Show Summary Details

Page of

PRINTED FROM OXFORD HANDBOOKS ONLINE ( © Oxford University Press, 2018. All Rights Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a title in Oxford Handbooks Online for personal use (for details see Privacy Policy and Legal Notice).

Subscriber: null; date: 17 November 2018

Accountability and Ambiguity

Abstract and Keywords

Ambiguity is intrinsic to life. Yet, mainstream accountability theory reduces its area of application by not taking ambiguity seriously. The literature treats too many aspects as exogenous to politics and is unlikely to capture accountability dynamics in contemporary representative democracies in transformation, including the European Union. Theorizing accountability requires going beyond predetermined principals and agents, detecting deviance from authoritative orders and rules, and disciplining unruly agents. An institutional perspective is offered, relaxing assumptions regarding what accountability means and implies; what is involved in demanding, rendering, assessing and responding to accounts; what factors foster effective accountability; and how accountability regimes emerge and change. The fluidity and unresolved conflicts of political life make it difficult to correctly assign causal responsibility and to learn from experience. Accountability processes, nevertheless, provide occasions for searching for and testing collective purpose, intelligence, meaning and political equality, as part of institutionalization and de-institutionalization processes.

Keywords: accountability, ambiguity, compliance and control, dynamics of accountability, European Union, institutional perspective, learning, representative democracies, unresolved conflict

Beyond Compliance and Control

The argument of this chapter is that mainstream accountability theory reduces its own area of application by not taking ambiguity seriously enough. Theorizing accountability requires going beyond the dominant concern in the literature organized around compliance and control: to detect and measure deviance from authoritative orders, rules, standards and contracts; to discipline unruly agents and make them comply, thereby solving the problems of principals.1

The compliance-control perspective is too static, treating too many aspects of accountability as exogenous to politics and accountability processes. For example, principal–agent approaches usually assume pre-determined principals and agents. Authority and success criteria are embedded in normative theories and formal-legal institutions prescribing chains of delegation/authorization and representation/accountability. A priori assumptions are made about information asymmetry and conflict and the mechanisms through which accountability works. These assumptions about political institutions and agency are likely to apply to some political orders, settings, and situations, and not to others. They are in particular unlikely to capture accountability in polities in transformation, and the aim of the chapter is to supplement mainstream theory by offering an institutional perspective.

Compliance-control is only one aspect of accountability relations and processes. It may, therefore, be fruitful to relax assumptions regarding what accountability means and implies; what is involved in demanding, rendering, assessing, and responding to accounts; what factors foster effective accountability; and how accountability regimes emerge and change. It may be useful to examine a priori assumptions regarding purposes formulated by pre-determined principals, mediated through competitive (p. 107) elections, laws, hierarchical arrangements, and dyadic relationships between the people, a representative assembly, government, and non-elected agents.

“Accountability” refers to being answerable to somebody else, to be obligated to explain and justify action and inaction—how mandates, authority, and resources have been applied, with what results, and whether outcomes meet relevant standards and principles. However, the precise content of and assumed best way to ensure accountability vary across polities, policy areas, groups, and time (Stone 1995; Bovens 2007; Bovens, Curtin, and t’Hart 2010). In European democracies committed to the idea that rulers should be accountable to the ruled, there are huge variations in what accountability means and implies and when the demand for compliance is met (Verhey, Broeksteeg, and Driessche 2008).

“Ambiguity” implies a state of having more than one meaning and accountability under ambiguity refers to calling for, rendering, assessing, and sanctioning accounts in situations where objectives, technology, and experience are unclear, participation is fluid, and the legitimacy of different resources is contested. It is commonplace to assert that accountability thrives on clear and consistent authority, mandates, responsibility, rules, standards, goals, and expectations. Ambiguity is the enemy of effective accountability and should be eliminated. In contrast, the chapter explores implications of seeing ambiguity as intrinsic to democratic politics and human existence.

Ambiguity as the Enemy of Accountability

In democracies accountability is linked to a belief in human agency, purposeful choice and history determined by human will, causal understanding, and control. Actors are accountable for the exercise of their powers, differentiating between authorized and unauthorized use of discretion. “Accountability abhors ambiguity” and requires that “clarity of accountability and contribution must be one of the attributes of every single role that makes up the system” (Porter-O’Grady, Hawkins, and Parker 1997, 54). The aim of accountability regimes is to reveal incompetence, fraud, malpractice, and abuse of power.

The idea that ambiguity is the enemy of accountability has roots in democratic theory and organization theory. A democratic article of faith is that “the public is ultimately the sole source of sovereign authority, and it is the public to whom all public officials ought ultimately to be accountable” (Goodin 2008, 164). Public accounts are vehicles for popular sovereignty (March and Olsen 1995) and effective accountability requires constitutional and formal-legal clarity regarding who is obligated to render an account, and who can call whom to account for what under different circumstances and sanction malpractice.

The idea is also central to conceptions of formal organizations as instruments for problem-solving and conflict resolution. Organizations are tools of rationality, (p. 108) effectiveness, and efficiency characterized by clear and consistent goals and formalized division of work, information, power, and responsibilities. Organizations are deliberately structured and restructured and there is a constant need to provide mechanisms of control to check on fulfillment of orders and adherence to rules. Supervision is built into the hierarchy of authority. The role of “higher-in-rank” includes the obligation to check on the performance of the “lower-in-rank” (Etzioni 1964, 25).

