attribution of privilege to an elite; this state of affairs may well nourish public outrage and revolt. As it has by now become apparent, through a practice of government licencing and cooperation in such crucial sectors as transportation, education, and the media, profits were channelled to companies operated by the extended family and political allies of the Tunisian dictator Ben Ali (see Nucifora, Churchill, and Rijkers 2015;Transparency International 2017). The ability of Ben Ali's family to transcend the rule of law was exposed in one of the US government cables disclosed by WikiLeaks (via @spbaines). In this way, Tunisians became aware of the high rates of political corruption in the country, its reach into a number of business sectors, and the related ability of the governing leadership to manipulate the law and turn it into an instrument to the advantage of the president's family. The disclosure of these corrupt activities can easily be seen as a major contribution to triggering popular protest and uprising. The Tunisian case can thus be taken as an illustration of how social and political inequalities may foster people's resentment in a way that often results in conflicts and even more political corruption.
How does the alleged empirical correlation between political corruption and social conflict matter from a normative point of view? By now, we have the normative resources to see how a full response to this question should also consider the dysfunctional institutional interactions in which, as argued earlier, political corruption consists. Reconsider, for example, the Tunisian context, but also that of the many 'colour revolutions' across the countries of the former Soviet Union and in the Balkans during the early 2000s. In those contexts, the violence of the protests against the high corruption rates can be taken as indicative of a deeper structural problem irreducible to complaints of poverty and discrimination. 11 The vehemence of those protests arguably indicates a profound sense of structural injustice, as it derives from a feeling of being entangled in an irredeemable web of corrupt institutional interactions, in which it is impossible to hold officials to answer for their conduct. Such a feeling matters from a normative point of view not just as it pinpoints social material inequalities per se. Our earlier discussion allows us to see how the frustration triggered by such a web of corrupt uses of powers of office may be read also and fundamentally as a reaction to a violation of office accountability. Because, as seen, office accountability designates a fundamental duty of office in public institutions, its structural violation can be the ground for overthrowing the existent political order. The main normative implication of this conjecture is that the fight against political corruption may not be confined to (while it encompasses) discrete actions tackling the material consequences of corrupt public institutions. Anticorruption also requires what we have presented as a generalized effort to uphold the public institutional system by sustaining systematically an institutional ethics of office accountability.
A concrete illustration of the importance of a systematic, ethically sustained anticorruption action along the lines we have sketched comes from an increasing number of international programmes. Consider, for example, how the Tunisian government has asked the support of the MENA -OECD Governance Programme in the area of the ethics and values of the public sector in order to improve their administrative culture. 12 The OECD supported the Tunisian administration through programmes of capacity building aimed at promoting an inclusive discussion among various stakeholders on the foundations of institutional integrity. The Tunisian government launched a national campaign on the values of ethics and integrity, which resulted in the development of 11 For a characterization and an analysis of 'colour revolutions,' see (Mitchell 2014). 12 For a general presentation of the programme, see https://www.oecd.org/mena/governance/; for the Tunisian initiative, see https://www.oecd.org/mena/governance/anti-corruption-tunisia.htm. Some observers have pointed out that one main failure in dismantling corrupted networks is the persistent relation between local kleptocratic elites and foreign interest (see, e.g. Anderson 2011). In such cases, mobilizing and sustaining new ethical resources from within an institution may be the only viable path for discarding and changing those relations.
a sustainable training environment for public administrators. The promotion of these kinds of ethics-based initiatives, and the relative attribution of public resources in a context of general scarcity, is not always easily justifiable. One of the reasons is that the impact of these initiatives is difficult to assess in the mid-term, since they do not offer outputs that can be easily indexable and measured in terms of their costs and benefits. By vindicating the importance of an institutional ethics of office accountability, our approach offers a strong normative justification for this kind of initiatives. It does so by pointing out that political corruption operates from within public institutions and triggers a vicious circle where the net of dysfunctional institutional interactions expands and legality sinks. As the problem is internal to an institution, it also requires an inward perspective to emerge in full, and the activation of internal resources to be tackled in a sustained manner. As we expound in the next section, this activation is crucial as long as it can empower officeholders to take the interrelated responsibility for institutional capacity building and corruption prevention.

