International political authority: on the meaning and scope of justified hierarchy in international relations

Abstract

Traditionally, states were widely believed to be the only institutions claiming political authority. More recently, though, a number of authors have argued that we find various instances of political authority on the international level. We discuss three prominent proposals for conceptualizing international authority: Michael Barnett and Martha Finnemore’s account of the authority of international bureaucracies, David Lake’s extension of ‘relational’ authority to the international realm, and Michael Zürn’s recent proposal for ‘reflexive’ authority. These authors provide a nuanced and empirically rich picture of hitherto mostly overlooked forms of power in world politics. Yet, we argue that in doing so they lose sight of the distinctly normative character of political authority relations: these relations are built on the explicit normative claim to the right to rule. When such a claim is considered to be justified, authority relations generate content-independent reasons for compliance. Thus understood, authority serves an important function, namely, to facilitate broadly accepted and normatively justified forms of hierarchical coordination. From a normative perspective, therefore, broadening the concept of authority to include various other forms of power deprives us of a critical yardstick against which international organizations should be evaluated. Moreover, it creates a distorted picture of the scope of international authority. Our world is shaped by highly problematic power relations. Yet, in order to meet current challenges of global governance, we need more, not less authority. To illustrate this argument we examine the case of the World Bank, an organization that exercises considerable power while explicitly avoiding any claim to political authority.

1. Introduction

Ever since the Westphalian peace treaty of 1648, states were widely believed to be the central instances of political authority. More recently, though, a number of scholars have highlighted various instances of international authority. Michael Barnett and Martha Finnemore, ¹ for example, argue that over recent decades international bureaucracies have developed autonomous authority in relation to their member states. David Lake prominently proposes to go even further and to conceive of international politics as shaped by relations of authority that encompass states, international organizations (IOs) but also private actors such as credit-rating agencies. ² Michael Zürn follows a similar path, aiming to capture new and more liquid forms of authority in International Relations (IR) with his concept of ‘reflexive authority’. ³ The renewed interest in international authority suggests that the phenomenon is far more widespread than previously assumed. This view of world politics, in turn, has strong normative implications: to the extent that we live in a world shaped by various forms of international political authority, the most pressing normative question seems to be whether, and how, these can be justified.

In this article, we question this view of power in world politics. Indeed, we argue that the debate on international authority suffers from a distorted conception of what political authority is essentially about. The debate has unearthed a nuanced and empirically rich picture of hitherto mostly overlooked forms of power in world politics. For both normative and analytic reasons, however, it is misleading to describe all these forms of power as instances of international political authority. To be able to arrive at such a conclusion, authors like Lake and Zürn need to considerably expand the concept of political authority. This, however, comes at the cost of losing sight of the distinctly normative character of political authority: when entity A has political authority over B, this means that A has a content-independent right to issue commands to B and B has a corresponding content-independent obligation to comply with A’s commands. Thus understood, relations of political authority serve an important social function, namely to facilitate broadly accepted forms of hierarchical coordination. From a normative perspective, then, broadening the concept of political authority to include various other forms of power deprives us of a distinct form of regulating our political life. Moreover, doing so is also misleading in analytic terms. Our world is shaped by highly problematic power relations. Yet, the scope of international political authority remains limited.

Our argument proceeds in three steps: first, we clarify our understanding of authority by drawing on the respective debates in political philosophy. Our aim is not to develop a new understanding of political authority. Instead, for the purposes of our discussion we draw on Max Weber’s conception of authority, and the way this conception has been further refined by authors like Joseph Raz. Based on these conceptual clarifications, we then turn to the debate on international political authority. Here, we discuss in detail three proposals for conceptualizing such authority: Barnett and Finnemore’s account of the authority of IOs, Lake’s concept of ‘relational’ authority, and Zürn’s notion of ‘reflexive authority’. We argue that in their attempt to broaden the concept of authority, these authors lose sight of the distinct character of political authority. In particular, they conflate political authority with epistemic authority and with exercises of power that do not involve a claim to the right to rule. Also, we show that, on closer inspection, many empirical examples cited by these authors do not qualify as instances of international authority. Importantly, our argument here is not that ‘our’ concept of political authority is somehow superior per se. Like the authors we discuss, we believe that the value of concepts and definitions is ultimately instrumental. Our insistence on a narrower definition of international political authority, therefore, is likewise motivated by instrumental considerations: such a deliberately narrow concept of political authority is uniquely suited to capture a distinct form of international relations.

In order to further clarify our argument, in the final section of this article we turn to the World Bank (WB) as an example of an IO that has considerable power, but explicitly refrains from claiming political authority. This example shows the significance of the phenomena discussed in the debate on international authority, and at the same time illustrates the need to distinguish political authority from other forms of power.

2. The meaning of authority

Authority has always been a core concept in analyses of politics and so it comes as no surprise that this concept has spurred many controversies. ⁴ In this section, we draw on the relevant debates in political philosophy in order to clarify our understanding of political authority as a normative relation that assigns an entity the ‘right to rule’. We begin by clarifying the inherently normative character of authority relations, and their social function. To further explain our understanding of authority, we then highlight two crucial distinctions: that between practical and epistemic authority, and that between interpersonal practical authority and institutionalized political authority.

2.1 Authority as a normative relationship

It is common to use the term ‘authority’ to describe a personal trait. If a person is said to ‘have authority’, that person is assigned a particular status that gives others reasons to follow his or her instructions. Often, the basis for this kind of personal authority is expertise: for instance, we might recognize a physician as an authority based on her reputation as an expert. If we do so, within her field of expertise we are likely to take her advice as authoritative. Other times, people are said to have ‘natural authority’, which does not refer to some special expertise but rather to a person’s charismatic ability to get others to follow her commands. Although inherently mythical, Weber showed that this kind of charisma is an important source of authority. ⁵ Yet another common source of personal authority is tradition.