Ambiguity as Intrinsic to Life

Ambiguity is not necessarily transitory or pathological. Ambiguity is intrinsic to human existence according to existential philosophy (Beauvoir 1972) and a core aspect of political and organizational life (Cohen, March, and Olsen 1972, 2012; March and Olsen 1976; March and Weissinger-Baylon 1986). When theorizing accountability it is necessary to take into account the ambiguous character of the foundational standards and principles of democracies (Connolly 1987). Demands for clarity and consistency are hard to meet because western culture is incoherent. There are unresolved conflicts between rival traditions regarding how concepts such as rationality and justice should be interpreted (MacIntyre 1988) and a cacophony of interpretations of what democracy means and implies (Hanson 1987, 69). Normative theories and formal-legal institutions give limited behavioral guidance. Democracies legitimize competing opinions and there is disagreement regarding what institutions make accountability effective (March and Olsen 1995, 59,162).

Since modern politics unfolds within and between organizations and institutions, it is worthwhile to observe that organizations are sometimes “organized anarchies” characterized by ambiguous preferences, unclear causal understanding, contested authority, and fluid participation (Cohen, March, and Olsen 1972, 2012). Accountability processes are then less integrated by institutions, and politics is event-driven, involving improvisation and elements of chance. Ambiguity is also embraced because detailed rules may be counterproductive in terms of outcomes and adaptability (Goodin 1982, 59–72).

The relevance of ambiguity is illustrated by processes of political transformation such as those currently taking place in Europe. For centuries the sovereign state has been the basic unit of political organization and identification and conceptions of accountability have developed in the context of the emerging state. The institutional foundations of the Westphalian state have, however, been challenged. Challenges embrace Europeanization, internationalization, a new Zeitgeist, and an international wave of public sector reforms. The role and significance of the territorial state, representative democracy, elected representatives, experts, and citizens have changed in ways that require reconsideration of conceptions of accountability. “Monitoring democracy” is presented as a new historical type of democracy involving continuous public scrutiny and control over power-holders across territorial boundaries and institutional settings (Keane 2009). The development calls attention to how institutions affect ambiguity and accountability.

(p. 109) Institutions, Ambiguity, and Accountability

Institutions draw the borders of, but do not eliminate, ambiguity. The ordering effect is strongest in settled polities with well-entrenched institutions staffed by experienced, well-socialized people performing socially standardized activities. It is less strong in unsettled polities with weak or competing institutions staffed by inexperienced personnel doing novel things (March and Olsen 1976, 50).

Institutionalization refers to processes through which something diffuse, unstable, and unfixed turns into something that is settled and integrated (Selznick 1992). Institutionalization implies increasing clarity and agreement about accountability practices. Standardization and formalization reduce ambiguity, uncertainty, and conflict concerning who is accountable for what, to whom, when, and how. Some ways of acting are perceived as natural and legitimate. Thus there is less need for using incentives or coercion to make people follow prescribed accountability rules, the need to explain and justify practices is reduced, and there is less need to struggle for resources (Olsen 2008, 196). Attribution of accountability is guided by socially validated, relatively well-known, clear, and stable roles, rules, routines, procedures, doctrines, expectations, and resources. However, institutions exist because people believe they exist. They require continued public recognition and acceptance (Searle 1995, 1, 45) and sometimes institutions face erosion, decay, and revolt. De-institutionalization generates ambiguity, uncertainty, disorientation, and conflict.

As a supplement to formal-legal interpretations of accountability institutions, an institutional approach offers ideas regarding how “living” institutions affect conceptions and practices of accountability and how accounts are influenced by and influence a political order. A living institution prescribes rules of appropriate behavior for different actors in different situations. Organizing ideas provide norms of assessment and conceptions of reality that explain and justify rules and practices, and institutionalized endowments make it (more or less) possible to act in accordance with prescriptions. Institutions constitute and influence actors, identities, and affective ties. They have some autonomy and dynamics of their own and robustness regarding deliberate reform efforts and environmental change, generating “historical inefficiency,” rather than simple equilibria (March and Olsen 1989).

The organization of accountability in contemporary democracies is characterized by varying degrees of institutionalization, specialization, and coordination. Over the years a multitude of institutions, organizations, networks, and communities of account formation, authorization, legitimation, dissemination, and assessment have emerged. Some have specialized in providing normative standards and certifying institutions, or monitoring, analyzing, and assessing performance without authority or resources to sanction misconduct. Others sanction abuse of power, but depend on information and analysis from outside. Besides legislative and judicial scrutiny, there are independent auditors, ombudsmen and other complaint mechanisms, epistemic communities, think-tanks, (p. 110) credit-rating and standard-settings agencies, tribunals, and committees of inquiry operating at an arm’s length of popular control. New accountability relationships have been added to old ones, creating complex layers and combinations of co-existing, dynamic accountability structures and processes.

Multiplication of scrutiny and account-giving may foster redundancy that generates waste and inefficiencies. Yet redundancy may also make accountability arrangements more reliable and facilitate adaptation to shifting circumstances (Landau 1969). Theorizing accountability, therefore, requires attention to the complexity and dynamics of accountability relations and processes, rather than assuming hierarchical, dyadic, and predetermined principal–agent relations. Theorizing implies exploring how accountability regimes work and change in more or less institutionalized settings and well-structured, recurring, and consensual situations, involving different degrees of ambiguity.

Institutional Sources of Ambiguity

To legitimately call someone to account requires that an actor has some autonomy and discretion that can be used or misused judged by some normative standard. Principals and agents are less likely to deviate from what they are supposed to do when (Olsen 2013):

  • Compliance is secured through external controls of opportunity and incentive structures and the expected utility of complying is always greater than not complying.

  • Reciprocal control is established through vertical or horizontal separation of power and expertise, checks and balances.

  • Congruence in norms, common understandings, and role expectations are achieved through recruitment. Reliable and competent behavior is guaranteed by selecting and removing actors with certain characteristics (e.g. members of a specific party or profession) to/from office.

  • Agreement is reached through communicative action and reciprocal discovery of normative validity and the best argument through deliberation among initially conflicting parties.

  • Self-control is achieved through socialization, internalization, habituation, and character formation, rendering the appropriate norms and codes of conduct understandable and respected.