Upholding the public institutional system and the inward commitment to anticorruption
Our normative discussion of anticorruption in institutional reconstruction programmes shows, inter alia, the ambivalence of the assessments of political corruption that mainly focus on its negative social, political, and economic effects. Besides its impoverishing and destabilizing effects, political corruption may arguably provide the resources for consolidating the power of some political and social agents, provide access to ready-touse resources for (re)building infrastructures, and even secure services to the population. In all these ways, in the nonideal circumstances of conflict, political corruption may even appear to serve peace-building goals. What is more, as seen in the previous section, many anticorruption initiatives of institutional reform and capacity building (like the MENA -OECD Programme discussed earlier) necessarily pursue mid-or long-term objectives, which bear immediate costs but have effects that are not immediately visible or easily measurable. Insofar as political corruption is assessed only in view of its negative costs, it is not unreasonable for local governments as well as international providers of aid to base their anticorruption interventions on weighing the alleged benefits of political corruption against the costs of combating this phenomenon firmly.
If the justification of specific anticorruption interventions is subject to a cost/benefit analysis, support for anticorruption interventions may become erratic, and the very commitment to anticorruption may falter under the pressure of a generalized scepticism. This scepticism towards the commitment to anticorruption clashes with the proclaimed recognition of the importance of establishing the fight against political corruption as a priority in postconflict reconstruction programmes. Telling illustrations of this recognition can be found at the core of the two reports by the UNDP and TI we have considered in the previous sections. In line with the ambitions of the 2030 Sustainable Development Agenda, the two reports point to the concrete risk that political corruption squanders peaceful settlements and, more generally, hinders reconstruction programmes at the social, economic, and political levels. These considerations are aligned with our normative claim concerning the structural and fundamental challenges of political corruption for the public institutional system. In view of this claim, we can now cash out the normative implications as concerns what approach to anticorruption is justified and required in postconflict situations and in the context of newly established public institutions.
A more systematic anticorruption action is required beyond reforming the administrative sector. Anticorruption must aim also at laying the very structural foundations of a new public institutional system grounded in security and peace. Moreover, integrating a comprehensive anticorruption agenda into institutional reconstruction programmes means that the fight against political corruption may not be left entirely to administrative law (corroborated by the corrective instruments of criminal law), and the actions of external entities, such as aid providers. As seen at the end of the previous section, this integration requires conceiving of anticorruption within an institutional ethics of office accountability, which (also with the instruments of soft law) calls on officeholders to take action from within their institution. This inward perspective becomes accessible only by understanding political corruption as a dysfunction of institutional interactions, which erodes institutional action from within.
To be sure, this warning is by itself hardly a novelty, since many institutional observers have remarked on this predicament (see, for example the United Nations Office on Drugs' and Crime's 'Doha Declaration' and Oxfam America 2010). Many international bodies engaged in the fight against political corruption have recommended that a mentality of anticorruption should be ignited in the target society. Instruments of soft law -although not uncontroversial -have often proved able to trigger important positive changes (Rose 2015). Concretely, this kind of anticorruption efforts may be sustained by developing safe internal reporting procedures (e.g. a hotline ensuring the confidentiality for whistleblowers) and ethical training programmes for current and prospective officeholders. Such training programmes are crucial for enhancing norms of conduct and internal self-discipline among officeholders (Transparency International 2017, 18). Our discussion provides a solid normative justification of such programmes insofar as they aim at promoting an institutional ethics of office accountability. This justificatory work is necessary and important to give coherence to many proposals that are already been advocated by scholars and practitioners alike, but currently lack the support of a unifying normative framework.