To some extent, this understanding of individual authority also informs our thinking about institutional authority. This is not just a coincidence: after all, historically it is quite recent that institutional rule has replaced more personalized forms of rule, such as the rule of monarchs. Indeed, when applied to institutions, the general logic remains the same: to say that an institution possesses authority means assigning that institution a specific normative status; it is recognized as having the right to issue instructions.

This emphasis on the normative dimension of authority can already be found in Weber’s seminal discussion of the issue. Weber equates domination (Herrschaft) and authority, defining both as ‘the probability that certain specific commands (or all commands) will be obeyed by a given group of persons’. ⁶ Importantly, however, he distinguishes authority from other forms of power by noting that people obey an authority’s orders not out of self-interest or fear but because they recognize the authority’s right to issue commands. ⁷ Thus understood, authority describes a normative relation in which one entity has the right to rule, and another entity has the corresponding duty to obey. As Raz notes succinctly, ‘It is common to regard authority over persons as centrally involving a right to rule, where that is understood as correlated with an obligation to obey on the part of those subject to the authority’. ⁸

Conceptually, this understanding of authority as a normative relation is synonymous with certain usages of the term ‘legitimacy’. Indeed, if one conceives of legitimacy as the ‘right to rule’, as is common among political philosophers, the notion of legitimate authority becomes tautological. To be sure, one might define legitimacy more broadly as ‘moral justifiability’. This resonates with everyday usages of the term such as, ‘She has a legitimate interest in …’ and makes it possible to apply the attribute of legitimacy to very different social phenomena. In the context of authority, however, this broader usage of the term converges with the more specific philosophical definition: legitimate authority understood as ‘morally justified authority’ is equivalent to the ‘right to rule’ and thus again tautological. In the words of Ian Hurd, ‘the phrase legitimate authority is, strictly speaking, redundant’. ⁹ Yet another alternative would be to conceive of authority as a power relation in non-normative terms, that is, as a power relation that does not involve a claim to the right to rule but merely consists of a de facto exercise of rule. This would make it possible to distinguish authority from legitimacy. The price however, would be to blur the distinction between authority and other forms of power. Indeed, a non-normative account of authority misses the crucial point of authority relations. As noted by Weber, people comply with an authority’s instructions if and to the extent that the claim to the right to rule is (perceived as) morally justified. If an entity is (perceived as) illegitimate in the broader sense of not being morally justified, this entity cannot exercise authority. It can still project power and enforce a certain level of compliance, but it does not have authority. Illegitimate authority, in other words, is an oxymoron.

Terminologically, this emphasis on the normative nature of authority relations implies that it is important to distinguish between the claim to authority and the exercise of authority. Any entity can claim authority. Only when that claim is recognized as justified by those toward whom it is addressed, however, does an entity have a right to rule – and can it be described as exercising authority.

Methodologically, we can approach authority relations from both a normative and an empirical perspective. From the normative perspective, the central question is whether there are good reasons to justify an entity’s claim to authority. From the empirical perspective, on the contrary, the focus is on which entities claim authority in which ways and toward what end, or whether such claims are perceived as justified. These two perspectives can converge, but they need not, and often do not. The crucial point, however, is that at the center of both of these two perspectives lies a normative relation. An empirical approach to authority may refrain from making explicit normative judgments about a particular instance of authority; and yet, to identify relations of authority it will be necessary to trace what Weber described as the ‘legitimacy beliefs’ of those subject to an entity’s authority, that is, their normative judgments. ¹⁰ From both perspectives, authority relations usually come in degrees. For instance, an entity’s claim to authority can be (perceived to be) justified only to a certain extent; it can be (considered to be) justified in relation to some of those over whom authority is claimed but not in relation to others; or it can be (seen as) justified with regard to some questions but not with regard to others. Also, authority can be exercised differently; with restraint or excessively, in a particular domain or comprehensively.

Authority relations serve many different specific purposes. Nonetheless, it is possible to describe the general social function of authority as allowing us to shape our normative relations with others through hierarchical coordination. The thought here is that in many social contexts, hierarchy is beneficial for us because it allows us to better achieve certain ends. ¹¹ Relations of authority constitute a form of hierarchy that is (perceived to be) morally justified and, in doing so, simplify social interactions in two important ways: first, all those involved in these relations are assigned specific roles. Awareness of the rights and obligations associated with these roles simplifies interactions. A classical example for this benefit of authority relations is that of a ship’s captain. If the other boatmen aboard the ship recognize his right to command and their obligation to comply with his commands, complex maneuvers are possible even under duress. Second, relations of authority assign content-independent rights and obligations. Within the parameters defined by the specific relation, an entity that possesses authority can issue commands that are binding for those subject to the authority irrespective of their content. This, too, simplifies interactions, for it is no longer necessary to evaluate the merits of each individual command based on its content.

One important upshot of the inherent normativity of authority relations is that these relations must be made explicit: for an entity to have authority, it must be recognized by those subject to it as morally justified. This, however, is not possible if the entity itself denies, or hides, its claim to authority. Only when an entity makes its claim to authority explicit is it possible for others to evaluate the justification provided for this claim. Moreover, the social function of relations of authority just described also requires that an authority issues its commands explicitly. The benefits of justified hierarchical coordination only become available if the authoritative entity clearly states its general claim to authority, as well as its more specific commands. What kind of communicative act this requires will vary depending on the social context. It will not always have to take the form ‘I hereby command you to ….’ Yet, an authority must communicate its commands adequately in order to generate an obligation to comply.