  • Available resources make it possible to follow behavioral prescriptions and proscriptions.

In settled polities there is widespread agreement regarding who is accountable to whom, for what, under what circumstances, according to which criteria, and therefore who should be blamed if things go wrong. Normative concerns and causal beliefs are embedded in traditions, morals, and institutional spheres with some autonomy (p. 111) and resources of their own. There are repertoires of socially constructed and validated accounts and responses to accounts, influenced by what is intelligible and appropriate in specific political-cultural contexts. Favored stories are embedded in ideologies and traditions and there are continuous attempts to fit events to those stories. Routinized accounts focus on deviances from shared rules and expectations and provide reassurance that events are controllable or have the appearance of control (March and Olsen 1995, 149,161). Those authorized to call someone to account do so with reference to shared norms, commitments, purposes, and expectations. Those who have an obligation to render accounts to some legitimate authority do what they are supposed to do. Activities are recorded in routinized reports, but there is little perceived need to explain and justify what is done. Whereas there is interpretation, deliberation, and bargaining regarding what accountability means and implies in individual cases and situations, the politics of accountability takes place within institutional constraints and with modest controversy.

Under such conditions it is not unreasonable to treat authority, power, mandates, standards, goals, and conflicts as exogenous and pre-determined. The task is to identify the technically most efficient means for reaching pre-determined ends within existing constraints and sanction violation of rules and contracts. For students of politics, however, accountability is more interesting when such factors are seen as endogenous to politics and accountability processes. Accountability is related to fundamental issues in political life and accountability processes provide occasions for debates and struggles over authority, power, norms, worldviews, and responsibility, crucial for the legitimacy of a political order. A complication is that the meaning of “accountability” becomes more complex, a fact with special relevance for unsettled polities characterized by weak or competing institutions with a modest ordering effect.

In unsettled polities there are competing claims of representation and role-conceptions, procedural and result-oriented assessments. Normative standards and rules are rarely exhaustive or self-interpreting. Understanding causal systems and assigning accountability is difficult. It is not clear what has happened, why it happened, and whether what happened was good, making it difficult to establish how well different institutions or actors have performed. Under conditions of complex and dynamic interdependencies, where the performance of one actor is dependent on the performance of others, chance events, and long and uncertain causal chains with significant effects surfacing years later, the assumption that actors can be made accountable by disentangling their contribution to fiascos and successes is problematic (March and Olsen 1995, 157–58). The assignment of accountability may be capricious, mistaken, controversial, and politicized.

The politics of accountability

Institutions affect but do not determine ambiguity and accountability. If institutions were perfect, there would be no discretion and no reason for studying accountability. (p. 112) Or, accountability would be limited to detecting misunderstandings, incompetence, and criminal acts. However, claims about accountability deficits and disillusion with traditional forms of representative democracy suggest that practice is less than perfect, and when causality is too obscure to be established validly in terms of correctly assigning responsibility politics acquires a more important role in accountability processes. Possibilities arise for exercising political visions, claiming virtue and victory, blaming, shaming and identifying scapegoats. Polities in transformation, therefore, provide “windows” to understanding aspects of accountability largely ignored by compliance-control approaches.

Democratic politics involves more than making, implementing, and enforcing policies. Democracies are also collections of interlocking communities of explanation, justification, and criticism. Making the world intelligible in normative and causal terms is central to political life (March and Olsen 1976) and the search for identity, belonging, purpose, direction, and meaning can be as important an aspect of accountability processes as compliance and control. Accountability processes may promote a culture of cooperation, compromise, and integration, or suspicion, confrontation, and disintegration. Under some conditions accountability processes have an integrative effect and are conducive to intellectual and moral self-development as well as self-government. They ameliorate the moral qualities of individuals and society through the internalization of a democratic and civic ethos; improve communication, learning, and epistemic quality; contribute to power-equalization and political equality; and help actors find meaning in life through reflection and reasoning together. Under other conditions accountability processes provide a rhetorical arena for political struggle, splitting the population.

Rendering and responding to accounts may create controversies and denial, or become accepted as legitimate. An event may be interpreted as the result of purposeful acts or a necessity, accident, or misunderstanding. Responsibility may be disclaimed and shifted to someone else, or poor results said to be caused by inadequate resources. Behavior may be excused, forgiven, and justified, or disapproved of and punished. The style of presentation of an account, for example, humbly submitting to criticism, expressing regret; or excusing, justifying, defending, or reframing interpretations may be more important for responses than the act to be accounted for (Dubnick 2005). Sometimes account-giving may become a ritual providing illusions (Gustavsson, Karlsson, and Persson 2009).

Ambiguity may be seen as a political necessity. In order to reach agreement, compromises are made sufficiently ambiguous to allow competing interpretations, and change may be easier when it involves modifying the relations between co-existing, ambiguous principles rather than one clear principle replacing another. Under some conditions ambiguity, concealing one’s preferences rather than taking a clear stand, may be an electoral strategy (Shepsle 1972; Page 1976; Alesina and Cukierman 1990). Superiors wanting to avoid association with torture may prefer ambiguity regarding how to handle prisoners. By not defining precisely what it means to “apply pressure” they shift accountability onto the lowest ranking and least powerful (Manning 2004).

(p. 113) Who to praise or blame does not necessarily depend on hard evidence and correct causal understanding. Interpretations of experience compete for acceptance on the basis of evidence or power (March 2010) and the ability to gain acceptance for a specific frame, type of discourse, language, interpretative community, or a particular interpretation, is a source and indicator of power (March and Olsen 1995, 180).The politics of accountability involve both the pursuit of accountability within an accountability regime and efforts to change established regimes. There is no clear-cut separation between the two, yet they are discussed separately in the next sections.