Our framework makes it possible to find a normative justification for a variety of institutional anticorruption practices from 'within.' In particular, it offers the instruments to better see why anticorruption should not entirely be resolved into a call for external interventions and the provision of either positive or negative sanctions to cultivate the appropriate officeholders' mindset. Where anticorruption initiatives come only from the outside (e.g. when they are led by international organizations), the risk is that those initiatives come as a mere formal list of standards, which fails to engage the very actors that partake in the relations that political corruption compromises. Moreover, especially in contexts of newly established public institutions, new rules and codes may be difficult to enforce and become a useless, if not a counterproductive, burden for the judicial system with the result that censoring corrupt behaviour becomes an impossible task. Our proposal to rethink anticorruption as a component of an institutional ethics of office offers the normative resources for justifying initiatives aimed at the promotion of changes in the structural interactions between officeholders, by engaging them directly in practices of accountability and, thus, drawing on their interrelated responsibility to uphold the public institutional system. Without such a commitment to office accountability, it is highly improbable that programmes of institutional reform alone can fight an enemy of public institutions that is primarily internal to those institutions and entrenched in their basic structure.
In the previous sections, we have suggested how this commitment justified in terms of institutional ethics may shed a new light on the kind of anticorruption initiatives identified in some such capacity building projects as the UNDP -ACIAC Conceptual Framework for Corruption Risk Assessment at Sectoral Level and the MENA -OECD Governance Programme. But our proposal for rethinking the standard legalistic approach to anticorruption through the realization of office accountability may also justify new practices for enabling officeholders to answer for their conduct in ways that sustain their taking direct anticorruption responsibility. 13 Consider, for example, the management of infrastructure-reconstruction projects. In postconflict scenarios, such projects seem paradigmatically open to political corruption (via bribes and various forms of clientelism) in a way revealing of and treading on deficits of office accountability. A standard legalistic approach to this problem would try to pre-empt corrupt behaviour by imposing harsher regulations and, perhaps, reducing the margins of officeholders' discretion in making decisions concerning public procurements. Instead, from the perspective of an institutional ethics of office accountability, the way is open for developing new practices that enhance, rather than restrict, the margins of the officeholders' action as they aim at mobilizing anticorruption resources from the inside of the relevant institutions.
One concrete proposal for the mobilization of officeholders in the fight against political corruption is the so-called 'four eyes' procedure. This kind of procedure requires that at least two officeholders approve of any decision before the parties can sign a contract. In a refinement of this basic idea, a random rotation of authorized individuals serves as the second pair of eyes, so that it cannot be known in advance who exactly will be called to vet any given decision. 14 Another concrete form to uphold office accountability consists in projects for the digitalization of decision-making records. By facilitating the access to relevant information by other institutional members, digitalization may enhance officeholders' capacity mutually to respond for their decisions by rendering the decision-making process more readily open to internal (self-)scrutiny. Digitalization may also promote the officeholders' commitment to office accountability by encouraging them to act in a way which, if asked, they could vindicate in good faith as coherent with the terms of their power mandate. The lack, 'disappearance,' or inaccessibility of documents is a typical feature of corrupt public institutions; information and communication technologies may contribute to enhance office 13 For a discussion of whistleblowing as an answerability anticorruption practice of this kind see Ceva and Bocchiola 2018;Ceva and Ferretti 2021: chapter 5. 14 Some experimental studies have casted doubts on the proficiency of this methodology in actually preventing corrupt practices as they risk reproducing in decision making officeholders' asymmetries of power, see for example Schikora 2011; Lambsdorff and Frank 2011. Such remarks are important to assess the efficiency of the particular method to initiate a critical reflection about the cluster of accountability practices into which the institutional ethics of office accountability we have defended could (or could not) be translated in specific circumstances.