2.2 Practical, epistemic, and political authority

Authority, as defined so far, can take different forms. One basic distinction is that between practical and epistemic authority. Practical authority gives actors content-independent reasons for performing action X. Those subject to a practical authority perform X because a command by that authority is (perceived to be) binding for them. Epistemic authority, on the contrary, provides content-independent reasons for believing Y. Those subject to an epistemic authority believe Y because a truth-claim by that authority is (perceived to be) binding for them. As Linda Zagzebski emphasizes, ‘My reason for adopting the authority’s belief is not dependent on the content of that belief; if he/she/it had believed q instead of p, I would have reason to believe q instead of p’. ¹²

Believing Y may, in turn, lead those subject to an epistemic authority to change their course of action. Indeed, if a certain belief Y unambiguously entails that one ought to perform X, then the claim to epistemic authority is inextricably linked with a claim to practical authority. When, for example, the pope declares his belief Y that X is a sin, he claims the epistemic authority to determine what status X has in the context of catholic dogma. At the same time, for a devout Christian the belief that X is a sin clearly implies that she should not engage in X. However, epistemic and practical authorities are not always coupled like this. On the one hand, a practical authority need not care whether I believe in Y as long as I follow its command to do X. On the other hand, an epistemic authority may give me reason to believe Y without requiring me to perform any action. For instance, a commission of historians may be assigned the epistemic authority to determine the facts about a historical event long past. When the commission does so, this may give me a reason to believe in their reconstruction of the events. Believing in their historical account, however, need not imply that I should pursue any particular course of action.

Certain forms of practical authority can be found in interpersonal relations. The prime example here is the relation between parents and small children. While parents may want to provide them with specific explanations for their commands, in the end children are thought to have a content-independent reason to do as their parents tell them to do (at least within certain limits). A relation of practical authority also emerges when one person makes a promise to another: the promisor thereby assumes an obligation to perform a certain action, unless the promisee relieves him of the promise. The promisee thus gains a certain kind of practical authority, for she can decide whether the promisor has to perform a certain action. While limited to a particular action, this authority is content-independent: the promisor has to perform the action not because it is an inherently good action but because the promisee demands he fulfill his promise. ¹³

In the following, though, we will focus primarily on institutionalized forms of practical authority. More specifically, we are interested in instances in which the entity claiming authority is an institution. We use the term political authority to describe these forms of institutionalized practical authority. This is not to deny that epistemic authority can have political implications, or can become politicized. Yet, if politics is ultimately about collective decision-making, then epistemic authority as such is not intrinsically political in the way that institutionalized practical authority is. The authority of the state in relation to its citizens is the most prominent example of institutionalized political authority. It is not, however, the only one. Religious organizations routinely claim practical authority over their members. Private companies, likewise, claim the authority to direct the actions of those who work for them. As we will discuss in more detail in the following section, in principle it is also possible for international institutions to exercise authority.

Before turning to the discussion of international authority, however, one further clarification is necessary. If a claim to political authority is (perceived to be) justified, those subject to the authority have a content-independent reason to follow its commands. From a normative perspective, not complying with an authority’s command constitutes a moral wrong. An actor that intends to act morally thus has a normative reason for following the command. Empirically, too, we would expect that a person that believes that an authority is justified will do her best to act in accordance with that belief. In an ideal world of morally motivated and fully informed actors the command by a political authority, by itself, would thus lead to full compliance. Under less idealized circumstances, however, there will often be situations in which a certain level of coercion may be instrumentally necessary to ensure a sufficient level of compliance. Mild forms of coercion may simply be a way of ‘reminding’ those subject to an authority of their obligations. More forceful forms of coercion, moreover, may be necessary when individuals or groups resist the commands of an authority. The crucial point, however, is that authority should not be conflated with the use of coercive force. Authority can be part of a justification for the use of force; authority alone, however, is neither sufficient nor necessary for the justification of coercion. For both normative and empirical reasons, it would therefore be wrong to identify relations of authority by looking for instances of coercion.

3. International authority

Having clarified the general concept of authority, we can now return to the more specific issue of political authority in international relations. Until quite recently, the state was seen as the only relevant instance of political authority. Conceptually, however, there is no reason why authority should be limited to the state. Moreover, empirical research on the growing influence of IOs and international law, ¹⁴ the expansion of international law, ¹⁵ and the changing nature of statehood and new forms of governance ¹⁶ has raised awareness of instances of authority beyond the state.

Indeed, it is uncontroversial that in a number of instances international institutions claim political authority. It is less clear to which extent these claims are (perceived to be) justified and to which extent these institutions actually make use of their authority. A first example is the United Nations Security Council (UNSC), which explicitly claims the authority to take all measures necessary in order to preserve international peace (UN Charter, ch. VII). ¹⁷ The International Criminal Court (ICC) is another example of international political authority, for it claims the right to issue commands that are binding for all states that have submitted themselves to the court by signing the ‘Rome Statute’. ¹⁸ As the ‘Rome Statute’ explicitly states, the ICC ‘shall have the power to exercise its jurisdiction over persons for the most serious crimes of international concern’. Importantly, however, its claim to authority – beyond the referral of a case from the UN Security Council – is limited to the signatories of the ‘Rome Statute’. International transitional administrations, such as those that have been established in Bosnia, Kosovo, or East Timor, are yet another example of international political authority. Unlike the previous examples, these international administrations make a very broad claim to authority: at least temporarily, they claim the right to assume the role of a state government. ¹⁹ Finally, transnational religious organizations constitute an interesting case of international authority. They claim the right to issue binding commands to their members irrespective of where they reside and can very effectively enforce their decisions, primarily by using the threat of social exclusion or stigmatization. Religious institutions such as the Al-Azhar University in Cairo, the Al-Medina University in Saudi Arabia, and the Al-Mustafa University in Iran, for instance, provide authoritative guidance to Muslims around the world. ²⁰ Likewise, the Catholic Church, with its emphasis on papal supremacy and strong reliance on transnational religious cadres and missionary activities, exercises substantial authority over Catholics around the globe. ²¹

Unfortunately, however, the renewed interest in international authority has been accompanied by a tendency to conflate the concept of authority with other forms of power. As noted in the introduction, conceptual definitions do not have intrinsic value. Their instrumental value depends on what phenomena they help us to identify and analyze. On the basis of the Weberian account of political authority explicated in the previous section, then, the broadening of the concept of political authority in the discussion of international authority is problematic: an overly broad conception of international authority loses sight of the distinctly normative character of political authority that distinguishes authority from other forms of power, and thereby also of the distinct social functions that political authority serves. As a consequence, moreover, it suggests a distorted picture of world politics as significantly shaped by political authority relations. In this section, we discuss three different proposals for conceptualizing international authority. In doing so, we want to clarify how these proposals depart from the conception of authority developed in the previous section, and why this is problematic. In particular, we will focus on Barnett and Finnemore’s account of the authority of IOs, Lake’s concept of ‘relational’ authority, and finally Zürn’s notion of ‘reflexive’ authority. As will become clear in our discussion of these different conceptions of international authority, we do not question that there are a few genuine instances of international authority. Our argument, however, is that these are far more limited in their reach, and far more seldom, than has been argued by the authors we discuss.