The Pursuit of Accountability

Through what mechanisms is effective accountability obtained? Democratic theories emphasize providing information and imposing sanctions as requirements for effective accountability. Because in democracies accounts are likely to be contested, they are often constructed in encounters among contending accounts. The influence of citizens is seen to depend on competing accounts and sources of information, and mechanisms for exploring, testing, gathering support for, and authorizing accounts are assumed to be supported by institutions securing freedom of expression, legitimate opposition, independent auditors, an active civil society, and a free press (March and Olsen 1995, 163).

The standard story of representative democracy gives priority to accountability through electoral mechanisms, and compliance-control approaches mainly deal with accountability in settled polities embracing ideas such as the sovereign people, sovereign parliament, and responsible minister. Popular rule is assumed to be expressed, implemented, and enforced through elections (Przeworski, Stokes, and Manin 1999). An example is parliamentarianism portrayed as a chain of dyadic relations. The authentic will of the people is expressed and control is exercised through competitive elections. The legislature makes laws and delegates authority to a hierarchy of non-elected public officials and holds them accountable (Strøm, Müller, and Bergman 2003). Self-interested principals control self-interested agents through incentives.

The problems of the standard story are well known. Problems do not necessarily originate with agents. Accountability deficit is caused by institutions and actors who are assumed to call others to account, but lack the motivation, time and energy, knowledge, or capabilities for reliably monitoring, assessing, and sanctioning agents’ behavior and performance (Busuioc 2010, 220–23). For example, democratic accountability assumes an informed citizenry that knows what powerful agents are doing and the evidence and reasons behind their behavior. This ideal is difficult to realize and the European Union (EU) illustrates that theorizing accountability must include situations where there is no unitary and stable “people” to be served. There are competing claims regarding what citizens want and what their interests and needs are. What normative criteria and causal understandings shall have priority, and what jurisdictions and responsibilities different levels of government, institutions, and actors shall have are contested. Many criteria (p. 114) and audiences are mobilized in partly uncoordinated ways and it is not clear who can legitimately call whom to account for what, or who controls relevant incentives and information.

Concerns regarding citizens’ inability to make rulers accountable need to be supplemented by citizens’ responsibilities toward the community at large as the foundation of democracy and ultimate source of power. Rather than assuming a sovereign people accountable to no one, it is important to study how citizen exemption from accountability masks the contribution of ordinary citizens to democratic successes and failures (March and Olsen 1995, 153). An institutional approach takes into account processes of internalization and identification through which citizens come to accept codes of appropriate behavior as legitimate, and at stake is whether citizenship is primarily seen as carrying rights and consuming public services or implies adherence to socially validated behavioral rules. It is a question of whether citizens see themselves as responsible members of a political community, accountable to each other not only for following laws but for participating in political and civic life, keeping informed, not making unattainable demands, and accepting duties towards the welfare of other citizens. The issue is not only whether representative democracies are able to live up to the expectations of citizens, but also whether citizens’ expectations make democracy possible and where expectations and aspirations come from.

Principals may provide agents with contradictory or unrealistic criteria, expectations, and aspirations. A culture of compromise with ambiguous lines between incumbents and opposition makes responsibility and accountability unclear. Political rhetoric fit for mobilizing support is not always fit for making programs work well. Reforms driven by political ideology rather than evidence and analysis produce perverse effects. Trust and deference are weakened through intense monitoring. Tripartite accountability relations among logics of politics, administration, and markets have turned out to be difficult to reconcile. Administrators have to decipher political decisions, legal documents, and a variety of pressures, and develop practical, viable solutions, and it has been argued that the current wave of delegation requires a reinvention of accountability (Gilmour and Jensen 1998; Klinger, Nalbandian, and Romzek 2001, 134–7).

For example, in the European Union, actors are required to provide accounts to a multitude of forums and satisfy multiple, contested, and ambiguous accountability claims. They are expected to do so in a variety of ways and through different channels, requiring different types of information, explanations, and justifications in terms of competing normative standards (Bovens, Curtin, and t’Hart 2010, 40). At different points in time, different accountability relationships are activated by and in relation to elected representatives, individual citizens, specific constituencies, mass media, democratic and civil standards, laws, financial auditing institutions, epistemic communities, professional codes of good practice, and the ethos of office. Calling to account and rendering, assessing, and sanctioning accounts often require pooling resources from several institutions and actors (Magnette 2000; Busuioc 2010).

Conceptions of accountability have come to rely on non-majoritarian, guardian institutions populated by officials neither directly elected nor managed by elected officials (p. 115) and loosely coupled to public opinion (Thatcher and Stone Sweet 2002). Judges, central bankers, and experts are assumed to act in a principled way, with competence, integrity, and impartiality, in accordance with laws, professional codes, the mandates of their office, and the common good. However, self-control and checks and balances among guardians have been assessed as democratically inadequate and simultaneously there are perceived to be accountability deficits and overloads. For instance, the Commission has become more accountable horizontally to other EU institutions, but less so to citizens. As a result, the Commission perceives itself as overloaded with accountability claims. In the eyes of the public it is unaccountable and deserves closer public scrutiny (Wille 2010, 83–85).

Living with Ambiguity and Conflict

When democracies face inconsistent claims, standards, and understandings, the task is usually seen as “striking a balance.” Claims are weighted. Trade-offs are made. Conflict is resolved and equilibrium reached. An alternative to assuming authoritative conflict resolution is to consider how democracies live with ambiguity and unresolved conflict. Democracies search for “satisficing” or acceptable solutions and accountability processes may be institution-specific or involve the interaction of a variety of institutions.