accountability and thus to a reduced level of political corruption, especially important in the area of fiscal management and public procurement. 15 Our discussion offers the normative resources to appreciate why, in the context of postconflict institutional action, it is particularly important to ensure that primary systems of financial control are in place within the newly established institutions before creating secondary bodies such as anticorruption commissions. The officeholders' active engagement is such internal systems is, as seen, of primary importance and directly entailed by their normative commitment to office accountability. Similarly, we can see how it is best to avoid rushing the coercively regulated introduction of rules of behaviour, unless they intercept the officeholders' ethical standards. The centrality of office accountability and the officeholders' call to direct action explains why it may in fact be more promising to resort to codes of conduct instead. Such codes are not primarily intended to constrain the officeholders' action, but they can enhance office accountability by specifying the officeholders' reciprocal expectations. By their very nature, the more such expectations are formulated by involving local stakeholders, the greater action-guiding capacity may be expected of them (Johnston 2011;Rose-Ackerman 2001). Of course, such an exercise must be sensitive to the risk of the socalled 'expectation trap,' in which local officeholders' mutual expectations turn out to be too low to live up to office accountability and progress may be slower than desirable. This risk, however, seems worth taking if it is necessary to ignite positive, durable, and ethically sustained institutional changes (Johnston 2014).
These strategies are but an illustration of the possible components of a programmatic framework centred on i) internal preventive (rather than just external and remedial) anticorruption practices; ii) the development of procedures and platforms to promote the officeholders capacity mutually to respond for their conduct; and iii) the introduction of tools and procedures to foster the officeholders' risk alertness and awareness of their mutual expectations about how their conduct can sustain (or hamper) institutional action. Our aim has been to show that these strategies -in part already being experimented in some postconflict contexts -can be generally justified within a comprehensive effort to sustain institutional action on the ground of an institutional ethics of office accountability. Crucial to this ethics is the promotion of the mutuality of the interaction between officeholders, thus strengthening the interdependence of their uses of their powers of office in their institutional capacity. In this way, the justified scope of anticorruption within postconflict institutional reconstruction programmes is not restricted to limiting the occasions for individual corrupt conduct. Anticorruption extends to the improvement of the structural institutional interactions between officeholders that may otherwise prolong the legacies of conflict and possibly fuel new ones.

Conclusions
We have devoted this article to the development of a new normative framework for justifying the integration of anticorruption into postconflict institutional reconstruction programmes, within an effort to promote an institutional ethics of office 15 For example, for an account of e-Taxation and e-Procurement project in Croatia, see Council of Europe 2006, 89. accountability centred on the mobilization of the officeholders to opposing political corruption.
We have developed such a framework in a critical engagement with a common approach to anticorruption within postconflict institutional reconstruction programmes. This common approach aims primarily to counter the extrinsic negative political, social, and economic effects of political corruption, and implement legal anticorruption standards and punitive measures, often initiated and sustained by external entities. We have argued that political corruption is an urgent condition to address not only because it is a costly consequence of some conflicts and the cause of many others. Rather, our normative analysis has suggested a fundamental sense in which political corruption is a condition that reveals a deficit of office accountability inherent to the structural relations between officeholders, due to fragile structures of institutional interaction, and that must be targeted as such. It follows that anticorruption should not be understood simply as a set of administrative or legal standards aimed at sanctioning corrupt individual conduct, and remedying the negative effects of political corruption on people's rights and opportunities. Anticorruption should aim at developing institutional interventions to sustain the interrelated actions of officeholders within public institutions by upholding an institutional ethics of office accountability, which engages officeholders directly to take anticorruption interrelated responsibilities.
Our main argument thus leads to the conclusion that anticorruption should aim at restoring or establishing institutional interactions that realize a normative ideal of office accountability as the cornerstone of any successful programme for upholding institutional action. This approach may result in a very slow pace change and solidification of institutional interactions, but would constitute the basis for a robust project of institutional reconstruction. The task of upholding office accountability in public institutions is always crucial, but in phases of postconflict institutional reconstruction, it seems even more urgent in order to overcome the legacies of disrupted relations that political corruption instantiates.