3.1 The authority of international bureaucracies

In their seminal book ‘Rules for the World: International Organizations in Global Politics’, Michael Barnett and Martha Finnemore made an important contribution to the debate on international authority by extending the concept to IOs. ²² As they write:

Our claim that IOs possess authority puts us at odd with much of IR theory, which presumes that only states can possess authority because sovereignty is the only basis of authority. ²³

More specifically, Barnett and Finnemore focus on the independent authority of international bureaucracies. They conceptualize authority as ‘the ability of one actor to use institutional and discursive resources to induce deference from others’. ²⁴ As they point out, this ability not only involves getting other actors to do what they otherwise would not, but also to tell right from wrong. Moreover, Barnett and Finnemore emphasize that authority relations are socially constructed by both the rulers and the ruled. Like all authorities, international secretariats depend on the widespread belief that they indeed have the right to rule. Barnett and Finnemore further conceive of bureaucracies as the embodiment of Weberian rational-legal authority characterized by technocratic expertise, formalistic procedures, impartiality, and nonviolence. In this sense, they hold that ‘bureaucracies are, by definition, authorities’. ²⁵

Barnett and Finnemore show that in addition to this emphasis on expertise, IO bureaucracies also draw on other sources of authority. Member states delegate authority to them in order ‘to take charge of a problem and sort it out’. ²⁶ According to Barnett and Finnemore, IO bureaucracies also claim authority in virtue of their useful and noble goals. Formally, the discretionary power of IO bureaucracies is limited to interpret broadly defined goals and thus confined ‘to a zone demarcated by state interests’. ²⁷ Yet, Barnett and Finnemore highlight how the capacity of IO bureaucracies to classify and organize information, their capacity to define problems in the first place and to coin key concepts, as well as their ability to spread global norms provides them with extraordinary powers to shape international affairs. As a result, IO bureaucracies may act against the interests even of powerful member states and/or shape states’ preferences so that they subsequently align with IO interests. ²⁸

In ‘Rules for the World’, Barnett and Finnemore discuss select activities of the International Monetary Fund (IMF), the United Nations High Commissioner for Refugees (UNHCR), and UN Department of Peacekeeping Operations (UNPKO). While Barnett and Finnemore are primarily interested in the ‘pathologies’ of bureaucratic organizations, these examples also illustrate their understanding of authority. On their account, the IMF commands authority in virtue of the ability to provide technical advice on matters of monetary, fiscal, labor, and environmental policies. Drawing on its expertise, and in conjuncture with a practice of conditional lending, the IMF shapes and reconfigures ‘domestic political and business institutions of all kinds, advising countries on appropriate configurations of everything from their social spending to their stock markets and banking sectors’. ²⁹ The UNHCR, in turn, exercises authority through (re-)defining key concepts, which then shape the discourse around refugees at large. Specifically, Barnett and Finnemore show how UNHCR expanded its scope of responsibility by operationalizing (and widening) the meaning of the term ‘refugee’. Finally, the authors illustrate the reluctance of the UN Secretariat and UNPKO to classify the mass killings in Rwanda in 1994 as ‘genocide’. Instead, their classification of the atrocities as ‘civil war’ contributed to the improper reaction to the catastrophe by the international community. Again, the UN’s framing of a situation induced other actors to behave in particular ways, including omissions.

While Barnett and Finnemore’s book offered a new perspective on the autonomous influence of IO bureaucracies, their conception of IO authority is problematic for three reasons. First, the definition of authority as ‘the ability of one actor to use institutional and discursive resources to induce deference from others’ ³⁰ does not reflect the inherently normative character of authority relationships. Instead, their definition of authority comes very close to Robert Dahl’s widely accepted definition of power as a relationship in which A causes (or has the ability to cause) B to do something that B would not otherwise do. ³¹ They can thus show that IO bureaucracies exercise considerable power but they cannot substantiate the further claim that these bureaucracies exercise political authority. ³²

Second, Barnett and Finnemore conflate epistemic authority with practical authority. They thereby lose sight of the different justifications required for these different forms of authority. Closely related, Barnett and Finnemore are empirically mistaken when they characterize IMF and UN activities under the heading of ‘expert authority’. In contrast to expert, or epistemic, authorities, both these organizations actively and deliberately shape discourses and regulate through publishing rankings and indicators. They do not merely provide expertise, but substantially exercise power. As noted, the definition of concepts such as ‘refugee’ or ‘genocide’ has direct implications for international policy-making. These organizations, therefore, do not pass as merely epistemic authorities.

Third, Barnett and Finnemore show that IOs, over time, tend to expand their operational independence by reinterpreting and pushing the boundaries of their original mandates in creative ways. As they write with regard to the IMF, ‘through its technical advice and conditionality programs, it has become intimately involved in members’ domestic economies in ways explicitly rejected by its founders’. ³³ Yet, neither the IMF, nor UNHCR or UNPKO publicly claim a right to rule. Quite to the contrary, they avoid such a claim, which would clearly violate the terms of their mandates.

3.2 Relational authority

With his approach to international authority, Lake draws on his earlier work on relational authority. Conceptually, he starts out with a Razian account of authority that is very similar to the one we propose. ³⁴ His intention, however, is to extend our understanding of political authority beyond the ‘formal-legal’ authority of established states. In order to do so, Lake proposes to conceive of political authority as a social relation that takes the form of a very distinct social contract. As he writes:

Relational authority is premised on an exchange between ruler and ruled in which A provides a social order of value to B sufficient to offset the loss of freedom incurred in his subordination to A, and B confers the right on A to exert the restraints on his behavior necessary to provide that social order. ³⁵

On this account of political authority, the stability of authority relations depends on whether the ruled, and the rulers, each uphold their part of the social contract. If they do so for extended periods, both sides develop ‘vested interests’, which significantly increases the stability of the authority relation. In an intriguing way, thus, Lake combines an account of authority as a normative relation with a rational choice explanation for why people conceive of an entity’s claim to authority as morally justified: the ‘legitimacy-belief’, on his account, follows from the rational self-interest of the ruled.