In settled polities and routine situations accountability processes are likely to take place in parallel, relatively autonomous, institutional spheres. They are incremental, institution-specific, and influenced by the history, organization, and dynamics of the institution, yet in the shadow of a common understanding of political order. Existing beliefs, status, and power relationships bias interpretation towards familiar templates. Inspired by Cyert and March (1963) it is hypothesized that in settled polities and well-understood situations repeatedly confronted, there are co-existing dyadic accountability relations, routines, standard operating procedures, self-restraint, and de-politicization. There is local rationality, where different institutions define accountability and hold different types of actors accountable for different things and in accordance with different normative standards. There is also sequential attention to claims, and demands for coordination are buffered by limited attention and slack resources. Rather than being collapsed into a coherent preference order through trade-offs, competing accountability claims are treated as independent constraints, together defining a satisfying solution. Reduced slack or austerity may generate demands for coordination. The search for new alternatives is driven by performance crises and change follows from shifts in attention and aspiration levels.

In unsettled polities and unchartered waters institutions and actors have to learn their place in the larger political order through experience with what is seen as acceptable and politically possible. There are likely to be demands for coordinated behavior, and accountability relationships and processes develop in a dynamic interplay between (p. 116) levels of governance and institutional spheres. The pursuit of accountability and the dynamics of accountability become indistinguishable.

The Dynamics of Accountability Regimes

Accountability processes are part of the emergence and transformation of accountability regimes. They provide opportunities for exploring how accountability may be enhanced, inhibited, and legitimized by alternative regimes for challenging old accountability relations and processes and developing new ones. There are debates and struggles over whether and why the principles organizing and governing common affairs deserve the allegiance of citizens. Questions are raised about why specific institutions and office-holders should have the right and capacity to call to account, render and assess accounts, and sanction unacceptable behavior. These are issues closely linked to competing visions of political order. In the European Union, for example, proposals for making institutions and actors accountable have been based upon different conceptions of the nature, purpose, and desired future of the EU. Visions have varied among those seeing the Union as an intergovernmental, or supranational, or regulatory entity (Bovens 2007; Harlow and Rawlings 2007; Bovens, Curtin, and t’Hart 2010, 180, 189).

The Belief in Design and Learning

Core democratic beliefs are that citizens should decide through what institutional arrangements they are to be governed and that democracies have a unique ability to learn from experience. Correctly assigning causal credit or blame is assumed to increase the frequency of success and reduce the frequency of failure (March and Olsen 1995). Issues are resolved over time, with agreed-upon results encoded into trustworthy institutions (Warren 2011, 523,526). Contemporary democracies are also involved in reconsideration of accountability relations and processes and reformers believe in auditing and account-giving as instruments for collective intelligence and improved performance (Pollitt and Summa 1997, 331). Principal–agent approaches likewise assume that “agency loss” is mitigated and citizens and legislators cope with conflicting preferences and information asymmetry through a variety of institutional devices, providing information about what agents are doing and why, as well as capacities for holding them to account. Control is embedded in the shifting design of electoral, legislative, and bureaucratic institutions, in processes of selecting agents, monitoring structures, reports from agents and from third parties, penalties for lying and misrepresenting (p. 117) facts, overturning and reversing agents’ decisions (Waterman and Meier 1998; Lupia and McCubbins 2000).

Deliberate design and reform is, however, only one among several change processes and it is problematic to assume that accountability regimes emerge and develop through institutional engineering and choice or competitive selection. Accountability arrangements may evolve slowly in relation to repeated encounters with well-structured, recurring problems in standardized situations. Yet, what are exceptions at one point in time become more common and gradually replace historically evolved routines, standards, and causal understandings. Key institutions, for example the Parliament, have developed through historical processes rather than deliberate design. In Westminster democracies the principle of ministerial responsibility has been adapted to shifting circumstances (Stone 1995), and the European Parliament (EP) has only recently succeeded in its appeal to “parliamentary democracy,” moving its relationship to the Commission closer to a traditional Parliament–Executive relationship with the Commission accountable to the EP.

Democracies do learn from experience on the basis of accounts, but experience is an unreliable teacher. It is difficult to give precise and valid accounts of what has happened and what could have happened, and to conclude precisely what lessons to extract. Lessons are often ambiguous and contested and they are sometimes lost. Learning may be spurious and contribute to mistakes rather than intelligence (March 2010). New routines for information collection are not automatically followed by routines of analysis, debate, and sanctioning. Information is gathered but not attended to or acted upon (Feldman and March 1981). Accounts may be perverted by incompetence, fraud, deceptions, disinformation, and indoctrination, raising questions regarding how citizens and office-holders actually interpret the relevance, validity, and implications of accounts, and how different accountability institutions support intelligence and learning (March and Olsen 1995).

In brief, institutional design and experiential learning do not guarantee efficient adaptation to human purposes or environmental change, and neither does competitive selection, bargaining, deliberation, imitation, diffusion, or other adaptive processes. An empirical lesson is that practice is likely to come before theory in complex, contested, and dynamic polities with multiple accountability claims. The meaning and implications of “accountability” are defined and changed when principles meet practical situations and generate specific implications, cleavages, and relationships.

Practice Precedes Theory

The European Union, as an unsettled polity with no shared vision of how accountability is to be organized and legitimized, provides several examples of evolving practice being theorized, explained, and justified post hoc. The Union has been concerned with how codes of conduct may nurture accountability, responsibility, and integrity. There (p. 118) has been criticism of the weak links between the European Parliament and citizens, the modest position of the EP and member state parliaments, and the lack of transparency and accountability of the Council. Bargaining in networks has created an accountability problem by making it difficult to identify who has been influential and should be held accountable. “Guardian” institutions such as the Commission, the Court of Justice, and the Central Bank have been shielded from popular control, making it difficult to hold them accountable.

Consistent with continental traditions the EU has great confidence in formal-legal institutions and decisions. Nevertheless, after institutions have been formally established, living institutions and their practices have evolved gradually through a more or less coordinated struggle for attention, resources, and support, and ahead of political visions and academic theories (Olsen 2007; 2010; 2012). Processes of delegation and accountability have developed in parallel and partly informally and independent of each other during “normal times.” Performance crises have generated demands for coordination and improved accountability.

Consider legal accountability and the birth of a new legal order in Europe. The origins of what has been called a juridical coup d’état, because courts in fundamental ways have transformed the normative foundations of the legal order through constitutional law-making (Stone Sweet 2007, 915), have a history. The ambiguities of the Treaties and different legal styles of interpretation allowed turf wars and struggle over competence and accountability. There have been gradual changes in judicial institutions, networks, and culture, based upon interaction between the European Court of Justice (ECJ), national courts, and the law profession. Through several partly autonomous events and decisions, and with significant elements of coincidence and chance, legal actors provided a “sufficient legal basis for the ECJ to revolutionize European law” and create an autonomous European legal order (Rasmussen 2008, 86, 98). A result was competing legal orders and ambiguous accountability relations.

Similar processes have taken place regarding administrative and financial accountability. European agencies (Busuioc 2010) and ombudsmen have struggled to establish an institutional identity, winning acceptance for the office, and building a Europe-wide professional network (Curtin 2007; Harlow and Rawlings 2007; Bovens, Curtin, and t’Hart 2010). The European Court of Auditors had difficulties developing an institutional identity as the guardian of sound financial management and gaining the attention of the EU’s executive powers. Contest and resistance arose before action capacity was built and the Court was able to assert its role as a central node in a matrix of financial accountability. The Court gradually gained confidence and was accepted, helped by the growing power of the Union, bigger budgets, more net contributors to the budget, broad media coverage of fraud and scandals, and support from the European Parliament (Laffan 2003).

The EU cohesion policy saw an audit explosion triggered by financial management scandals and loss of legitimacy. However, whereas the aim was to achieve better policy performance, attention became focused on compliance and traditional financial accounting practice related to legality, regularity, and detecting errors, fraud, (p. 119) and corruption more than on learning and discovering how performance could be improved. There were complaints regarding the scale, intensity, redundancy, inefficiency, and overlapping of audits, diverting attention from results, strangling risk-taking and innovation, and generating disappointment and distrust. Resources and power were redistributed in favor of the “watchdogs” rather than those implementing cohesion policies and new auditing practices and regimes were layered on top of old ones (Mendez and Bachtler 2011).

Ambiguity and Accountability

Understanding how contemporary representative democracies live with ambiguity and multiple, contested, and dynamic conceptions of accountability that are not subject to enduring resolution requires going beyond mainstream compliance-control approaches. Accountability cannot be captured assuming a single, formal-legal chain of dyadic relationships or a checks-and-balances system in equilibrium. Theorizing what accountability is all about, through what mechanisms accountability is pursued, assigned, assessed, and sanctioned, and how regimes change, makes it necessary to go beyond a historical and non-contextual conception of accountability, making normative standards, epistemic quality, and the distribution of authority and power endogenous to politics and accountability processes. However, approaches based on different foundational assumptions are likely to be complementary rather than excluding each other. Theory-builders need to explore competing ideas, their areas of application, strengths and weaknesses, and the circumstances under which different processes are likely to foster or hinder effective accountability and regime change.

An institutional approach based on an expanded conception of accountability has been offered as a starting point for understanding how accountability regimes are organized, work, and change in representative democracies characterized by ambiguity, uncertainty, and tensions. Whereas there is a long way to go before a proper theory of accountability is within reach, some elements of a research agenda can be offered. Understanding accountability assumes reconstructing how decisions are made and whether things could and should have been done differently.

Political and organizational decision-making is often messy. The fluidity, inconsistencies, and conflicts of political life—incompletely reconciled desires and understandings that evolve in the context of pursuing them, variable participation, and shifting legitimacy of resources—make it difficult to correctly assign causal responsibility, credit, and blame. Democratic politics, nevertheless, emphasizing human will, understanding, and control and responsive and accountable government, foster demands for calling someone to account, and institutions are social conventions that create elements of temporary and imperfect order and predictability. Rules and practices define appropriate behavior and specify what is natural, normal, reasonable, right, good, and true; what must be expected, can be relied on and makes sense in a community; and institutionalized (p. 120) resources affect the degree to which actors are capable of complying with what rules prescribe and proscribe. Beliefs in a legitimate accountability regime simplify political life and reduce the degree of ambiguity by ensuring that many things are taken as given.

The institutional approach offered has gone beyond the stylized categories of compliance-control approaches privileging the electoral channel, legislation, hierarchies, and dyadic relations. It has invited comparisons between accountability regimes and argued that standard assumptions are most applicable in settled polities and routine situations characterized by low degrees of ambiguity. The assumptions are likely to be fruitful in polities with a sovereign center of normative and coercive authority or a single dominant set of normative and organizational principles, and in polities where formal-legal authority and external control dominate other concerns. Compliance-control approaches are less applicable in capturing accountability in unsettled polities and unfamiliar situations, especially in multi-level and multi-centered polities embedded in heterogeneous, pluralistic, and dynamic societies, such as the European Union. Under such conditions ambiguity plays a significant role because there are a variety of accountability regimes stretching over levels of government and institutional spheres. There is no unified, sovereign people, legitimate way of formulating unambiguous mandates, or authoritative political center. There are competing sources of information and conflicting role demands and internalized identities. Several actors hold roles as both “principals” and “agents,” differently mobilized in different settings, and resources and capabilities are not necessarily adequate for prescribed roles. Possible next steps are to address in detail the actual operation and interaction of contested, co-existing accountability institutions, networks, and regimes, to develop typologies of accountability arrangements, and relate them to competing conceptions of what accountability is all about and what regimes foster effective accountability.