On the international level, Lake identifies a number of instances in which such patterns of relational authority have developed. A case that well illustrates his understanding of international authority is that of the World Trade Organization (WTO) Appellate Body. The WTO’s ‘Dispute Settlement Understanding’ states that its members ‘recognize that it [the Appellate Body] serves to preserve the rights and obligations of Members under the covered agreements, and to clarify the existing provisions of those agreements in accordance with customary rules of interpretation of public international law’. ³⁶ As Lake notes, the Appellate Body ‘issues binding rules, expects and receives broad compliance from member states, and authorizes punishment against violators’. ³⁷ What makes this a particularly good example for Lake’s account of relational authority is that here it is indeed plausible to assume that this instance of authority, to paraphrase Lake, provides a social order of value to the WTO members that offsets the loss of freedom incurred by their subordination to this authority. In short, it serves their interests well to recognize the authority of the Appellate Body.

Other examples that Lake uses to illustrate his account of international political authority, however, are less convincing and illustrate what we take to be a major problem of his account: by turning away from ‘formal-legal’ forms of authority to social patterns, Lake deliberately gives up the notion that authority relations are based on an explicit claim to authority. As a consequence, he can no longer explain why, and how, authority relations differ from other power relations.

This becomes particularly evident when he discusses the United States’ ‘informal empire over states on the Caribbean littoral’ ³⁸ as an instance of international authority. It is undisputable that the United States has long sought to impose its will on its Caribbean neighbors, variously relying on its military, economic, or diplomatic power. It is also well documented that many states in the Caribbean, or at least their rulers, have at times been willing to submit to the demands of the United States. In line with Lake’s notion of an implicit social contract, they felt that doing so would serve their interests better than resisting the United States. Yet, the notion of an ‘informal empire’ already indicates that the United States never publicly claimed the right to exercise this kind of power over their Caribbean neighbors. Likewise, the tyrants who placed their own interests over those of their citizens never publicly acknowledged the existence of such an empire.

The conceptually relevant point here is that, as noted above, a claim to authority must be made explicit to fulfill its normative purpose of establishing a justified and broadly accepted form of hierarchy. Empirically, it will often be important for the stability of an authority relation that those ruled will conceive of this relation as serving their own interests. Yet, this alone is not enough to constitute an authority relation. By drawing on Raz’s account of authority, Lake himself emphasizes that authority is a distinct normative relation. Those subjected to an exercise of power must believe it to be a justified exercise of the right to rule. When the entity exercising power itself does not claim the right to rule, and indeed denies to rule, this is not possible. Those subjected to this exercise of power cannot recognize it as an instance of authority. Moreover, when an entity does not claim authority, it also cannot create content-independent obligations for following its commands (e.g. to follow a given course of action in its economic policy). It is not plausible, therefore, to hold that those subject to such an exercise of power have a content-independent obligation to perform certain actions.

Our suggestion, thus, is not to describe relations of power such as those between the United States and its Caribbean neighbors as instances of international political authority. As significant as these power relations are, they are not relations of practical authority. This also shifts the normative perspective on these power relations: The problem in cases like these is not an excess of international political authority. Instead, the problem is that there is no international authority capable of ending such unjustifiable exercises of power.

3.3 Reflexive authority

Zürn presents yet another account of international authority. His conceptual starting point is Weber’s account of authority. Zürn, however, wants to broaden the notion of authority to go beyond what he at times describes as the solid authority of states. Conceptually, thus, he explicitly wants to go even further than Lake. As he notes:

it seems odd to conceive the practice of international and transnational political authorities in line with the mental map of solid authority as ‘commands’ issued by one actor that are expected to be obeyed by a second, and regularly backed up by force. (Lake 2009, 18) ³⁹

To do so, Zürn proposes the notion of ‘reflexive’ authority. To begin with, he distinguishes reflexive authority from traditional forms of authority. Traditional authority, on his account, requires the suspension of one’s own reasoning and thus amounts to ‘blind and mindless submission’. ⁴⁰ What he seems to have in mind here is the kind of authority ascribed to pre-enlightenment societies. Reflexive authority, on the contrary, is reason-based: people recognize an entity’s claim to authority as justified, because they have reasons to do so. An important motivation, though not the only one, is that they believe the entity to have superior expertise. ⁴¹

This emphasis on reasons is compatible with most current conceptions of political authority. Indeed, it even seems to fit with Lake’s notion of self-interested recognition of authority. After all, on Lake’s account, too, people accept a claim to authority because they have reasons to do so. Importantly, however, Zürn goes further by holding that reflexive authority is not limited to relations of command and obedience. As he writes, ‘Reflexive authority does not usually work with ‘commands to do x’, but rather with ‘requests to consider y’ […] Reflexive authorities do not usually give reasons that underlie a command’. ⁴²

At this point, Zürn’s notion of (international) authority clearly departs from the Razian conceptual framework as discussed above. Zürn explicitly argues that his conception of reflexive authority is meant to include both practical and epistemic authority. ⁴³ Indeed, in both the cases, the recognition of an authority is reason-based. The difference, however, is that epistemic authority does not, by itself, imply a claim to the right to rule. Epistemic authority provides reasons for believing a certain proposition, but not reasons to act in a certain way. Zürn’s core move, thus, is to extend the notion of international authority by including various forms of epistemic authority.