The institutional approach has gone beyond the assumptions that discovering and sanctioning deviance from pre-determined rules is the dominant aspect of accountability relations and processes, and that efficient accountability is best achieved through incentives and external control motivating a self-interested agent to behave as a self-interested principal desires. The approach has called attention to how accountability processes, especially in unsettled polities, provide occasions for searching for and testing collective purpose, intelligence, and meaning and political equality, and how internalization of different logics of appropriate behavior works as a mechanism through which institutions have an impact on calls for, rendering of, and responses to accounts. An institutional research agenda calls attention to the conditions under which accountability processes attract few or a variety of issues and participants; how training, socialization, and internalization may be stronger or weaker and more or less consistent across institutions; and how incentives and internalized rules interact in forming behavior. An underdeveloped issue is how accountability regimes affect how resources, rights, and capabilities are developed, distributed, regulated, and mobilized; how they make it more or less possible to act in accordance with key identities, roles, mandates, and responsibilities in representative democracies and, possibly, empower the powerless.

(p. 121) Finally, the institutional approach goes beyond the assumption that accountability regimes emerge, are maintained, and change as a result of the deliberate choices of pre-determined principals. It interprets the dynamics of accountability regimes as parts of complex institutionalization and de-institutionalization processes, and holds that compliance-control approaches are more likely to apply in a short-term perspective than in capturing long-term developments. The institutional agenda calls attention to how accountability processes provide opportunities for debates and struggles over what is a legitimate political order; how government is and should be organized; and where responsibility and accountability over different issues reside and should reside. Possible next steps are to examine what are stabilizing and de-stabilizing factors and frictions in change processes such as deliberate design, bargaining, deliberation, experiential learning, diffusion, and evolving historical processes. An institutional agenda in particular has to examine the internal dynamics of accountability regimes. In contemporary democracies accountability institutions are structured according to different principles and rules, providing competing analyses and prescriptions. Each institution is, furthermore, a collection of more or less tightly-coupled action programs determining possible actions over a range of circumstances that satisfies complex sets of standards, goals, rules, requirements, and constraints. There is therefore a need to explore how change follows because tensions and disputes are endemic in institutions and accountability regimes are never fully integrated or accepted by everyone.


Alesina, A. and Cukierman, A. 1990. “The Politics of Ambiguity.” Quarterly Journal of Economics, 105: 829–50.Find this resource:

    Beauvoir, S. de. 1972. The Ethics of Ambiguity. Secaucus, NJ: Citadel Press; originally published 1947.Find this resource:

      Bovens, M. 2007. “Analysing and Assessing Accountability: A Conceptual Framework.” European Law Journal, 13: 447–68.Find this resource:

        Bovens, M., Curtin, D., and ’t Hart, P. 2010. The Real World of EU Accountability: What Deficit? Oxford: Oxford University Press.Find this resource:

          Busuioc, M. 2010. The Accountability of European Agencies: Legal Provisions and Ongoing Practices. Delft: Eburon.Find this resource:

            Cohen, M. D., March, J. G., and Olsen, J. P. 1972. “A Garbage Can Model of Organizational Choice.” Administrative Science Quarterly, 17: 1–25.Find this resource:

              Cohen, M. D., March, J. G., and Olsen, J. P. 2012, “A garbage can model” at forty: A solution that still attracts problems, pp. 19–30 in The Grabage Can Model of Organizational Choice: Looking Forward at Forty, eds. A. Lomi and J. R. Harrison. Bingley, UK: Emerald Books.Find this resource:

                (p. 122) Connolly, W. E. 1987. Politics and Ambiguity. Madison: University of Wisconsin Press.Find this resource:

                  Curtin, D. 2007. “Holding (Quasi-) Autonomous EU Administrative Actors to Public Account.” European Law Journal, 13: 523–41.Find this resource:

                    Cyert, R. M. and March, J. G. 1963. A Behavioral Theory of the Firm. Englewood Cliffs, NJ: Prentice Hall.Find this resource:

                      Dubnick, M.J. 2005. “Accountability and the Promise of Performance: In Search of Mechanisms.” Public Performance & Management Review, 28: 376–417.Find this resource:

                        Etzioni, A. 1964. Modern Organizations. Englewood Cliffs, NJ: Prentice-Hall.Find this resource:

                          Feldman, M. S. and March, J. G. 1981. “Information in Organizations as Signal and Symbol.” Administrative Science Quarterly, 26: 171–86.Find this resource:

                            Gilmour, R. S. and Jensen, L. J. 1998. “Reinventing Government Accountability: Public Functions, Privatization, and the Meaning of ‘State Action.’” Public Administration Review, 58: 247–58.Find this resource:

                              Goodin, R. E. 1982. Political Theory and Public Policy. Chicago: University of Chicago Press.Find this resource:

                                Goodin, R. E. 2008. Innovating Democracy: Democratic Theory and Practice After the Deliberative Turn. Oxford: Oxford University Press.Find this resource:

                                  Gustavsson, S., Karlsson, C., and Persson, T. 2009. The Illusion of Accountability in the European Union. London: Routledge.Find this resource:

                                    Hanson, R. L. 1987. Democracy, pp. 68–69 in Political Innovation and Conceptual Change, eds. T. Ball, J. Farr and R.L. Hanson. Cambridge: Cambridge University Press.Find this resource:

                                      Harlow, C. and Rawlings, R. 2007. “Promoting Accountability in Multilevel Governance: A Network Approach.” European Law Journal, 13: 542–62.Find this resource:

                                        Keane, J. 2009. The Life and Death of Democracy. London: Simon & Schuster.Find this resource:

                                          Klinger, D. E., Nalbandian, J., and Romzek, B. S. 2001. “Politics, Administration, and Markets: Conflicting Expectations and Accountability.” American Review of Public Administration, 32: 117–44.Find this resource:

                                            Laffan, B. 2003. “Auditing and Accountability in the European Union.” Journal of European Public Policy, 10: 762–77.Find this resource:

                                              Landau, M. 1969. “Redundancy, Rationality, and the Problem of Duplication and Overlap.” Public Administration Review, 29: 346–58.Find this resource:

                                                Lupia, A. and McCubbins, M. D. 2000. “Representation or Abdication? How Citizens Use Institutions to Help Delegation Succeed.” European Journal of Political Research, 37: 291–307.Find this resource:

                                                  MacIntyre, A. 1988. Whose Justice? Which Rationality? Notre Dame: University of Notre Dame Press.Find this resource:

                                                    Magnette, P. 2000. “Towards ‘Accountable Independence’? Parliamentary Controls of the European Central Bank and the Rise of a New Democratic Model.” European Law Journal, 6: 326–40.Find this resource:

                                                      Manning, M. 2004. Torture and the Politics of Ambiguity: Project Syndicate. Retrieved from <>.

                                                      March, J. G. 2010. The Ambiguities of Experience. Ithaca: Cornell University Press.Find this resource:

                                                        March, J.G. and Olsen, J. P. 1976. Ambiguity and Choice in Organizations. Bergen: Universitetsforlaget.Find this resource:

                                                          March, J. G. and Olsen, J. P. 1989. Rediscovering Institutions: The Organizational Basis of Politics. New York: Free Press.Find this resource:

                                                            March, J. G. and Olsen, J. P. 1995. Democratic Governance. New York: Free Press.Find this resource:

                                                              March, J. G. and Weissinger-Baylon, R. 1986. Ambiguity and Command: Organizational Perspectives on Military Decision Making. Marshfield: Pitman. (p. 123) Find this resource:

                                                                Mendez, C. and Bachtler, J. 2011. “Administrative Reform and Unintended Consequences: An Assessment of the EU Cohesion Policy ‘Audit Explosion.’” Journal of European Public Policy, 18: 746–65.Find this resource:

                                                                  Olsen, J. P. 2007. Europe in Search of Political Order. Oxford: Oxford University Press.Find this resource:

                                                                    Olsen, J. P. 2008. Explorations in Institutions and Logics of Appropriateness: An Introductory Essay, pp. 191–202 in Explorations in Organizations, ed. J.G. March. Stanford: Stanford Business Books.Find this resource:

                                                                      Olsen, J. P. 2010. Governing Through Institution Building: Institutional Theory and Recent European Experiments in Democratic Organization. Oxford: Oxford University Press.Find this resource:

                                                                        Olsen, J. P. 2013. “The Institutional Basis of Democratic Accountability.” West European Politics, 36 (3): 447-73.Find this resource:

                                                                          Page, B. I. 1976. “The Theory of Political Ambiguity.” The American Political Science Review, 70: 742–52.Find this resource:

                                                                            Pollitt, C. and Summa, H. 1997. “Reflexive Watchdogs? How Supreme Audit Institutions Account for Themselves.” Public Administration, 75: 313–36.Find this resource:

                                                                              Porter-O’Grady, T., Hawkins, M. A., and Parker, M. L. 1997. Whole-Systems Shared Governance: Architecture for Integration. Gaithersburg, MD: Aspen Publishers.Find this resource:

                                                                                Przeworski, A., Stokes, S. C., and Manin, B. 1999. Democracy, Accountability, and Representation. Cambridge: Cambridge University Press.Find this resource:

                                                                                  Rasmussen, M. 2008. “The Origins of a Legal Revolution: The Early History of the European Court of Justice.” Journal of European Integration History, 14: 77–98.Find this resource:

                                                                                    Searle, J. R. 1995. The Construction of Social Reality. London: Penguin.Find this resource:

                                                                                      Selznick, P. 1992. The Moral Commonwealth: Social Theory and the Compromise of Community. Berkeley: University of California Press.Find this resource:

                                                                                        Shepsle, K. A. 1972. “The Strategy of Ambiguity: Uncertainty and Electoral Competition.” The American Political Science Review, 66: 555–68.Find this resource:

                                                                                          Stone Sweet, A. 2007. “The Juridical Coup d’État and the Problem of Authority.” German Law Journal, 8: 915–28.Find this resource:

                                                                                            Stone, B. 1995. “Administrative Accountability in the ‘Westminster’ Democracies: Towards a New Conceptual Framework.” Governance, 8: 505–26.Find this resource:

                                                                                              Strøm, K., Müller, W. C., and Bergman, T. 2003. Delegation and Accountability in Parliamentary Democracies. Oxford: Oxford University Press.Find this resource:

                                                                                                Thatcher, M. and Stone Sweet, A. 2002. “Theory and Practice of Delegation to Non-Majoritarian Institutions.” West European Politics, 25: 1–22.Find this resource:

                                                                                                  Verhey, L., Broeksteeg, H., and Driessche, I. van den 2008. Political Accountability in Europe: Which Way Forward: A Traditional Concept of Parliamentary Democracy in a EU Context. Groningen: Europa Law Publishing.Find this resource:

                                                                                                    Warren, M. E. 2011. Democracy, pp. 517–29 in The Oxford Handbook of the History of Political Philosophy, ed. G. Klosko. Oxford: Oxford University Press.Find this resource:

                                                                                                      Waterman, R. W. and Meier, K. J. 1998. “Principal-Agent Models: An Expansion?” Journal of Public Administration Research and Theory, 8: 173–202.Find this resource:

                                                                                                        Wille, A. 2010. The European Commission’s Accountability Paradox, pp. 63–86 in The Real World of EU Accountability: What Deficit? eds. M. Bovens, D. Curtin and P. ‘t Hart. Oxford: Oxford University Press. (p. 124) Find this resource:


                                                                                                          (1) . Thanks to Mark Bovens, Robert E. Goodin, and Thomas Schillemans for challenging comments and to James G. March for constructive suggestions and more than 40 years of discussions of ambiguity.