And indeed, empirically Zürn can show convincingly that there are many institutions on the international level that claim epistemic authority. In particular, he points to the activities of international nongovernmental organizations (NGOs) and expert committees. In an interesting twist, he emphasizes the phenomenon of politically assigned epistemic authorities, that is, entities that are formally assigned the status of an epistemic authority by a political institution. His prime examples here are credit-rating agencies and expert committees such as the International Panel for Climate Change (IPCC). As noted above, it is possible for epistemic authority to be closely tied with practical authority. Indeed, many of the examples that Zürn uses to illustrate the idea of politically assigned epistemic authority are almost borderline cases. When the IPCC provides us with reasons to believe in a certain understanding of the causes and likely effects of climate change, to anyone not totally oblivious to the suffering of others caused by climate change, it thereby also provides us with reasons to act in a certain way.

Yet, such borderline cases notwithstanding, it seems problematic to conceptually conflate practical and epistemic authority as Zürn does with his broad understanding of reflexive authority. Because epistemic authority does not have the normative function of providing us with justified forms of hierarchical coordination, the standards of legitimation are different. For one thing, the reasons for accepting a claim to epistemic authority differ from the kind of reasons for accepting a claim to practical authority. Moreover, given that practical authority severely limits our freedom of action (for it imposes on us an obligation to act in certain ways), we usually demand a very stringent explanation for why we should accept a claim to practical authority. Purely epistemic authority, on the other hand, does not affect our freedom of action in such a direct and severe way, and so the standards of legitimation are usually lower.

Emphasizing the differences between practical and epistemic authority is important, moreover, to get an adequate picture of the scope of international authority. It is true that there are quite a number of entities in international politics that claim epistemic authority. Yet, Zürn also admits that practical authority is rare on the international level. The broad notion of reflexive authority tends to obscure this more nuanced picture by capturing instances of both epistemic and political authority and thus creating the false impression of a general expansion of international authority.

To further clarify and illustrate our conceptual arguments, we now take a closer look at the World Bank (WB) as a prime example of an IO that increasingly relies on the kind of exercise of power discussed by Barnett, Finnemore, Lake, and Zürn but refrains from claiming authority. The WB walks the line between IOs that undisputedly claim political authority (e.g. the UN Security Council) on the one hand and those who explicitly portray themselves as epistemic authorities (e.g. the IPCC). It therefore makes for a particularly interesting case to show why distinguishing international political authority from other forms of power is normatively indispensable.

4. The World Bank

Created in Bretton Woods, New Hampshire, in 1944, the main task of the WB (then named the ‘International Bank for Reconstruction and Development’ (IBRD)) was to address capital deficiencies and thereby stabilize the global economy after World War II. ⁴⁴ During the 1980s, the WB focused on macroeconomic policies and efforts to increase private capital flows. Specifically, it tried to enhance the privatization of public services, trade liberalization, and deregulation through conditionalities attached to its ‘structural adjustment programmes’ (SAPs). ⁴⁵ After mounting critique and a realization that SAPs had disastrous effects on human rights ⁴⁶ and poverty alleviation, ⁴⁷ the WB abandoned its SAPs and increasingly identified the lack of institutional capacity as the main obstacle to development. ⁴⁸ Since the early 1990s and especially since the turn of the millennium the WB strengthened its role as a ‘knowledge bank’. ⁴⁹

Throughout its history, the WB has thus exercised power in different ways. ⁵⁰ Yet, the WB has always refrained from claiming political authority vis-a-vis recipient states. To avoid misunderstandings from the beginning, the WB does exercise practical authority in one specific respect. In line with its mandate, it claims and exercises practical authority over its member states on budgetary matters concerning the organization. Similar to a sports club in which the managing board levies membership fees, the WB’s Board of Directors (BoD) sets and adjusts the quotas each member has to pay. Formally, the BoD is an organizational body of the WB. In practice, its members – the Executive Directors – are delegates of governments and represent national interests accordingly. Yet, the debate around the expansion of IO authority, rightly, does not focus on this limited internal authority, but rather on the attempts by IOs to project authority externally.

In the context of its development work, then, one can distinguish two major forms of WB power: economic and discursive power. The WB exercises economic power over borrowing states by making flows of aid conditional on changes in policy. ⁵¹ The more recipient states depend on the disbursement of loans, the higher the degree of dependency and thus WB power vis-a-vis the recipient state. This exercise of economic power in direct interactions has been well researched. ⁵²

Yet, since the 1990s the WB has shifted toward a new form of power, namely, discursive power. An important milestone in this regard was the publication of the World Development Report 1998/1999 on ‘Knowledge for Development’. Starting with this publication, the WB began to present itself as the most important knowledge provider in the field of development, as a ‘knowledge bank’, or even a ‘knowledge empire’. ⁵³

The notion of discursive power highlights how power can be exercised by shaping social discourses, for example, through spreading propaganda, framing perceptions, or pushing certain narratives. ⁵⁴ On a conceptual level, the exercise of discursive power is distinct both from epistemic and practical authority. It goes beyond epistemic authority because an entity exercising discursive power does not merely state a truth-claim but actively aims to force and/or manipulate others to accept that claim. At the same time, discursive power is not an instance of practical authority, unless it is combined with an explicit claim to the right to rule.

There are some instances, in which the WB can plausibly be described as claiming epistemic authority. For instance, numerous Working Papers, Journal Articles, and Flagship Reports (e.g. on approaches to prevent violent conflict) ⁵⁵ can reasonably be located in the realm of epistemic authority. In these publications, the WB draws on its research capacities to make cause and effect statements about the world, and thus claims epistemic authority in the field of global development.

However, the bulk of the WB’s activities goes beyond that and can be described as exercises of discursive power. Specifically, the WB defines key concepts and rates and ranks the performance of states. The WB thus integrates different forms of discursive power under one roof. All of these activities have in common that they go beyond purely epistemic authority. Crucially, however, the WB deliberately refrains from any explicit claim to the right to rule, and thus does not claim, or exercise, international political authority.

4.1 Defining core concepts

Actively shaping the core concepts of a policy field constitutes a major form of discursive power that goes beyond merely offering truth-claims. As noted above, in Rules for the World, Barnett and Finnemore discuss how the UNHCR exercised discursive power in the field of humanitarian aid by defining who qualifies as a ‘refugee’ and thus directly affected the fate of millions of people. In the realm of development policies, the WB is arguably the most important organization in actively shaping the transnational discourse. ⁵⁶ According to Jones, it was around the year 2000 that the WB sought to become the world’s leading ‘knowledge bank’ that would invent and shape key concepts in both academia and the policy world. ⁵⁷ Even before that, the WB had shaped the discourse on development, for instance, by introducing the term ‘governance’ in the late 1980s and early 1990s. ⁵⁸ Over the next two decades, ‘good governance’ became the dominant concept informing development approaches for the Western donor community at large. While WB talk about ‘government’ deficits would have gone beyond its mandate formulated in the Bank’s Articles of Agreement (which clearly state that the WB only has a ‘non-political mandate’), the term governance was advocated, allowing the Bank to rank and evaluate the political achievements of states. ⁵⁹ What is more, the term governance was coined in a way that allowed the WB to adopt a more comprehensive perspective on the provision of goods and services. Instead of focusing on state institutions only, governance was broad enough to include private actors, NGOs, as well as the WB itself in governance coalitions beyond the state. The WB’s focus on governance allowed to redefine itself and its orbit of operations. Rather than being a purely economic actor, the WB began working on a range of governance issues that had formerly been under the supervision of states, including economic and social policies, transportation, communication, health and nutrition, to energy and extractive industries. ⁶⁰

Indeed, the WB not only defined key concepts in these fields but also put them into practice. By providing ‘technical assistance’, it inserted these concepts into the domestic politics of recipient states. The term ‘technical assistance’ refers to a wide range of activities from trainings, assistance in carrying out reforms in a specific policy field, or in building whole institutions from scratch. For example, WB projects involving efforts at privatization, private sector development, and financial sector modernization as core components may include assistance to the government of the recipient country to draft legislation, to build new institutions, or to train public officials. Hence, technical assistance is a standard mechanism through which discourses on ‘governance’ and ‘development’ materialize in recipient states. ⁶¹

With a large staff of economists and specialists in a range of development fields, the WB’s research and communications budget by far exceeds the resources available to governments and universities in most of its member countries. Due to its self-ascribed and barely unquestioned role as an architect of global development policies defining norms of best practice, the WB is able to shape the development discourse, and thereby to exercise discursive power over states and other development organizations. Yet, it does so without claiming a right to rule, precluding the normative demands associated with political authority.

4.2 Governance by rating and ranking

David Lake and Michael Zürn discuss the example of credit-rating agencies that exercise considerable influence on states’ economic performance (see sections ‘Relational authority’ and ‘Reflexive authority’). Similarly, merging epistemic authority and discursive power, the WB develops indicators for the governance performance of countries as a whole. While it portrays its ratings and rankings as objective and apolitical, the choice of criteria and methodology reflects the perspective of the WB’s staff and influential member states. ⁶² Among the most important ones, the Country Policy and Institutional Assessment (CPIA) scores determine the future distribution of loans and subsidies. With the help of an ‘assessment questionnaire’, WB country teams evaluate member states according to 16 criteria grouped together into four clusters (‘Economic Management’, ‘Structural Policies’, ‘Polices for Social Inclusion’, and ‘Public Sector Management and Institutions’). In combination with a benchmarking report, this aggregated evaluation is consequential: it constitutes the basis on which the Board of Governors decides over the conditions for loans of a given country. ⁶³ Thus, CPIA assessments clearly go beyond the mere exercise of epistemic authority.

Rankings, likewise, amount to effective instruments of discursive power. ⁶⁴ A prominent example is the Worldwide Governance Indicators (WGIs) which have been developed by the WB to capture the governance performance of member states since 1996. Covering six key dimensions of ‘governance’, countries are ranked according to, for example, government effectiveness, regulatory quality, political stability, lack of violence, and corruption. In contrast to the output-oriented CPIA focusing on institutions, the WGI covers a wide range of governance indicators from 30 organizations and relies on expert assessments. Because of its public nature, its wide recognition as a tool for measuring governance quality and because it compares countries, negative results on the WGI follow the logic of ‘naming and shaming’. ⁶⁵ Although rankings are not immediately relevant for the allocation of funds, they do exercise pressure on governments and have been rightly described as attempts at ‘global governance by indicators’. ⁶⁶ Yet, again, no public claim to authority is made.

4.3 Why the World Bank does not claim authority

A claim to authority exposes an international institution to demands for adequate justification. If successful, such justification creates a normative bond between the international institution and those subject to it. The latter accept the institution’s right to rule, which in turn allows for demanding and yet stable forms of rule. However, in light of the legitimation most IOs have to offer, a claim to authority would get them in hot water – and the WB knows that.

In part, the WB as an organization exercises power without a clear awareness regarding the scope, domain, and strength of its influence. Due to the amount of actors in international relations with conflicting, coexisting, or overlapping mandates and spheres of influence, the attribution of responsibility becomes increasingly difficult – for civil society and academics as well as for WB staff. ⁶⁷ However, the lack of a clear center of power does not imply the absence of particularly powerful actors who are in a position to shape transnational norms and policies in their favor, and who often also have an interest in doing so. In the field of development, the WB is a particularly influential actor – but avoids publicly claiming authority. After reviewing the academic literature, the WB’s Articles of Agreement, Project Documents, and after conducting 20 interviews with WB staff in its headquarters in Washington D.C., we suggest that the avoidance of authority has systemic causes. Among them, three stand out.

First, considerations of legality are paramount. Unlike the few IOs with an explicit mandate to interfere with the sovereignty of its member states, such as the USNC or the ICC, the WB shares with the vast majority of IOs that its articles of agreement oblige it to be politically neutral:

The Association and its officers shall not interfere in the political affairs of any member, nor shall they be influenced in their decisions by the political character of the member or members concerned. ⁶⁸

In the history of the WB, there is a tension between this nonpolitical mandate and its desire to have an impact in a number of policy fields. For instance, in 1990 its chief lawyer, Ibrahim Shihata, explicitly claimed the legal competence to interfere in those sectors which he deemed immediately relevant for macroeconomic reforms. ⁶⁹ Yet, attempts to adopt a wide interpretation of its legal mandate have been subject to increasing criticism on behalf of its member states, especially since the rise of emerging powers as Brazil, Russia, India, South Africa, and, especially, China (BRICS). ⁷⁰ Overall, the WB is careful to emphasize the authority of recipient countries’ governments. The usual project cycle is indicative here to understand how it reconciles the desire for impact with a simultaneous respect for (formal) sovereignty. In a first step, proposals for projects may come from WB staff as well as from the government. The exercise of discursive power is relevant here as governments may consciously or unconsciously propose projects in anticipatory obedience to the dominant development discourse. Where the government fails to recognize a need for action, WB staff may propose and recommend identifying a given problem in an ongoing, informal dialogue. Here, the circle closes as WB staff can refer back to a normative order (consisting of concepts, ratings, and rankings) it brought into being in the first place. After projects have been identified, their preparation and approval lies with the WB. The end of the project preparation cycle marks a turning point, since the signing of a contract passing the sole legal responsibility for the project’s implementation onto the government. ⁷¹ The WB is thus not only able to shield any legal action against it, it also avoids accusations of paternalism on behalf of recipient governments. Hence, while formal legal rules prohibit the WB from adopting direct coercion, it may nevertheless shape outcomes through an increasing reliance on discursive power. ⁷²

Second, the choice of indirect means of (discursive) influence is partly motivated by considerations of effectiveness. In the course of the ‘Structural Adjustment Programmes’ (SAPs) in the 1980s and 1990s, the WB experienced how the direct exercise of power (e.g. by dictating conditionalities to states who depended on WB loans) was met by recipient governments with suspicion and increasing mistrust of the WB. Since the WB depends on governments to implement projects or reforms, this threatened its effectiveness. Today, it is well understood among WB staff that the informal exercise of power geared towards shaping the intentions of states is much more effective than direct exercises of power. Several interviewees hinted at the fact that coercive tools do not work. Where the WB used to force governments to engage in reforms, these governments simply would agree but not actually implement the reforms. To cite an experienced staff member who worked for the WB’s Internal Evaluation Group (IEG) for more than two decades:

Coercive tools don’t work. If you try to push things down anybody’s throat (…) and they are not convinced, they will say yes to take the money but they won’t do anything on the ground. The World Bank has learned the hard way whenever they have tried coercion is has not worked. So the big mantra now is country-ownership. ⁷³

Third, the WB is well aware of its restrictions when confronted with demands for democratic procedures of decision-making and accountability. It witnessed the rise of global protest movements in resistance to the IMF, WTO, and itself since the early 1990s. ⁷⁴ In response, the WB pursued a variety of strategies to mitigate the critique. For instance, it created an accountability regime in the late 1980s and early 1990s comprising social and environmental safeguards and an Inspection Panel (IP), where citizens can directly request an examination of projects without a prior detour through their national governments. ⁷⁵ Moreover, it has improved its information and disclosure policy in the mid-1990s as well as its dialogue with civil society (particularly through the regular ‘Civil Society Policy Forum’ (CSPF)). The IEG critically evaluates activities of the WB since the 1970s, and has initiated several policy reforms since then. These examples represented important innovations. Yet, if the WB were to openly claim the right to rule, these rather limited forms of democratic accountability would most likely not satisfy the public demand for an appropriate justification of the right to rule.

In sum, the WB faces a challenge: it needs to demonstrate to its its shareholders that its actions have a substantial impact, while avoiding a claim to authority. The organizational adaptation that we sought to demonstrate involves the exercise of discursive power to enhance its effectiveness, to delegate legal responsibility, and to mitigate resistance from global civil society it faces several constraints in its external environment. These constraints involve a legal dimension, considerations for effectiveness as well as one of democratic accountability. ⁷⁶ Yet, fully aware of the normative demands coming on the heels of claims to authority, the WB refrains from making such claims.

5. Conclusion

Can we observe patterns of political authority in international relations? In recent years, a growing number of IR scholars have used the conceptual framework of authority to analyze patterns of hierarchical relationships among actors, thus fundamentally questioning the premise of structural anarchy in international relations. There is consensus among all the authors discussed in this article that the international system is far from being anarchic and that IOs increasingly shape global politics. Yet, we hope to have shown that Michael Barnett and Martha Finnemore, David Lake, as well as Michael Zürn tend to conflate international political authority with related, but different forms of power. We argued that they do so by emptying the concept of political authority of its distinct normative components, in particular, the claim to having a right to rule. As discussed above, a political institution that exercises authority does not merely impose its will on others but stands in a normative relation with those who recognize its right to rule.

A problematic consequence of the expansion of the concept of international political authority, then, is that it leads to a distorted picture of the current scope of international political authority. It suggests a view of world politics as shaped by a significant increase in international authority. Closer attention to the specifics of authority relations, on the contrary, leads to a more nuanced assessment. International organizations are increasingly powerful but, for the most part, they do not claim authority. There are a few cases of genuine international political authority, such as the UNSC or the ICC, but these cases are fewer than suggested by Barnett, Finnemore, Lake, and Zürn. The example of the WB illustrates this important distinction. The WB is an IO that exercises considerable influence in virtue of its economic power and, increasingly, its discursive power to coin key concepts of the development discourse, to govern by rating and ranking and to shape normative orders through projects on the ground. Yet, the WB explicitly refrains from claiming political authority.

The expansive use of the concept of international authority, finally, makes it difficult to grasp the normative challenge we face. The increasing scope and depth of IO governance in world politics is not a sign of too much political authority, but instead indicates a dearth. If the social function of authority is to allow for morally justified forms of hierarchical coordination, it seems that we need more, not less, international political authority. How different IOs could – according to their competencies and functions – transform into political authorities, and how we might achieve such reforms, are thus critical questions political scientists and political philosophers should pursue in the future.